US20150175031A1 – Hoverboard – Google Brevets

RENVOI AUX APPLICATIONS CONNEXES

  • Cette demande de brevet revendique une priorité inférieure à 35 U.S.C. §119 (e) à la demande de brevet provisoire américaine n ° 62/066 891, déposée le 21 octobre 2014, intitulée «Hoverboard», par Henderson et al., Qui est incorporée dans son intégralité à toutes fins utiles aux présentes. Cette demande de brevet revendique une priorité inférieure à 35 U.S.C. §119 (e) de la demande de brevet US n ° 61 / 977.045, déposée le 8 avril 2014, intitulée «Applications de dispositifs magnétiques à distribution de flux unilatérale», par Henderson et al., Qui est incorporée dans son intégralité à titre de référence à toutes fins utiles ici. Cette demande de brevet revendique une priorité inférieure à 35 U.S.C. §119 (e) de la demande de brevet US n ° 62 / 011.011, déposée le 11 juin 2014, intitulée «Applications de dispositifs magnétiques à distribution de flux magnétique unilatérale» de Henderson et al., Qui est incorporée par référence à dans son intégralité à toutes fins utiles. Cette demande de brevet revendique une priorité inférieure à 35 U.S.C. §119 (e) de la demande de brevet US n ° 62/031 756, déposée le 31 juillet 2014, intitulée «Propulsion et contrôle d'un véhicule à levage magnétique», par Henderson et al., Qui est incorporée dans son intégralité à titre de référence fins de la présente. Cette application revendique une priorité inférieure à 35 U.S.C. §120 et est une continuation-en-partie de la demande de brevet US Ser. N ° 14 / 069,359, intitulée «Lévitation magnétique d'un objet fixe ou mobile», déposée le 31 octobre 2013 par Henderson, qui revendique une priorité inférieure à 35 U.S.C. §119 (e) à la demande de brevet provisoire américaine n ° de ser. 61/999 695, intitulé «Lévitation magnétique stationnaire» par Henderson, déposé le 15 mars 2013, dont chacun est incorporé par référence dans son intégralité et à toutes fins utiles.

  • DOMAINE DE L'INVENTION

  • La présente invention concerne de manière générale des systèmes à lévitation électromagnétique, et plus particulièrement un hoverboard, qui utilise une lévitation électromagnétique.

  • CONTEXTE

  • Avant octobre 2014, Wikipedia, l'encyclopédie en ligne, décrivait un hoverboard comme une planche fictive en lévitation utilisée pour le transport personnel dans les films «Retour vers le futur, partie II» et «Retour vers le futur, partie III». planche à roulettes sans roues. À travers des effets spéciaux, les cinéastes ont représenté les planches planant au-dessus du sol.

  • Depuis ces films, les gens rêvent de faire du hoverboard. Quelques groupes ont prétendu avoir créé un hoverboard. Cependant, les dispositifs se sont révélés être des canulars élaborés. Comme suit, un hoverboard, qui utilise des forces magnétiques pour survoler, est décrit.

  • SOMMAIRE

  • Un hoverboard est décrit. Le hoverboard génère des forces magnétiques qui peuvent être utilisées pour soulever et propulser le hoverboard. Les forces magnétiques sont générées à l'aide d'un ou plusieurs moteurs en vol stationnaire. Dans un mode de réalisation, les moteurs de vol stationnaire comprennent chacun un moteur électrique qui est configuré pour faire tourner un réseau d'aimants. Le réseau d'aimants en rotation est configuré pour induire des courants de Foucault dans un substrat conducteur. Les courants de Foucault induits génèrent un champ magnétique qui repousse les aimants dans le moteur en vol stationnaire, ce qui le fait planer dans les airs.

  • Dans un mode de réalisation, le hoverboard comprend quatre moteurs en vol stationnaire et une plate-forme de pilote. Un coureur peut se tenir sur la plate-forme du coureur pour piloter le hoverboard. En vol, un pilote peut modifier sa position et son poids sur la plate-forme du pilote pour incliner l’ensemble du hoverboard et le propulser dans une direction donnée. Les forces de translation générées peuvent être suffisantes pour permettre au hoverboard de grimper ou de se maintenir en position sur une surface inclinée. De plus, des mécanismes peuvent être fournis qui permettent aux moteurs en vol stationnaire d'être inclinés individuellement pour contrôler un équilibre des forces agissant sur le vol stationnaire. Par exemple, il est possible de prévoir des liaisons mécaniques permettant au cycliste de saisir des forces qui actionnent un ou plusieurs des mécanismes d’inclinaison à la fois afin de contrôler plus finement la direction du hoverboard.

  • Dans un mode de réalisation particulier, le vol stationnaire peut être caractérisé en ce qu’il comprend deux moteurs volants ou plus. Chacun des moteurs en vol stationnaire peut comporter un moteur électrique, un régulateur de vitesse électronique, une antenne et un hauban. Le moteur électrique peut comprendre un enroulement, un premier ensemble d'aimants permanents et une première structure contenant les premiers aimants permanents, un courant électrique étant appliqué à l'enroulement pour faire tourner l'un des enroulements ou le premier ensemble d'aimants permanents.

  • Le régulateur de vitesse électronique peut contrôler une vitesse de rotation du moteur électrique. Le bras peut être configuré pour recevoir un couple de rotation du moteur électrique qui fait tourner le bras. Le carénage peut renfermer le moteur électrique et au moins une partie supérieure de la branche.

  • Le STARM peut inclure un deuxième ensemble d’aimants permanents et une deuxième structure qui maintient le deuxième ensemble d’aimants permanents. Le second ensemble d'aimants permanents peut être tourné pour induire des courants de Foucault dans un substrat, de sorte que les courants de Foucault induits et le second ensemble d'aimants permanents interagissent pour générer des forces qui font passer le hoverboard et le cycliste en position stationnaire à l'emplacement le long du substrat. Le hoverboard peut être configuré pour accueillir des mouvements complexes impliquant une rotation et une translation simultanées.

  • Une plate-forme de pilote sur laquelle la personne monte peut avoir une extrémité avant et une extrémité arrière. Un premier moteur en vol stationnaire peut être monté sous la plateforme du pilote près de l'extrémité avant et un second moteur en vol stationnaire peut être monté sous la plate-forme du pilote près de l'extrémité arrière de la plate-forme. De plus, une source d’énergie électrique qui transmet le courant électrique aux deux ou plusieurs moteurs en vol stationnaire peut être montée sous la plate-forme du pilote entre le premier moteur en vol stationnaire et le deuxième moteur en vol stationnaire.

  • Dans des modes de réalisation particuliers, une charnière mécanique peut être couplée à un premier moteur en vol stationnaire qui permet au bras, au moteur électrique et à l'enveloppe du premier moteur en vol stationnaire de s'incliner par rapport à la plate-forme du cycliste. Une amplitude d'une force de translation, produite par le premier moteur en vol stationnaire, peut changer en réponse à l'inclinaison de la tige, du moteur électrique et du carénage. La force de translation peut être utilisée pour contrôler l’orientation du hoverboard. Un mécanisme d'inclinaison peut être couplé au premier moteur de vol stationnaire configuré pour recevoir une force externe et / ou générer une force interne qui incline le premier moteur de vol stationnaire en réponse à une force d'entrée générée par la personne.

  • Un autre mode de réalisation peut être caractérisé en tant que dispositif de levage magnétique. L'élévation magnétique peut comprendre un ou plusieurs moteurs de vol stationnaire, une source d'énergie électrique et une structure destinée à contenir le ou les moteurs de vol stationnaire et la source d'énergie électrique. Chaque moteur en vol stationnaire peut comprendre un moteur électrique, un régulateur de vitesse électronique et une antenne. Le régulateur de vitesse électronique est généralement disposé électroniquement entre le moteur électrique et la source d'alimentation électrique. Il peut contrôler la puissance atteignant le moteur électrique pour contrôler une vitesse de rotation du moteur.

  • Le moteur électrique peut comprendre un enroulement, un premier ensemble d'aimants permanents et une première structure contenant les premiers aimants permanents, un courant électrique étant appliqué à l'enroulement pour faire tourner l'un des enroulements ou le premier ensemble d'aimants permanents. Le bras peut être configuré pour recevoir un couple de rotation du moteur électrique qui fait tourner le bras. Le bras peut comprendre un deuxième ensemble d’aimants permanents et une deuxième structure qui maintient le deuxième ensemble d’aimants permanents, le deuxième ensemble d’aimants permanents étant mis en rotation pour induire des courants de Foucault dans un substrat tels que les courants de Foucault induits et le deuxième ensemble d’éléments permanents. Les aimants interagissent pour générer des forces qui permettent au dispositif de levage magnétique de planer et de se déplacer librement dans n'importe quelle direction au-dessus du substrat.

  • DESCRIPTION BRÈVE DES DESSINS

  • Les dessins inclus ne sont donnés qu'à des fins d'illustration et servent uniquement à fournir des exemples de structures et d'étapes de processus possibles pour les systèmes et procédés de l'invention décrits. Ces dessins ne limitent en aucune manière les modifications de forme et de détail qui peuvent être apportées à l'invention par l'homme du métier sans s'écarter de l'esprit et du cadre de l'invention.

  • figure. 1 est une illustration d'une personne chevauchant un hoverboard conformément aux modes de réalisation décrits.

  • FIGUES. 2 et 3 sont des illustrations de courants de Foucault générés sur une plaque conductrice en réponse à des agencements d'aimants entraînés en rotation au-dessus des plaques conformément aux modes de réalisation décrits.

  • figure. 4A est un tracé des courbes de portance et de traînée associé à un agencement d'aimants rotatifs selon les modes de réalisation décrits.

  • figure. 4B est un tracé de la portance associé à un agencement d'aimants rotatifs en fonction de la distance d'un substrat conducteur conformément aux modes de réalisation décrits.

  • figure. 4C La figure 1 est un tracé de courbes de portance associé à un agencement d'aimants rotatifs en fonction de l'épaisseur d'un substrat conducteur et de la vitesse de rotation en fonction des modes de réalisation décrits.

  • FIGUES. 5A et 5B sont des schémas fonctionnels illustrant les effets de vol stationnaire et de propulsion des agencements en rotation d'aimants conformément aux modes de réalisation décrits.

  • figure. 6 La figure 1 est une vue de dessous d'un hoverboard avec des moteurs en vol stationnaire qui basculent conformément aux modes de réalisation décrits.

  • figure. sept La figure 1 est une vue de côté d'un moteur en vol stationnaire dans une position d'inclinaison conforme aux modes de réalisation décrits.

  • figure. 8A La figure 1 est une vue en perspective d'un STARM conforme aux modes de réalisation décrits.

  • figure. 8B est une vue de côté de STARM avec un moteur intégré conformément aux modes de réalisation décrits.

  • figure. 8C La figure 1 est une vue de côté d'un STARM intégré à un moteur conforme aux modes de réalisation décrits.

  • FIGUES. 9, dix et 11 sont des vues de dessous, de dessus et de côté d'un hoverboard alimenté par batterie conformément aux modes de réalisation décrits.

  • FIGUES. 12 et 13 Les figures 1 et 2 sont des vues en perspective et latérale d'un hoverboard avec des moteurs en vol stationnaire couplés à des entretoises conformément aux modes de réalisation décrits.

  • figure. 14 La figure 3 est une illustration d'un moteur en vol stationnaire couplé à un bras de levier conformément aux modes de réalisation décrits.

  • figure. 15 est une illustration d'un moteur en vol stationnaire couplé à des pédales conformément aux modes de réalisation décrits.

  • figure. 16 est une illustration d'un moteur en vol stationnaire couplé à des rênes conformément aux modes de réalisation décrits.

  • FIGUES. 17A à 17E Les illustrations ci-dessous illustrent des configurations de carénage pour moteurs en vol stationnaire conformément aux modes de réalisation décrits.

  • FIGUES. 18A à 18D Les illustrations ci-dessous illustrent des configurations de carénage pour les moteurs en vol stationnaire où un sommet du carénage forme une partie de la plate-forme du pilote conformément aux modes de réalisation décrits.

  • FIGUES. 19A à 19C sont des illustrations de vues de côté et de dessus de moteurs de vol stationnaire inclinables selon les modes de réalisation décrits.

  • figure. 20 La figure 1 est une illustration d'un vol stationnaire à quatre moteurs en vol stationnaire conformément aux modes de réalisation décrits.

  • figure. 21 La figure 1 est une illustration d'un hoverboard avec deux moteurs de vol stationnaire rotatifs selon les modes de réalisation décrits.

  • FIGUES. 22A à 22D et 23 sont des illustrations d'un hoverboard avec deux moteurs en vol stationnaire conformément aux modes de réalisation décrits.

  • figure. 24 La figure 1 est une illustration d'un hoverboard avec huit moteurs en vol stationnaire conformément aux modes de réalisation décrits.

  • FIGUES. 25A à 25E sont des illustrations des directions de mouvement en réponse à une entrée de force à différents endroits sur un hoverboard conformément aux modes de réalisation décrits.

  • DESCRIPTION DES MODES DE REALISATION PREFERES

  • La présente invention va maintenant être décrite en détail en faisant référence à quelques modes de réalisation préférés de celle-ci, tels qu'illustrés dans les dessins annexés. Dans la description suivante, de nombreux détails spécifiques sont exposés afin de permettre une compréhension complète de la présente invention. Cependant, il sera évident pour l'homme du métier que la présente invention peut être mise en pratique sans tout ou partie de ces détails spécifiques. Dans d'autres cas, des étapes de processus et / ou des structures bien connues n'ont pas été décrites en détail afin de ne pas obscurcir inutilement la présente invention.

  • En ce qui concerne les figures suivantes, un hoverboard est décrit. Le hoverboard est configuré pour générer une portance utilisant des forces magnétiques. Dans des modes de réalisation particuliers, un individu peut monter sur le hoverboard de la même manière que sur une planche à roulettes, une planche de surf ou une planche à neige, c’est-à-dire utiliser les mouvements de son corps pour contrôler le mouvement du dispositif. Cependant, contrairement à ces dispositifs, le hoverboard peut générer ses propres forces de propulsion. Ainsi, la propulsion de l'appareil ne dépend pas uniquement des forces générées par l'utilisateur, de la gravité ou de l'énergie des vagues.

  • De plus, en vol stationnaire, les forces liées au frottement sont très faibles. Ainsi, le hoverboard peut être configuré pour offrir une plus grande liberté de mouvement et une plus grande maniabilité par rapport à une planche à roulettes, une planche de surf ou une planche à neige. Les détails d'un hoverboard sont décrits par rapport à FIGUES. 1-15. En particulier en ce qui concerne figure. 1, un individu chevauchant un hoverboard est illustré. Par rapport à FIGUES. 2 et 3, deux agencements d'aimants pouvant être utilisés pour générer une portance magnétique et leur interaction avec un substrat conducteur sont décrits.

  • Par rapport à figure. 4A, soulève et traîne générée en fonction de RPM est discuté. Avec le respect de figure. 4BOn discute de la portance en fonction de la distance par rapport à un substrat conducteur. La portance en fonction de la vitesse de rotation et de l’épaisseur du substrat conducteur est décrite par rapport à figure. 4C.

  • Les forces propulsives et les forces de commande pouvant être générées par un moteur en vol stationnaire, tel qu'un moteur en vol stationnaire couplé à un vol stationnaire, sont décrites par rapport à FIGUES. 6 et 7. Par rapport à FIGUES. 8A, 8Bande 8C, une configuration STARM et un moteur pouvant être utilisé pour former un moteur en vol stationnaire sont discutés. Deux configurations de hoverboard sont décrites en ce qui concerne FIGUES. 9-13. Des configurations supplémentaires de hoverboard sont décrites en ce qui concerne FIGUES. 22A à 25.

  • Par rapport à FIGUES. 14-16 Des mécanismes analogiques et numériques utilisés pour déplacer un moteur en vol stationnaire à des fins de contrôle et de propulsion sont décrits. Par rapport à FIGUES. 17A-17E la configuration du carénage pour les moteurs en vol stationnaire est discutée Enfin, des configurations d’intégration de moteurs en vol stationnaire rotatifs à une plate-forme de charge utile sont décrites en ce qui concerne: FIGUES. 18A-21.

  • Bien que beaucoup des modes de réalisation décrits ici soient discutés dans le contexte d'un hoverboard monté par une personne, divers aspects des modes de réalisation sont également applicables à de nombreux autres types de dispositifs de levage électromagnétiques. Par exemple, les moteurs en vol stationnaire qui s'inclinent peuvent être utilisés pour propulser et lever des appareils de levage électromagnétiques qui ne sont pas nécessairement configurés pour transporter une personne. Comme autre exemple, les configurations de moteur et STARM (STator et ARMature) décrites ici peuvent être utilisées dans des dispositifs de levage électromagnétiques configurés pour lever une palette. D'autres exemples d'appareils de levage électromagnétiques et d'applications sont fournis dans les demandes de brevet provisoires et les demandes de brevet de service public précédemment incorporées ici à titre de référence. Ainsi, l'exemple d'un hoverboard contrôlé par l'homme est fourni à des fins d'illustration et n'est pas censé être limitatif.

  • Système Hoverboard et fonctionnement

  • Par rapport à FIGUES. 1 à 5B, quelques exemples généraux et principes de fonctionnement d’un système hoverboard sont décrits. Le système hoverboard peut comprendre un hoverboard ayant des moteurs de vol stationnaire et un substrat sur lequel le hoverboard fonctionne. Le substrat peut comprendre une partie conductrice dans laquelle sont induits des courants de Foucault. L'interaction électromagnétique entre le dispositif qui induit les courants de Foucault et les courants de Foucault induits peut être utilisée pour générer une portance électromagnétique et diverses forces de commande de translation et de rotation.

  • figure. 1 est une illustration d'une personne dix monter un hoverboard 12. Dans un mode de réalisation, le vol plané comprend quatre moteurs en vol stationnaire, tels que 16. Les moteurs de vol stationnaire 16 générer un champ magnétique qui change en fonction du temps. Le champ magnétique variant dans le temps interagit avec un matériau conducteur en piste 14 former des courants de Foucault. Les courants de Foucault et leurs champs magnétiques associés ainsi que les champs magnétiques du moteur en vol stationnaire interagissent pour générer des forces, telles qu'une force de levage ou une force de propulsion. Des exemples de courants de Foucault pouvant être générés sont décrits en relation avec FIGUES. 2 et 3. La portance et la traînée associées aux courants de Foucault induits sont décrites par rapport à FIGUES. 4A-4C.

  • Dans figure. 1, la piste 14 est formé de cuivre. En particulier, on utilise trois feuilles de cuivre superposées superposées. D'autres matériaux conducteurs peuvent être utilisés et le cuivre est décrit à des fins d'illustration uniquement. Les surfaces incurvées peuvent être formées plus facilement en utilisant un certain nombre de feuilles minces en couches. Par exemple, un half-pipe peut être formé. Dans figure. 1, une partie d'un half-pipe est montrée. La piste 14 peut inclure diverses surfaces inclinées et plates et l’exemple de demi-tuyau est fourni à titre illustratif seulement.

  • L'épaisseur du matériau conducteur utilisé peut dépendre des propriétés du matériau conducteur, telles que sa capacité de transport de courant et la quantité d'élévation magnétique souhaitée. Un moteur en vol stationnaire particulier, en fonction de facteurs tels que la force du champ magnétique de sortie, la vitesse de déplacement du champ magnétique et la distance du moteur en stationnaire par rapport à la surface d'une piste, peut induire des courants de Foucault plus ou moins puissants suivre le matériel. Différents moteurs de survol peuvent être configurés pour générer différentes quantités d'ascenseurs et induire ainsi des courants de Foucault plus forts ou plus faibles.

  • La densité de courant associée aux courants de Foucault induits dans le matériau peut être maximale à la surface, puis décroître avec la distance qui le sépare de la surface. Dans un mode de réalisation, la densité de courant induite à la surface peut être de l'ordre de un à dix mille ampères par centimètre carré. Lorsque le matériau conducteur devient plus mince, il peut atteindre une épaisseur où la quantité de courant potentiellement induite par le moteur en vol stationnaire est supérieure à celle que le matériau conducteur peut contenir. À ce stade, la quantité de portance magnétique produite par le moteur en vol stationnaire peut chuter par rapport à la portance potentiellement générée si le matériau conducteur était plus épais. Cet effet est discuté plus en détail en ce qui concerne figure. 4C.

  • À mesure que l'épaisseur du matériau augmente, les courants induits deviennent de plus en plus petits à mesure que l'on s'éloigne de la surface. Après avoir atteint une certaine épaisseur, l’addition de matériau entraîne très peu de soulèvement supplémentaire. Pour les moteurs de vol stationnaire utilisés pour le hoverboard 12, des simulations ont montré qu’utiliser ½ pouce de cuivre ne produirait pas beaucoup plus de portance que l’utilisation de pouce de cuivre.

  • Pour le périphérique montré dans figure. 1, des simulations ont prédit que l’utilisation d’une feuille de cuivre de seulement 1/4 pouce réduirait considérablement la portance par rapport à l’utilisation d’un demi-pouce de cuivre. Dans divers modes de réalisation, la quantité de cuivre pouvant être utilisée varie en fonction de l'application. Par exemple, pour un modèle de hoverboard à petite échelle configuré pour transporter une poupée, une feuille de cuivre de pouce peut être plus que suffisante. Comme autre exemple, une piste avec une quantité plus fine de matériau conducteur peut conduire à une génération de portance moins efficace par rapport à une piste avec une quantité plus épaisse d'un matériau plus conducteur. Cependant, le coût du matériau conducteur peut être échangé contre l'efficacité de la génération d'ascenseur.

  • Un substrat 14 peut inclure une partie configurée pour supporter les courants de Foucault induits. De plus, il peut comprendre des parties utilisées pour ajouter un support mécanique ou de la rigidité, pour assurer un refroidissement et / ou pour permettre l’assemblage d’une partie de la piste. Par exemple, des tuyaux ou des ailettes peuvent être fournis qui sont configurés pour éliminer et / ou déplacer la chaleur vers un emplacement particulier. Dans un autre exemple, le substrat 14 peuvent être formés comme une pluralité de carreaux qui sont configurés pour s'interfacer les uns avec les autres. Dans un autre exemple encore, la partie du substrat 14 qui est utilisé pour supporter les courants de Foucault induits peut être relativement mince et des matériaux supplémentaires peuvent être ajoutés pour fournir un support structurel et une rigidité.

  • Dans divers modes de réalisation, la partie du substrat 14 utilisé pour supporter les courants de Foucault induits peut être relativement homogène dans la mesure où ses propriétés sont sensiblement homogènes en profondeur et d'un emplacement à l'autre. Par exemple, une feuille de métal solide, telle que de l'argent, du cuivre ou de l'aluminium, peut être considérée comme sensiblement homogène quant à ses propriétés en profondeur et d'un endroit à l'autre. Comme autre exemple, un matériau composite conducteur, tel qu'un polymère ou un composite, peut être utilisé lorsque les propriétés du matériau sont en moyenne relativement homogènes d'un endroit à l'autre et en profondeur.

  • Dans d’autres modes de réalisation, la partie du substrat 14 utilisé pour supporter les courants de Foucault induits peut varier en profondeur mais il peut être relativement homogène d’un endroit à l’autre. Par exemple, la partie du substrat 14 qui supporte les courants de Foucault peuvent être formés à partir d'un matériau de base qui est dopé avec un autre matériau. La quantité de dopage peut varier en profondeur de sorte que les propriétés du matériau varient en profondeur.

  • Dans d’autres modes de réalisation, la partie du substrat 14 qui supporte les courants de Foucault peuvent être formés à partir de couches de matériaux différents. Par exemple, un isolant électrique peut être utilisé entre des couches d'un matériau conducteur, telles que des couches de cuivre isolées les unes des autres. Dans un autre exemple, une ou plusieurs couches d'un matériau ferromagnétique peuvent être utilisées avec un ou plusieurs matériaux paramagnétiques ou diamagnétiques.

  • Dans un autre exemple encore, la surface du substrat 14 qui supporte les courants de Foucault peut inclure une structure de surface, telle que des fossettes surélevées ou enfoncées, qui affecte les courants de Foucault induits ou toute autre propriété matérielle. Ainsi, d’un site à l’autre, il peut y avoir de légères variations dans les propriétés du matériau, mais en moyenne sur une zone particulière, les propriétés du matériau peuvent être relativement homogènes d’un endroit à l’autre.

  • L’avantage d’avoir des propriétés matérielles relativement homogènes d’un endroit à l’autre est que la portance et la traînée générées en moyenne seront relativement les mêmes quelle que soit la direction de translation du hoverboard le long de la surface. Dans d’autres modes de réalisation, les propriétés du matériau peuvent varier d’un endroit à l’autre pour modifier les caractéristiques de portance et de traînée locales associées au hoverboard lorsqu’il se déplace dans une direction particulière. Les configurations dans lesquelles les propriétés des matériaux varient peuvent être utilisées pour mettre le coureur à l’épreuve et rendre le trajet plus intéressant. Par exemple, des dalles de propriétés matérielles différentes peuvent être liées, telles que des dalles de cuivre et d’aluminium, pour simuler un effet tel que des bosses en ski.

  • Dans un autre exemple, la piste peut être amincie à un certain endroit. L’effet d’une voie plus mince peut entraîner une chute de l’ascenseur local et une augmentation de la traînée. Les déséquilibres dans les forces de traînée générées par un STARM peuvent être utilisés pour générer des forces de translation qui propulsent un hoverboard. Ainsi, une augmentation locale de la traînée résultant d’un amincissement local de la partie du substrat 14 peut être utilisé pour fournir un boost de vitesse de translation en localisation. Cette augmentation de vitesse peut être utilisée avant une pente ascendante ou ascendante pour aider le hoverboard ou un autre type d'appareil de levage électromagnétique à gravir la pente.

  • Dans un mode de réalisation, une section, telle que 18, de la partie du substrat qui supporte les courants de Foucault induits peuvent être configurés pour tourner. La rotation de la section peut être entraînée par un moteur situé sous le substrat. Par exemple, un disque circulaire, dont le sommet affleure le haut du substrat environnant, peut être configuré pour tourner tandis que le substrat environnant reste dans une position fixe.

  • Comme on le verra plus en détail ci-dessous en ce qui concerne FIGUES. 2-4C, les propriétés des courants de Foucault induits dépendent du mouvement des aimants par rapport à la partie du substrat qui supporte les courants de Foucault induits. Ainsi, une modification du mouvement relatif entre les aimants et le substrat peut modifier les propriétés de portance et de traînée locales émises par un moteur en vol stationnaire. Par exemple, la rotation du substrat peut être utilisée pour améliorer localement la portance et la traînée ou pour réduire localement la portance et la traînée sur la partie du substrat en rotation.

  • Dans un mode de réalisation, une pluralité de sections du substrat peuvent être configurées pour tourner. La pluralité de sections peut être configurée pour s'aligner sur un ou plusieurs moteurs de survol d'un hoverboard, c'est-à-dire que le moteur de survol peut être placé sur le dessus de la section rotative. Par exemple, un hoverboard peut comprendre quatre moteurs en vol stationnaire et quatre sections de substrat rotatif peuvent être fournies pour être alignées avec les quatre moteurs. Dans un autre exemple, un hoverboard peut comprendre huit moteurs en vol stationnaire et quatre sections de substrat rotatif peuvent être fournies, lesquelles s'alignent avec seulement quatre des moteurs en vol stationnaire.

  • Une section de substrat en rotation peut être utilisée conjointement avec un moteur en vol stationnaire pendant une phase de démarrage où un moteur en vol stationnaire est au repos et son STARM ne tourne pas. Dans un mode de réalisation, le substrat en rotation peut être utilisé pour réduire le couple de traînée subi par le moteur lors du démarrage. Comme décrit plus en détail en ce qui concerne figure. 4AÀ partir du repos, la traînée peut augmenter avec l’augmentation du régime, atteindre un pic puis diminuer avec l’augmentation du régime. Comme décrit en ce qui concerne figure. 4B, la portance et la traînée diminuent de façon exponentielle à mesure que la distance entre le moteur en stationnaire et la surface du substrat augmente. Ainsi, le substrat en rotation peut être utilisé pour aider à surmonter les conditions de démarrage lorsque le régime est faible et que la traînée est élevée et / ou lorsque le moteur en vol stationnaire est proche de la surface et que la traînée est élevée.

  • Un avantage de cette approche est que cela peut réduire les exigences de couple associées au moteur. Les exigences de couple plus faibles peuvent permettre l'utilisation d'un moteur moins puissant. Un moteur moins puissant peut offrir plus de temps de vol car le moteur consomme moins d'énergie au démarrage et peut peser moins lourd.

  • Par exemple, une ou plusieurs sections rotatives du substrat peuvent être remontées dans une direction opposée à celle du moteur en vol stationnaire. Initialement, le moteur en vol stationnaire peut rester au repos. Dans un mode de réalisation, le substrat peut être mis en rotation suffisamment rapidement pour générer une portance, telle que remontée jusqu'au régime opérationnel du moteur en vol stationnaire. Ensuite, le moteur en vol stationnaire peut être mis en rotation pendant que le substrat est en rotation de telle sorte qu'un régime relatif constant soit maintenu entre le substrat qui supporte les courants de Foucault induits et les aimants dans le moteur en vol stationnaire. Lorsque le régime relatif constant sélectionné est supérieur au régime où se produit le couple de traînée maximal, le couple de traînée maximal subi par le moteur sera inférieur à celui obtenu s'il était filé sur une surface non rotative.

  • Si le substrat en rotation génère suffisamment d’élévation, le hoverboard s’élève à une hauteur donnée avant le démarrage du moteur. Comme décrit ci-dessus, le couple de traînée peut chuter de manière exponentielle avec la distance de la surface. Si le moteur en vol stationnaire est démarré après que la hauteur particulière a été atteinte, le couple maximal subi par le moteur en vol stationnaire aura diminué de manière exponentielle en fonction de la hauteur réelle atteinte.

  • Dans un mode de réalisation particulier, une enceinte peut être prévue autour d'une ou de plusieurs parties de substrat en rotation. Un hoverboard peut être placé dans l'enceinte et soulevé à la suite de la rotation du substrat, puis les moteurs de hoverboard peuvent être démarrés de sorte que la génération de la force de portance soit transférée du substrat en rotation aux moteurs en vol stationnaire. L'enceinte peut maintenir le hoverboard en position pendant ce processus. Ensuite, un coureur peut alors marcher sur le hoverboard et une extrémité de l’enceinte peut être ouverte pour permettre à l’hoverboard de quitter l’enceinte.

  • Le processus décrit ci-dessus peut être utilisé en sens inverse pour abaisser le hoverboard dans une position de repos. Tout d'abord, tout ou partie des forces de portance peuvent être transférées des moteurs en vol stationnaire au substrat en rotation en faisant tourner les moteurs en vol stationnaire tout en faisant tourner le substrat en rotation. Pendant ce processus, le hoverboard peut planer à une hauteur approximativement constante. Comme décrit ci-dessus, le transfert peut impliquer que le substrat rotatif génère la totalité ou une partie de la vitesse relative entre les aimants du moteur en vol stationnaire et le substrat. Ensuite, les moteurs en vol stationnaire, seuls ou en combinaison avec le substrat en rotation, peuvent être utilisés pour abaisser le vol en stationnaire jusqu'à une position sur le sol où les moteurs en vol stationnaire sont arrêtés.

  • Dans un mode de réalisation, un dispositif logique, tel qu'un processeur embarqué, sur le panneau survolant peut être configuré pour communiquer avec un dispositif logique qui contrôle la vitesse de rotation d'un ou de plusieurs tronçons ou substrat en rotation. Les périphériques peuvent communiquer pour permettre la génération d’un profil RPM relatif. Dans un mode de réalisation, le dispositif logique sur la carte survolée ou le dispositif logique contrôlant les sections de substrat en rotation peut être configuré pour commander temporairement les moteurs en vol stationnaire et les sections de substrat en rotation afin de mettre en œuvre un profil RPM relatif. Dans un autre mode de réalisation, le dispositif logique peut être configuré pour échantillonner les vitesses de rotation des divers composants rotatifs et développer un profil RPM en temps réel.

  • Dans un autre mode de réalisation, le dispositif logique du hoverboard peut être configuré pour mettre en oeuvre des profils RPM de démarrage pour ses moteurs en vol stationnaire, ce profil étant mis en œuvre en réponse à la réception d'une commande d'un dispositif distant ou à la détection d'un actionnement d'un dispositif mécanique, tel que comme un interrupteur ou un bouton. Le dispositif logique contrôlant les sections rotatives du substrat peut également être configuré avec un ou plusieurs profils RPM de démarrage dans lesquels un profil de démarrage est mis en œuvre en réponse à la réception d'une commande d'un dispositif distant ou à la détection de l'actionnement d'un dispositif mécanique, tel que comme interrupteur ou bouton. Les sections de substrat hoverboard et en rotation peuvent chacune être commandées pour entrer dans un mode de démarrage dans un ordre quelconque afin de générer un profil de RPM relatif entre le hoverboard et la ou les sections du substrat en rotation.

  • Divers profils de taux de rotation relatifs peuvent être générés entre un moteur en vol stationnaire et une section de substrat en rotation. Ainsi, l'exemple consistant à faire tourner une section de substrat en rotation jusqu'à une vitesse de rotation opérationnelle du moteur en vol stationnaire tandis que le moteur en vol stationnaire reste au repos, puis à maintenir une vitesse de rotation relative constante entre les dispositifs lorsque le moteur en vol stationnaire tourne, est fourni aux fins suivantes: d’illustration seulement et n’est pas censé être limitatif. Par exemple, un moteur en vol stationnaire et un substrat rotatif peuvent être utilisés pour générer un profil de vitesse de rotation relative avec un pic supérieur à la vitesse de rotation opérationnelle du moteur en vol stationnaire. La vitesse de rotation relative maximale peut être d'abord générée, puis les dispositifs peuvent être contrôlés de sorte que la vitesse de rotation opérationnelle soit approchée d'en haut lorsque les vitesses de rotation de chaque dispositif sont ajustées.

  • Retournant vers figure. 1, la personne peut contrôler le hoverboard 12 en déplaçant leur poids et leur position sur le hoverboard. Le changement de poids peut changer l'orientation d'un ou de plusieurs moteurs en vol stationnaire 16 par rapport à la surface de la piste 14. L'orientation peut inclure une distance de chaque piste du moteur en vol stationnaire. L'orientation de chaque moteur en vol stationnaire, tel que 16, par rapport à la surface de la piste peut générer des forces parallèles à la surface. Further details, related to the orientation of a hover engine and the associated forces which are generated are described with respect to FIGUES. 5A, 5B, 6 et sept et 24 et 25.

  • The net force from the hover engines 16 can be used to propel the vehicle in a particular direction and control its spin. In addition, the individual may be able to lean down and push off the surface 14 to propel the hoverboard 12 in a particular direction or push and then jump onto to the hoverboard 12 to get it moving in a particular direction.

  • In other embodiments, as described in more detail with respect to FIGUES. 9-15, mechanisms can be provided which allow the direct control of one or more hover engines separately from the other hover engines. The control mechanisms can be analog, digital or combinations thereof. An analog control mechanism can receive a user supplied force. In the mechanism, the user supplied force can be transferred to the hover engine such that the user supplied force affects an orientation of the hover engine. The orientation of the hover engine can cause a force to be generated in a particular direction.

  • A digital control mechanism can be configured to detect a movement of a person or a movement of a device by the person. For example, a movement of a person's head, arms or a device held by a person can be detected or a pressure exerted on a sensor in a person's shoe or on the hoverboard 12 can be detected. In one embodiment, multi-axis accelerometer system can be used to detect and quantify an amount of movement. In another embodiment, a camera system can be used to detect and quantify an amount of movement. In yet other embodiments, an optical system, such as a laser system, can be used, to detect position and quantify an amount of movement. Further, wireless systems, which employ triangulation, including GPS can be employed.

  • The movement or the detected pressure can be converted to a control signal by an onboard processor. The control signal can be sent to a controller, separate from the onboard processor, associated with one or more actuators coupled to a hover engine or a group of hover engines. The control signal can cause the controller to operate the actuators in a manner which changes the orientation of the hover engine.

  • Digital or analog controls can also be used to control operating parameters of a motor associated with the hover engine. In particular, the power received by the motor and its associated RPMs can be controlled. As is described with respect to figure. 4, the relative amount of lift and drag generated by a hover engine can be changed in response to a change in an RPM value associated with a hover engine. Next, examples of magnet arrangements, which can be used with a hover engine, are described with respect to FIGUES. 2 and 3.

  • FIGUES. 2 and 3 are illustrations of eddy currents generated on a conductive plate in response to arrangements of magnets rotated above the plates. The conductive plate is the portion of the substrate which is configured to support induced eddy currents. The eddy currents and associated forces which are generated were simulated using Ansys Maxwell 3D (Canonsburg, Pa.). In each of the simulations, an arrangement of magnets is rotated at 1500 RPM at ½ inches height above copper plates 56 et 64, respectivement. The copper plates are modeled as ½ inch thick. The plate is modeled as being homogeneous in depth and from location to location. The width and length of the plate is selected such that edge effects that can occur when a STARM induces eddy currents near the edge of the plate are minimal.

  • The magnets are one inch cube Neodymium alloy magnets of strength N50, similar magnets can be purchased via K and J magnetics (Pipersville, Pa.). The magnets weigh about 3.6 ounces each. Magnets of different sizes, shapes and materials can be utilized and this example is provided for the purpose of illustration only.

  • Dans figure. 2, eight one inched cube magnets, such as 50, are arranged with an inner edge about two inches from the z axis. The magnets are modeled as embedded in an aluminum frame 52. The arrow head indicates the north pole of the magnets. The polarities of four of the magnets are perpendicular to the z axis. The open circle indicates a north pole of a magnet and circle with an x indicates a south pole of a magnet. A polarity pattern involving four magnets is repeated twice.

  • In various embodiments, the polarity pattern of the magnets shown in the figure can be repeated one or more times. One or more magnets of different sizes and shapes can be used to form a volume of magnets which match a polarity direction associated with a polarity pattern. For example, two one half inch wide rectangular magnets with a total volume of one cubic inch or two triangular magnets with a total volume of one cubic inch can be aligned in the same direction to provide a polarity direction in a polarity pattern. In the polarity pattern, a magnets with a polarity direction different than an adjacent magnet may touch the adjacent magnet or may be separate from the adjacent magnet.

  • For a given number of magnets of a particular cubic size, the distance from the z axis of the face of the magnets can be adjusted such that the magnet's edges are touching or are a small distance apart. With this example using eight magnets, an octagon shape would be formed. A configuration of twenty one inch cube magnets arranged around a circle with the polarity pattern shown in figure. 2 repeated five times is described with respect to FIGUES. 8A and 8B. The inner edge of this arrangement of magnets is about 3.75 inches from the rotational axis.

  • When the magnets are brought together, the magnitude of the lift and drag which is generated per magnet can be increased relative to when the magnets are spaced farther apart. In one embodiment, trapezoidal shaped magnets can be utilized to allow the magnets to touch one another when arranged around a rotational axis. A different trapezoidal angle can be used to accommodate different total number of magnets, such as four magnets (90 degrees), eight magnets (45 degrees), etc.

  • A combination of rectangular and triangular shaped magnets can also be used for this purpose. For example, triangular magnets can be placed between the cubic magnets shown in figure. 2. In one embodiment, the polarity pattern for groups of four trapezoidal magnets or combinations of rectangular and triangular magnets can be similar to what is shown in figure. 2.

  • When the arrangement of eight magnets is rotated above the copper plate, eddy currents are induced in the copper. In the example of figure. 2, the simulation indicates four circular eddy currents 56 are generated. The four eddy currents circle in alternating directions and are approximately centered beneath the circulating magnets. A current circulating in a circular coil generates a magnetic field which looks like a magnetic field of a bar magnet where the orientation (north/south) depends on the direction of the current. The strength of the magnetic field which is generated depends on the area of the circular coil and the amount of current flowing through the coil.

  • A reaction occurs where the circulating eddy currents generate a magnetic field which repels the arrangement of magnets such that a lifting forces and drag forces are generated. The eddy currents rotate as the magnets rotate. However, the eddy currents are not directly underneath the four magnets aligned with the z axis. Thus, the eddy currents can generate a magnetic field which attracts one of the poles of permanent magnets to which it is adjacent. The attractive force can act perpendicular to the lift to produce drag, which opposes a movement of the magnets. The drag can also be associated with a torque. The drag torque is overcome by an input torque supplied by a motor coupled to the arrangement of magnets.

  • In this example, there are not well defined circuits. Thus, one eddy current can interact with an adjacent eddy current. The interaction causes the magnitude of the current to increase at the interface between eddy currents such that magnitude of the current varies around circumference of each eddy current. Further, the current also varies in depth into the material with the greatest current per area occurring at the surface and then decreasing in depth in to the surface.

  • In addition, unlike circuits with a fixed position, the center of the eddy currents rotate as the magnets inducing the currents rotates. Unlike when a magnetic is moved linearly over a conductive material, separate eddy current forms in front of and behind the magnet. In this example, the four poles (magnets with north and south perpendicular to the surface of the plate) are close enough such that the eddy current formed in front of one pole merges with the eddy current formed behind the next adjacent pole. Thus, the number of eddy currents formed is equal to the number of poles which is four. In general, it was observed for this type of configuration that the number of eddy currents which formed was equal to the number of poles used in the magnet configuration.

  • In addition, material interfaces can affect the induced eddy currents such that an amount of lift and drag which is generated is different near the interfaces as opposed to away from the interfaces. For example, a surface on which eddy currents are induced can have edges where the material which supports the induced eddy currents ends. Near the boundaries, when a STARM approaches an edge, the eddy currents tend to get compressed which affects the resultant lift and drag.

  • In another example, a surface can have interfaces through which there are discontinuities in the conductivity. For example, edges of two adjacent copper sheets used to form a surface may not touch, may partially touch or may be conductively insulated from one another. The discontinuous conductivity can lessen or prevent current from flowing across the interface which affects the lift and drag generated from the induced eddy currents.

  • In one embodiment, a substrate which supports induced eddy currents can be formed from a number of sheets which are stacked in layers, such ⅛ inch copper sheets stacked on top of one another. A discontinuity may be formed in one layer where two adjacent sheets meet, such as small gaps between the two sheets which reduce the current which flows from a first sheet to an adjacent second sheet. The gaps may allow for thermal expansion and simplify the assembly process. To lessen the effect of the discontinuity, adjacent edges between sheets can be staggered from layer to layer. Thus, the discontinuity at particular location may occur in one layer but not the other adjacent layers.

  • In some instances, a conductive paste can be used to improve the conductivity between sheets. In another embodiment, adjacent sheets can be soldered together. In yet another embodiment, flexible contacts, which can be compressed and then expand, can be used to allow current to flow between different sheets.

  • Dans figure. 3, a three row by five column array of one inch cube magnets, such as 60, is rotated above a copper plate. The magnets are modeled as surrounded by an aluminum frame 62. The magnets in this example are configured to touch one another. A magnet pattern for each row of five magnets is shown. In alternate embodiment, a five magnet pattern of open circle, left arrow (pointing to open circle), circle with an “x”, right arrow (pointing away from circle with an x) and open circle can be used. This compares to the left arrow, circle with an “x”, left arrow, open circle and right arrow pattern shown in the Figure.

  • The magnet pattern is the same for each row and the magnet polarity is the same for each column. In various embodiments, a magnet array can include one or more rows. For example, a magnet array including only one row of the pattern shown in figure. 3 can be used.

  • Multiple arrays with one or more rows can be arranged on a rotating body, such that the rotating body is balanced. For example, magnet arrays of two, three, four, etc. arrays of the same number of magnets can be arranged on a rotating body. In another embodiment, two or more pairs of magnet arrays with a first number of magnets and two or more pairs of magnets arrays with a second number of magnets can be arranged opposite one another on a rotating body.

  • In the example of figure. 3, two eddy currents, 66, are generated under the magnet array and two eddy currents 70 et 68 are formed ahead and behind the array. These eddy currents move with the array as the array rotates around the plate. As the array is moved over the plate 64, eddy currents, such as 72 spin off. The eddy currents 66, 68 et 70 generate magnetic fields which can cause magnetic lift and drag on the array. When two of these types of arrays placed close to one another, the simulations indicated that the eddy current induced from one array could merge with the eddy current induced from the other array. This effect diminished as the arrays were spaced farther apart.

  • In the examples of FIGUES. 2 and 3, the simulations indicated that more lift force was generated per magnet in the configuration of figure. 3 as compared to figure. 2. Part of this result is attributed to the fact that a portion of the magnets in figure. 3 is at a greater radius than the magnets in figure. 2. For a constant RPM, a greater radius results in a greater speed of the magnet relative to the conductive plate which can result in more lift. If it is desirable to arrange the magnets around a motor as shown in FIGUES. 8A and 8B and use a disk of smallest diameter, then the configuration in figure. 2 is more suitable. Thus, in some embodiments, it may be desirable to trade lift generating efficiency for packing efficiency when selecting an arrangement of magnets to utilize.

  • The lift per magnet can be total lift divided by the total magnet volume in cubic inches. For one inch cube magnets, the volume is one cubic inch. Thus, the total number of magnets is equal to the volume in cubic inches. Hence, the use of lift force per magnet in the previous paragraph. The use of total lift divided by the magnet volume of a magnet arrangement provides one means of comparing the lift efficiency of different magnet arrangements. However, as noted above, the speed of the magnet relative to the substrate, which is a function of radius and RPM, effects lift and hence may be important to consider when comparing magnet configurations.

  • Dans FIGUES. 2 and 3, a portion of the magnet poles in the magnet polarity pattern are aligned such that the poles are parallel to an axis of rotation of the STARM (The poles labeled with “x” or “o” in the Figures). When the bottom of a STARM is parallel to a surface which supports the induced eddy currents, the portion of the magnet poles and the axis of rotation are approximately perpendicular to the surface. In an alternate embodiment, as described in more detail with respect to FIGUES. 24 and 25, the portion of the magnet poles in FIGUES. 2 and 3 which are substantially parallel to the rotation axis can be rotated ninety degrees inwards or outwards such the magnet poles become perpendicular to the axis of rotation.

  • In this configuration, to interact with a surface, a STARM can be rotated on its side, like a tire riding on a road, where the axis of rotation is approximately parallel to the surface. In particular embodiments, a mechanism, such as an actuator, can be provided which can dynamically rotates one or more of the magnet poles (again, “x” and “o” labeled magnets) during operation. For example, the magnet poles shown in FIGUES. 2 and 3 may be rotatable such that they can be moved from an orientation where they are perpendicular to the surface as shown in FIGUES. 2 and 3 to an orientation where they are parallel to the surface and back again. When the magnets are turned in this manner, the amount of lift and drag which are generated can be reduced. In additional embodiments, fixed magnet configurations can be utilized where the magnet poles shown in FIGUES. 2 and 3 are rotated by some angle between zero and ninety degrees relative to their orientation in the FIGUES. 2 and 3.

  • figure. 4A includes a plot 100 of lift 106 and drag 108 curves associated with an arrangement of rotating magnets in accordance with the described embodiments. The curves are force 102 versus rotational velocity 104. The curves can be determined via experimental measurements and/or simulations. It is noted the magnetic lift and drag is separate from any aerodynamic lift and drag which may be associated with the rotation of magnet arrangement associated with hover engine.

  • Although not shown, an amount of torque can be determined and plotted. As shown in figure. 2, an array of magnets can be radially symmetric. In some instances, such as when a radially symmetric array is parallel to the conductive substrate, the net drag force may be zero. Nevertheless, a torque which opposes the rotation of the array is generated. The rotational input from a motor can be used to overcome the torque.

  • As shown in figure. 4A, the magnetic drag increases as velocity increases, reaches a peak and then starts to decrease with velocity. Whereas, the magnetic lift increases with velocity. The velocity can be the velocity of the magnets relative to the surface which induces the eddy. When the magnets are rotating, this velocity is product of a distance from the axis of rotation times the angular velocity. The velocity can vary across a face of a magnet as distance from the axis of rotation varies across the face of the magnet.

  • In various simulations of a magnet configuration shown in figure. 3, the most drag was observed to occur between 250 and 350 RPM. However, the amount of drag including its peak can depends on such variables as the size and the shape of the magnets, a distance of the magnets from the substrate in which the eddy currents are induced, a speed of the magnets relative to the substrate which changes as a function of radius and a thickness of the substrate and a strength of the magnets. Also, for an arrangement of a plurality of magnets, the arrangement of their poles and spacing relative to one another can affect both the lift and drag, which is generated. Thus, the value range is provided for the purposes of illustration only.

  • figure. 4B is a plot of force 102 associated with an arrangement of rotating magnets as a function of distance 110 from a conductive substrate. In this example, a configuration of magnets similar to shown in figure. 3 was simulated. The plot is based upon a number of simulations at a constant RPM. The lift appears to follow an exponential decay curve as the distance from the surface 110 augmente.

  • figure. 4C is a plot of lift curves associated with an arrangement of rotating magnets as a function a thickness of a conductive substrate and RPM. In this example, a configuration similar to what is shown in figure. 3 was used. The conductive substrate is copper and thickness of the copper is varied between 0.05 and 0.5 inches in the simulation.

  • The simulations predicted that the amount of generated lift begins to decrease after a certain threshold thickness of copper is reached and is relatively constant above the threshold. The location of the threshold varies as a function of RPM. It may also vary according to the magnet configuration. In one simulation, negative lift was predicted, i.e., an attractive force was generated when the thickness was thin enough.

  • FIGUES. 5A and 5B are block diagrams illustrating hovering and propulsive effects from rotating arrangements of magnets. Dans figure. 5A, a motor 122 is coupled to a STARM 124. The rotatable components with magnets can be referred as a STARM (STator and ARMature). The motor 122 and STARM can be used as part of a hover engine.

  • The STARM 124 is coupled to the motor 122 and the motor 122 is coupled to a rotatable member 128. The rotatable member 128 is coupled to anchors 126 une et 126 b. The combination of the rotatable member 128 and the anchors 126 une et 126 b can be configured to constrain a range of rotation of the rotatable member. For example, the rotatable member 128 may be allowed to rotate through some angle range 134 around its axis.

  • The rotatable member 128 can be configured to receive and input torque from some mechanism. For example, in one embodiment, a mechanical linkage can be provided which allows a user to supply a force. The force can be converted into torque which causes the rotatable member 128 and hence the motor 122 and the STARM 124 to rotate.

  • In another embodiment, an actuator can be used to supply the torque to rotate rotatable member 128. An actuation of the actuator can cause the motor 122 and STARM 124 to tilt relative to the substrate 136. The actuator can include a servo motor which receives control commands from a controller. In one embodiment, the actuator can include its own controller which receives control commands from a separate processor, which is part of the control system.

  • In yet another embodiment, a hover engine can be configured to receive an input force from a user and can include an actuator. The actuator can be used to change a position of the STARM, such as returning it to a designated position after a user has tilted it. In another operation mode, the actuator can be used to provide automatic control around some tilt position initiated by user via an input force.

  • It yet another embodiment, the actuator can be used to provide automatic controls which may be used to correct a control input from a user. For example, if the control system detects that the board is about to slide out from underneath the user after a user provides a control input, the control system can control one or more STARMs to prevent this event from happening. The hoverboard can include one or more on-board sensors used to make these corrections.

  • For example, sensors, such as accelerometers, which measure translational and rotational accelerations, can be used to determine an orientation of the hoverboard and/or a position of a user relative to the hoverboard, which is used to generate an orientation correction in response to the user control input. In one embodiment, a first set of sensors, such as accelerometers, can be worn by a user, such as at their waist and a second set of sensors, such as accelerometers, can be located on the hoverboard. A processor, such as one onboard the hoverboard can be configured to receive the sensor data. The sensor data can be used to determine an orientation of the hoverboard relative to the person and possibly to make adjustments to a hover engine.

  • The hoverboard may also include one or more weight sensors for determining where a user is standing on the hoverboard. The weight distribution associated with the hoverboard and rider can affect the response of the hoverboard in response a command to change an orientation of the hoverboard via some mechanism, such as a tiltable hover engine. For example, the weight distribution associated with where a person is standing effects the magnitude of rotational moments. Thus, knowledge of the weight distribution may be used to more finely tune the commands used to control the orientation of the STARM, such as selecting which STARM to actuate and an amount to actuate it.

  • In one embodiment, a user riding the hoverboard can control actuation of the actuator via some action, such as a body movement or interaction with a control device, such as a foot pedal or control stick. The body movement or interaction with the control device can be translated by an on-board processor into one or more commands which are sent to the controller associated with the actuator. In response, the actuator can rotate the rotatable member by some amount.

  • As an example, the body movement can be characterized via a sensor in a device worn or held by the person. The data received from the sensor can be processed locally in the device worn or held by the person and/or via an onboard processor associated with the hoverboard. The processed data can be used to generate control commands which are sent to one or more actuators, which can tilt a STARM, or some other mechanism for generating a propulsive force.

  • In additional embodiments, a rider doesn't necessarily have to stand on the hoverboard. For example, a rider can sit on the board, lie on their back on the board or lie on their stomach on the board. Control mechanisms can be provided which enable a rider to control the board in these positions. In addition, structure can be provided which help support a rider in one of these positions. For example, a seat can be coupled to the hoverboard.

  • When the STARM 124 and motor 122 are rotating, a rotation of the rotatable member 128 changes the angular momentum of the STARM and the motor. It can also change the magnetic forces acting on the STARM 124 as the magnetic forces vary with the distance of the magnets in the STARM 124 from the substrate 136. Therefore, the amount of torque needed to rotate the member 128 can depend on the moment of inertia associated with the STARM 124 and motor 122, how fast the STARM 124 and motor 122 are spinning and the height of the STARM 124 above the substrate 136. The height of the STARM above the substrate can depend on its velocity, which affects how much lift is generated as well as a payload weight, i.e., the weight of a hoverboard rider and how the riders' weight is distributed on the hoverboard. The height of the STARM above the substrate can vary for different portions of the STARM and from STARM to STARM when a hoverboard includes multiple STARMs.

  • In the example of figure. 5A, the STARM 124 is approximately parallel to the substrate 136. The magnetic drag, such as 132 une et 132 b, opposes the rotation of the STARM 124. The motor 122 is configured to rotate in the clockwise direction 130. Thus, the drag torque is in the counter clockwise direction. Power is supplied to the motor 122 to overcome the drag torque.

  • When the STARM is parallel to the substrate 136, the magnetic drag is balanced on all sides of the STARM 124. Thus, there is no net translational force resulting from the magnetic drag. As is described with respect to figure. 5B, a net translational force is generated when the STARM 124 is tilted relative to the substrate.

  • Dans figure. 5B, the STARM 124 is in a titled position 140. Thus, one side of the side of STARM 124 is closer to the substrate 136 and one side of the STARM 124 is farther away from the substrate 136. The magnetic interaction between the magnets in the STARM 124 and substrate decreases as a distance between the magnets in the STARM and substrate 136 augmente. Thus, in tilted position 140, the drag force 138 b is increased on one side of the STARM 124 and the drag force 138 une is reduced on the opposite side of the STARM 124 as shown in figure. 5B. The drag force imbalance creates traction, which causes a translational force to be generated approximately in the direction of the axis of rotation of the rotational member 128.

  • When the STARM 124 is initially tilted, the translational force can result in an acceleration of the STARM 124 in the indicated direction and hence change in velocity in the indicated direction. In particular embodiments, with one or more STARMs configured to generate translational forces, a hoverboard can be configured to climb. In another embodiment, the hoverboard may be configured to maintain its position on a slope while hovering such that the gravitational forces acting on the rider and the board are balanced by the translational forces generated by the hoverboard and its associated hover engines.

  • A configuration and operational mode where a position of the hoverboard is maintained on a sloped substrate may be used as part of a virtual reality system where a rider wears a virtual reality headset. Via the headset, the user may only see images generated by the headset or may see images generated by the headset in conjunction with the local surrounding visible to the user. A virtual reality headset may be used to generate images of a user moving through some terrain, likes a snowy slope, while the hoverboard moves side to side and forward and back on the sloped substrate. The sloped substrate may provide the user with the feeling of moving on a tilted slope while the virtual reality images may provide the visual imagery associated with movement. Fans may be used to add an additional sensation of movement (e.g., the feeling of wind on a person's skin).

  • The hoverboard can have sufficient propulsive ability to allow it to hold its position on the slope against the force of gravity. For example, the hoverboard can be moved side to side while it maintains its position on the slope. Further, the hoverboard may be able to move downwards on the slope and then climb upwards on the slope against gravity. In some instance, the climbing can be done while the board's orientation remains relatively unchanged, i.e., the board doesn't have to be turned around to climb. This maneuver can be accomplished by changing an orientation of the hover engines relative to the substrate which supports the induced eddy currents.

  • Returning to FIGUES. 5A and 5B the amount of tilt in a particular direction can affect the amount of force imbalance and hence the magnitude of the acceleration. Because the magnetic drag is function of the distance of the magnets from the substrate, the magnetic drag increases on the side closer to substrate and decreases on the side father away from the substrate. As the magnetic forces vary non-linearly with the distance of the magnets from the surface, the amount of translational forces which are generated may vary non-linearly with the tilt position of the STARM.

  • After a STARM 124 (or both the STARM 124 and motor 122) has been rotated via member 128 in a counter clockwise direction and the STARM has started translating in a first direction, an input torque can be provided which tilts the STARM in a clockwise direction to reduce the amount of translational force which is generated by the STARM. When the STARM is tilted past the horizontal in the clockwise direction, the STARM may generate a translational force which is in an opposite direction of the first direction. The translational force opposing the direction of motion can slow the STARM and bring it to rest. If desired, the translational force can be applied such that the hoverboard stops and then the STARM can begin to translate in an opposite direction.

  • Hoverboard and Hover Engine Configurations

  • With respect to FIGUES. 6-25, a number of configurations of a hoverboard with hover engines are described. In particular, hover engine configurations and their integration into a hoverboard system are described. In addition, a number of different mechanisms, which enable a force to be transferred to a hover engine to cause an orientation of the hover engine to change, are described.

  • figure. 6 is a bottom view of a hoverboard 200 with hover engines with mechanisms which provide for tilting along a specified axis. The hover engines are coupled to a support structure 202. In one embodiment, the support structure 200 may also be used to provide a rider platform. The hoverboard 200 includes four hover engines, such as 206. The hover engines each include a motor (not shown) and a STARM 204.

  • A location 212 of a center of mass of the hoverboard 200 is shown. This location can denote the location when the hoverboard is unloaded, i.e., it doesn't include a rider. When a rider is riding the hoverboard 212, the individual can change their position on the hoverboard to shift the location of the center of mass and hence change various rotational moments associated with the hoverboard 200. Thus, the location 212 is shown for illustrative purpose only and is not meant to be limiting.

  • An axis 208 is shown drawn through the center of mass 212. A rider can shift their weight such that a moment is generated and the hoverboard 200 tilts around an axis, such as 208. The rider can change shift their weight in many different ways which can result in a moments in different directions being generated and a corresponding tilt of the board along different axes. Thus, the example of axis 208 is shown for illustrative purposes only.

  • In particular embodiments, a rider may be able to freely move their position over a top surface of the hoverboard 200. In another embodiment, the hover board can include foot restraints of some type, such as a strap which a rider can insert their feet through. When a rider's feet are constrained, the amount of movement of the center of mass and the possible moments which are generated may also be constrained.

  • A change in position can cause the hoverboard to tilt. The change can include a dynamic component where the hoverboard shifts between different orientations and then a relatively steady state orientation where the tilt position of the hoverboard is steady. Thus, the dynamic and steady state tilt position of the hoverboard can change the height positions of the hover engines relative to ground. The change in height positions can cause the hover engines to generate a net translation force which allows the hoverboard to be propelled in a particular direction.

  • While hovering, the lateral forces constraining the hoverboard can be very small. Thus, when a person shifts their weight in a particular direction, the hoverboard 200 will move in response and can slide out from underneath the person. Thus, the hoverboard can differ from a device, such as a skateboard or ice skates, where the lateral movement is constrained, due to the friction generated between the wheels or blades, with the ground or ice.

  • In a particular embodiment, to provide an ability of a rider to control forces on the hoverboard, the hoverboard can include one or more hover engines each with a tilt mechanism which allows the hover engines to be tilted in response to an input force generated by the rider. As described above, the input force can be used as part of an analog or digital control system. Additional details of different tilt mechanisms are described with respect to figure. 9-21. Alternate embodiments, where the orientation of the hover engines on the hoverboard are fixed, are described with respect to FIGUES. 22A-25.

  • In the example of figure. 6, tilt axes, such as 216, are shown through each STARM, such as 204. A mechanism is provided which tilts each STARM about at least one axis. As described with respect to figure. 14, a mechanism can be provided which allows a tilt of the STARM about multiple axes. Thus, the example in figure. 6 is provided for illustrative purposes only.

  • The tilt axes, such as 216, are used to describe a tilt of the hover engine relative to the hoverboard 200, such as relative to a plane parallel to a top surface of the hoverboard. As described above, the entire hoverboard can also be tilted relative to a substrate. For example, the rider, by shifting their weight, can tilt the hoverboard 200 about axis 210. Thus, the tilt of the hover engine, relative to the substrate and the associated height of the magnets in the hover engine above the substrate at particular time, depends on both the tilt of the hover engine relative to the hoverboard and the tilt of the hoverboard relative to the substrate.

  • Dans figure. 6, the tilt axes, 216, are configured to provide lateral spin control. When a rider shifts their weight to one of the corners, the hoverboard 200 can tend to keep moving in the direction where the rider shifted their weight which can cause the board to slide out from underneath the rider. The tilt mechanism can be configured to generate a force, such as 218, which negates this effect. Thus, the rider may continue to more easily ride board without falling off.

  • In the example of figure. 6, the forces, such as 218, which result from a tilt of a hover engine, are shown approximately perpendicular to a side of the hoverboard 200 and parallel to the tilt axes, such as 216. Further, the tilt axes are all parallel to one another. In various embodiments, the tilt axes don't have to be orientated in this manner, such as perpendicular to a side of the hoverboard 202 or all parallel to one another and this example is provided for illustrative purposes only. Thus, the tilt axis can be position at various angles relative to one another and the hoverboard 200.

  • In addition, not all of the hover engines may include tilt mechanisms which allow the hover engine to be tilted relative to the hoverboard. Thus, the tilt of some hover engines may be rider (or system) controllable while other hover engines may be configured in a fixed position and may not be tilted. For example, the hoverboard may include a large hover engine in the center which is configured for primarily generating lift and then may be surround by a plurality of smaller hover engines which are used to generate control forces. The larger hover engine can be installed in a fixed position such that its orientation relative to the hoverboard doesn't change whereas the smaller hover engines may be coupled to a tilt mechanism which allows them to tilt relative to the hoverboard. Both the larger and the smaller hover engines may be tilted relative to a substrate in response to dynamic changes in the center of mass of the hoverboard and rider system. An advantage of this approach is that the smaller hover engines may have a smaller moment of inertia and hence may be more easily tilted about an axis by a person or a by an actuator. In the case of the actuators, a smaller moment of inertia may allow for less powerful actuators to be employed.

  • In yet another embodiment, the hoverboard 200 can include an automatic control system. The automatic control system can be configured to detect movement in a particular direction. An acceleration, a velocity or a change in a position may be detected using an on-board sensor package. In response, a logic device, such as an on-board processor, can be configured to actuate one or more tilt mechanisms as a function of time to generate some complementary movement. For example, a shift in weight and a subsequent movement of the board can be detected. In response, the automatic control system can actuate one or more of the tilt mechanisms as a function of time to keep the board from sliding out from underneath the person.

  • In various embodiments, a tiltable hover engine may be moved only in response to rider inputs, moved only in response to commands from a control system or combinations thereof. For example, an actuator coupled to a hover engine can be configured to actuate in response to a control input generated by a rider at some times and at other times in response to control input generated by a control system. The control system may receive input from a sensor package to generate control signals which augment the control inputs from the rider. Further, an orientation of one or more first tiltable hover engines on a hoverboard may be only controlled by a rider, such as via a manual actuation from the rider, whereas an orientation of one or more second tiltable hover engines, tilted via actuators, on the hoverboard may only be controlled by a control system implemented on a logic device on the hoverboard. In another embodiment, a single hover engine may be controlled in response to user inputs at one time and then at another time be controlled based upon commands generated by the control system, which are generated independently from the user. For example, the control system may be configured to shut down a hoverboard and bring it to rest independently of any direct control inputs provided by a user.

  • The tilt mechanisms can be coupled to the hover engine, such that the amount of force and the direction of the force can be varied over a range. For example, the tilt mechanism can be configured to cause a force 218 which is only toward the vehicle in a range of zero to some maximum value. In another embodiment, the tilt mechanism can be configured to generate a force which can be directed toward the vehicle or outward from the vehicle where the maximum value in each direction may be substantially equal or the maximum value in one direction may be greater than the other direction. In yet another embodiment, which is described with respect to figure. sept as follows, the tilt mechanism can be configured to generate a force which is between a minimum value greater than zero and a maximum value.

  • figure. sept is a side view of a hover engine 230 coupled to a tilt mechanism in a tilt position. The hover engine includes a motor 122 and a STARM 124 which are positioned over the substrate 136. In one embodiment, the mechanism can include a minimum tilt off set angle 234. The minimum tilt off set angle 234 in this example is between the horizontal and line 232. The tilt range angle 236 is the angle amount through which the hover engine may rotate starting at the minimum tilt off set angle 234. The tilt mechanism can include one or more structures which constrain the motion of the tilt mechanism to the tilt angle range.

  • When the minimum tilt off set angle 234 is zero and the STARM 124 is parallel to the substrate 136, the STARM 124 may not generate a net translation force. As described above, during operation, a hoverboard to which a STARM is coupled can be tilted. Therefore, the angle of the STARM relative to the substrate can depend on the orientation of the STARM relative to some reference system associated with the hoverboard and the orientation of the hoverboard relative to the substrate where both orientations can change as a function of time. Thus, in some instances, a translation force can be generated even when the minimum tilt off-set is zero. When the minimum tilt off set angle is greater than zero, the STARM may generate a net translational force at its minimum position in a particular direction. When the minimum tilt off set angle is less than zero, then during the tilt angle range the magnitude of the force may be go to zero and the direction of the force which is generated can also change.

  • In some embodiments, the net minimum force generated by one hover engine can be balanced in some manner via translational forces associated with other hover engines. For example, as shown in figure. 6, two hover engines can be tilted to generate forces in opposite directions to cancel one another. Thus, although the net force for a single hover engine may be greater than zero at its minimum tilt off set angle position, it can be balanced by forces generated from another STARM such that the net force acting on the hoverboard is zero.

  • The forces which are generated from a tilted STARM can vary non-linearly with angle of the hover engine relative to the substrate. Thus, the change in force which is generated as a function of a change in angle can vary non-linearly. By utilizing, a minimum tilt angle offset, the hover engine can be configured to output more or less force in response to a change in a tilt angle over a selected tilt angle range. In this manner, the control characteristics of the hoverboard can be adjusted.

  • In one embodiment, the tilt mechanisms can include an adjustable tilt off set mechanism that allows the minimum tilt off set angle to be manually set. For example, a rotatable member with a protuberance can be provided where the protuberance is configured to impinge on a screw at one end of its range of rotation. As the screw is unscrewed, the range of rotation of the rotatable member can be decreased and the minimum tilt off set angle can be increased and vice versa. Using the adjustable tilt off set mechanism, a user or operator may be able to manually adjust the handling characteristics of the hoverboard.

  • In other embodiments, the motor 122 and STARM 124 can be tilted up to ninety degrees, such that side 152 is parallel to a top of the substrate 136. The side 152 and a portion of side 150 can be covered to allow contact the top of the substrate in this position. For example, a rubber tire can be used which covers side 152 and a portion of side 150. In yet another embodiment, the STARM 124 can have additional magnets with poles which point outwards through side 152. In yet another embodiment, a portion of the magnets on STARM 124 can be rotatable to allow the magnet poles to rotate from a position where they intersect side 150 to a position where they intersect side 152. For example, the magnet poles can range from a position where they are perpendicular to side 150 to a position where they are perpendicular to side 152.

  • figure. 8A is a perspective view of a STARM 400. The STARM 400 is 10 inches in diameter. In various embodiments, the STARMs used on a hoverboard can be between four and fourteen inches in diameter. The STARM includes a raised outer ring 405. A distance from a bottom of the STARM 400 to a top of the outer ring is about 1.13 inches. This height allows one inch cubed magnets to be accommodated. In one embodiment, twenty one inch cube magnets are arranged within the outer ring. In particular embodiments, the volume of magnets on the hoverboard, which can be distributed among one or more STARMS, can be between thirty and eighty cubic inches.

  • In one embodiment, the STARM 400 including the outer ring 405 can be formed from a number of layers, 402, 408, 410, 412, 404 et 414, from top to bottom, respectively. Layers 402 et 414 form a cover over the top and bottom portions of the magnets in the outer ring. In one embodiment, layers 402 et 408 are about 0.065 of an inch thick. In alternate embodiment, one or both of layers 402 et 408 can be eliminated. In one embodiment, the top and bottom layers can be formed from a material such as aluminum. In another embodiment, the top layer 402 can be formed from a material with magnetic properties, such as mu-metal, iron or nickel.

  • Layers 408, 410, 412, 404 each include twenty apertures to accommodate twenty magnets. More or less magnets and hence more or less apertures can be utilized and this example is provided for illustrative purposes only. The total thickness of the layers is one inch and each layer is 0.25 inch thick. In one embodiment, two layers are formed from polycarbonate plastic and two layers are formed from aluminum. The polycarbonate plastic can reduce weight. In various embodiments, the thickness of each layer, the material used for each layer and the number of layers can be varied. For example, different metals or types of plastics can be used. As another example, a single material can be used for each of the layers.

  • When the layers are aligned, the one inch cube magnets can be inserted through the layers. For different shaped or different size magnets, such as rectangular shaped magnets, trapezoidal shaped magnets or 1.5 cubic inch magnets, a different aperture shape or size can be used. In one embodiment, an adhesive can be used to secure the magnets in place, such as super glue. When secured, the bottoms of the magnets are approximately flush with the bottom of layer 404. This feature can maximize the height between the bottom of the magnets and the substrate when a vehicle using the STARM design 400 is hovering.

  • One or more layers can include apertures, such as 416, that allow fasteners to be inserted. The fasteners can secure the layers together. In another embodiment, an adhesive can be used to secure one or more of the layers to one another. In alternate embodiment, the layers 404, 408, 410 et 412 can be formed as a single piece.

  • figure. 8B is a side view of STARM 420 with an embedded motor 422. The cross sections of two magnets, 415, are shown within the outer ring 405. The top of the magnets is flush with the outer top of layer 408 and the bottom of the magnets is flush with the bottom of layer 404. In various embodiments, the STARM 420 can be configured to receive magnets between 0.5 and 2.5 inches of height.

  • In one embodiment, the top of the magnets may extend above the top of the 408. Thus, the outer ring 405 may only extend partially up the sides of each magnet. This feature may allow the magnets to be secured in place while reducing weight.

  • The inner radius 424 of the outer ring 405 is greater than a radius of the motor 422. Thus, the motor can be inserted within the outer ring and secured to layer 404 such that the STARM 420 can be rotated when the motor is operated. Thus, the outer ring extends along the side 430 of the motor. An advantage of mounting the motor in this manner is that the overall height profile of the hover engine may be reduced as compared to mounting the motor 422 at a height above the top of the outer ring.

  • In various embodiments, the height 428 height of the outer ring may be less than the height of the motor 426, such that the outer ring extends partially up the side 430 of the motor 422. In another embodiment, the height 428 of the outer ring 405 and the height of the motor can be approximately equal. In yet another embodiment, the height 428 of the outer ring can be greater than the height of the motor.

  • It may be desirable to increase the height 428 to accommodate taller magnets. Taller magnets may be used to increase the amount of magnetic lift which is generated when the magnets, such as 415 are at a greater distance from a substrate. The volume of a magnet including its height can affect the strength of the magnetic field at a particular distance which extends from a magnet.

  • The motor, such as 422, used to rotate a STARM can be electric or combustion based. In general, any type of motor which outputs a suitable amount of torque can be used. An electric motor requires a power source, such as battery or a fuel cell, to supply electricity. A combustion motor requires a fuel which is combusted to operate the motor. Battery types include but are not limited to batteries with a lithium or zinc anode, such as lithium ion, lithium polymer or a zinc-air system.

  • An electric motor can be configured to output torque about a rotational axis. The electric motor can include a configuration of wire windings and a configuration of permanent magnets. Current is provided through the windings to generate a magnetic field which varies as a function of time. The magnetic field from the windings interacts with magnetic field from the permanent magnets to generate a rotational torque. AC or DC motors can be utilized, such as an induction motor or a DC brushless motor.

  • In various embodiments, the windings can be configured to rotate while the magnets remain stationary or the magnets can be configured to rotate while the windings remain stationary. An interface, such as a shaft, can be provided which couples the rotating portion of the motor to the STARM 400. Dans figure. 8A, the STARM 400 is configured to interface with the motor at 406.

  • The non-rotating portion of the motor 422 can be integrated into a motor housing which surrounds the magnets and the windings. The motor housing can include an interface which enables it to be attached to one more structures associated with the hoverboard. In another embodiment, non-rotating portion of the motor can include an interface which allows it to be directly attached to one or more structures associated with the hoverboard.

  • In a particular embodiment, the core of the motor 422 can be stationary where both the magnets associated with the motor and the magnets associated with the STARM rotate around the stationary core. One non-rotating support structure can extend from the core which allows the motor and STARM to be coupled to the hoverboard. A second non-rotating support structure can extend from the core which provides support to a portion of a shroud which is interposed between a bottom of STARM and the substrate which supports the induced eddy currents (see figure. 8C). As is described in more detail with respect to FIGUES. 17A-17E, various shroud configurations may be used to surround a STARM.

  • The arrangement of magnets in the motor 422 can include poles which are substantially perpendicular to the axis of rotation of the motor (often referred to as a concentric electric motor) or can include poles which are substantially parallel to the axis of rotation of the motor (often referred to as an axial electric motor). In one embodiment, a winding configuration, such as the winding configuration associated with an axial motor, can be used to induce eddy currents in a substrate. In these embodiments, there are no rotating parts and the STARM and the magnets associated with an electric motor are eliminated. As part of a hover engine, the windings can be tilted relative to a hoverboard to generate control forces in a manner previously described above with respect to FIGUES. 5A and 5B.

  • In yet another embodiment, the magnets associated with the motor 422 can be removed and a motor winding can be designed which interacts directly with the magnets in the STARM. For example, a winding can be placed above magnets 415 to interact with the magnetic flux above the magnets or a winding can be placed around the outside of magnets 415 or around the inside of magnets 415. A current applied to the winding to cause the STARM to rotate. As described above, rotation of the STARM can cause eddy currents to be induced in a portion of a substrate.

  • As an example, the motor 422 can include an outer ring configured to rotate. The STARM 400 can mounted be to the outer ring of the motor 422 instead of to a shaft extending from the center of the motor. This type of motor design can be referred to as an outboard design. This feature may allow the portion of layers 404 et 412 within the inner radius 424 of the outer ring 405 to be removed such that the bottom of the motor is closer to the bottom of the outer ring 405. One advantage of this approach is that the overall height of the STARM 420 and motor 422 may be reduced.

  • In a particular embodiment, the outer ring 430 of the motor and the outer ring 405 of the STARM may be formed as an integrated unit. For example, the outer ring of the motor 422 can have a layer extending outwards from the side 430. The layer extending from the side 430 can include a number of apertures through which magnets can be inserted. Optionally, one or more layers with apertures, such as 408, 410 et 412, can be placed over the magnets.

  • In general, in a hover engine, the support structures associated with the STARM, the stator of the motor, the shroud and housing can be integrated with one another. For example, an enclosure for the motor and STARM can include an integrated shroud. In another example, the structure forming the rotor for the motor can be integrated with the structure for the STARM. In another example, all or a portion of the structure forming the stator of the motor can be integrated with a housing and/or shroud associated with the hover engine.

  • figure. 8C is a side view of a hover engine 450 having a STARM 465 integrated with a motor in accordance. The hover engine 450 includes a stationary core 456 with windings configured to interact with magnets 460 to rotate the magnets. The core is attached to the support structure 464. The support structure 464 can provide a first interface to attach the hover engine to a hover board. In addition, the support structure 464 can be coupled to a housing 452 which surrounds both motor and the STARM 465. The support structure 464 may be used to help maintain a gap between the bottom of the STARM 465 and the housing 452.

  • In one embodiment, a small protuberance 466 may be provided at the end of support structure 464. The small protuberance 466 can be formed from a metal or a material with a low friction coating, such as a Teflon coated material. The small protuberance can provide a small stand-off distance when the hover engine is near the ground, such as during take-off and landing. It can help prevent the STARM 465 from impinging the ground. In particular embodiments, the protuberance 466 can be coupled to a portion of the hover engine which rotates or a portion which remains static during operation.

  • The STARM 465 includes a structure 458 surrounds the magnets 454. As described above, the structure 462 surrounding magnets 460 and the structure 458 surrounding magnets 454 can be formed as a single piece. The magnets 454 et 460 may be shaped differently and have different sizes relative to one another.

  • In various embodiments, bearings (not shown) can be provided between the support structure 464 and the structure 458 to allow the STARM 465 to rotate about the stationary core. In lieu of or in addition to bearings between the STARM structure 458 and the support structure 464, bearings can be provided at one or more locations between the housing 452 and the structure 458. For example, bearings may be placed between the bottom of the STARM 465 and the housing 452 to help maintain the spacing between the housing 452 and the STARM 465 on the bottom of the STARM. In another example, a bearing may be placed between the side of the STARM and the side of the housing 452 to maintain the spacing between the inner side of the housing 452 and the side of the STARM.

  • In one embodiment, the height of the hover engine can be less than three inches. In another embodiment, the height of the hover engine can be less than two inches. In yet another embodiment, the height of the hover engine can be less than one inch. Next details of a hoverboard including a plurality of hover engines are described.

  • FIGUES. 9, dix et 11 are a bottom, top and side view of a battery powered hoverboard 300. Dans figure. dix, the hoverboard 300 includes four hover engines, 304 une, 304 b, 304 c et 304 . The hover engines are of equal size and identical to one other, i.e., similar motor, number of magnets, STARM diameter, etc. In alternate embodiments, additional or less hover engines can be provided where the size of the hover engines may vary from hover engine to hover engine.

  • The dimensions of the hoverboard 300 are about 37.5 inches long by 4.5 inches high by 18.5 inches wide. The weight of the hoverboard unloaded is about 93.5 pounds. The hoverboard can be configured to carry a rider weighing up to 392 pounds. In operation, the distance between the hoverboard 300 and the substrate which supports the induced eddy currents may vary depending on the weight of the rider.

  • Each hover engine includes a STARM, such as 325, with a motor (not shown) and engine shroud 318 with a gap between the shroud 318 and STARM 325 to allow for rotation. The STARM 325 is connected to the motor via connectors 322. The motor provides the input torque which rotates the STARM. In alternate embodiments, a single motor can be configured to drive more than one STARM, such as 325.

  • In one embodiment, the STARMs, such as 325 are 8.5 inches in diameter. The STARMs are configured to receive sixteen one inch cube magnets. Thus, the total volume of the magnets on the hoverboard is 64 cubic inches. In one embodiment, Neodymium N50 strength magnets are used. The magnets each weigh about 3.6 ounces (force). Therefore, the total magnet weight for one hover engine is about 3.6 pounds (force).

  • As described above, a total magnet weight and volume can vary from STARM to STARM mounted to hoverboard. For example, hover engines, 304 b et 304 , may each have STARMs with twenty one inch cube magnets and hover engines, 304 une et 304 c, may each have STARMs with twelve one inch cube magnets. In this embodiment, the STARMs are mounted beneath the motor as compared to within a center of the STARM as shown in figure. 8B.

  • In one embodiment, the motors can be a q150 DC brushless motor from Hacker Motor (Ergolding, Germany). The motor has a nominal voltage of 50 Volts and a no load current of 2 Amps. The weight is about 1995 grams. The speed is about 52.7/min. The RPM on eta max is about 2540. The torque on eta max is about 973.3 N-cm. The current on eta max is about 53.76 Amps. In one embodiment, the motor can be operated between about 1000 and 2300 RPMs. In different embodiments, motors can be operated at higher or lower RPMs, such up to 10,000 RPMs

  • The hover engines each have a shroud, such as 318. The shroud 318 partially encloses the STARM, such that a bottom of the STARM is exposed. A tilt mechanism 312 is coupled to the shroud 318 of each hover engine. The tilt mechanism 312 is coupled to a pivot arm 310. The hover engines 304 une, 304 b, 304 c et 304 are suspended beneath a support structure 302.

  • The pivot arms, such as 310, extend through an aperture in the support structure. The pivot arms can be configured to move and tilt the shroud in some manner. More complex mechanical linkages are possible and the pivot arm is provided for the purposes of illustration only and is not meant to be limiting. In general, a mechanical linkage configured to transfer a force can have one or more members and one or more pivot points.

  • The motors in each hover engine can be battery powered. In one embodiment, sixteen battery packs are used. The batteries are VENOM 50C 4S 5000MAH 14.8 Volt lithium polymer battery packs (Atomik RC, Rathdrum, Id.). Each battery weighs about 19.25 ounces. The dimensions of the batteries are 5.71 inches by 1.77 inches by 1.46 inches. The minimum voltage is 12 V and the maximum voltage is 16.8 V. Other types of batteries can be used and these are provided for the purposes of illustration and are not meant to be limiting.

  • The sixteen batteries are wired together in four groups of four batteries and each coupled to motor electronic speed controllers, such as 306 une et 306 b via connectors 316 une et 316 b to four adjacent battery packs. Connectors 316 c et 316 each connect to four batteries and to a motor electronic speed controllers, 306 c et 306 , which are stacked behind, 306 une et 306 b (see figure. 11). The wire connections for the batteries are not shown. The batteries are wired in series in this example to provide up to about 60 V to the electronic speed controllers.

  • In an alternate embodiment, an AC power source may be used. For example, a device, such as the hoverboard can be coupled to AC power source which powers AC electric motors. In this embodiment, the batteries can be removed and replaced with a power converter which is suspended beneath the structure 302.

  • The motor electronic speed controllers, 306 une, 306 b, 306 c et 306 are coupled to each of the four motors via wire bundles 308 une, 308 b, 308 c et 308 . The electronic speed controllers are stacked on top of one another. Hence, electronic speed controllers 306 c et 306 are not visible in figure. 9 (instead see figure. 11).

  • An electronic speed controller can be an electronic circuit which can vary an electric motor's speed, its direction and possibly also to act as a dynamic brake. Electronic speed controllers are often used for brushless motors to provide an electronically generated three-phase electric power low voltage source of energy for the motor. An example of brushless electronic speed controllers is a Jeti Spin Pro Opto brushless esc (Jeti USA, Palm Bay, Fla.).

  • Electronic speed controllers can be programmed to control a speed of the motor. Further, the speed controllers can be configured to detect and store a peak controller temperature, maximum current, minimum current, RPM and motor run time. The electronic speed controllers can be coupled to a wireless transmitter for sending telemetry data and receiving control commands from a wireless controller.

  • In one embodiment, each electronic speed controller can be coupled to a number of Lithium Polymer batteries, such as four to fourteen cells. Dans figure. 9, four battery backs are shown coupled to each controller to provide up to sixty Volts. Electronic speed controllers are normally rated according to maximum current, for example, 25 amperes. Generally, the weight of the electronic speed controller increases as the maximum current capacity increases. Many modern electronic speed controllers can support nickel metal hydride, lithium ion polymer and lithium iron phosphate batteries with a range of input and cut-off voltages.

  • For brushless electric motors, the correct phase varies with the motor rotation, which can be taken into account by the electronic speed controller. Usually, back EMF from the motor is used to detect this rotation, but variations exist that use magnetic (Hall Effect) or optical detectors. Computer-programmable speed controls generally have user-specified options which allow setting low voltage cut-off limits, timing, acceleration, braking and direction of rotation. In some instances, the motor's direction can be reversed by switching any two of the three leads from the electronic speed controller to the motor.

  • In alternate embodiments, other types of electric motors can be used. For example, one or more brushed electric motors can be used. In brushed electric motors, brushes carry power to windings which rotate. DC motors can include shunt motors, separately excited motors, series motors, permanent magnet DC motors and compounded motors. AC motors can include induction motors and synchronous motors. Other electric motor types can include stepper motors, hysteresis motors, reluctance motors and universal motors. As described above, non-electric motors, such as a combustion motor can also be utilized.

  • figure. dix is a top view 330 of the hoverboard. The hover engines are suspended beneath the central support structure 302 as described above with respect to figure. 9. The shrouds, such as 318, of the hover engines extend slightly beyond an edge of the support structure 302. The shrouds can be made strong enough to support a weight of a person without impinging any underlying parts, such as a rotating STARM.

  • A rider platform 332 is mounted above the support structure. The top of the rider platform 332 may substantially flat, i.e., a minimal amount of protuberances. The protuberances may be minimized to allow a rider to move around the rider platform without tripping. Although, as described below, the rider platform may be configured to bend and flex and hence may be curved. In one embodiment, the rider platform may include foot straps for securing a rider's feet in place. In various embodiments, the rider platform 332 and the support structure can be formed as a single integrated unit (e.g., see FIGUES. 22A-22D).

  • Some examples of materials which may be used to form support structure 302, shroud 318 and rider platform 332 include but are not limited to wood, plywood, plastic, reinforced plastic, polymers, glass filled nylon, fiber glass, reinforced composites, metals (e.g., aluminum), metal alloys, metal composite materials (e.g., an aluminum composite material), a hemp composite, composites with a honeycomb core or other inner structure, composites with a balsa core, expanded metal, etc.

  • The pivot arms 310, which are attached to each of the hover engine shrouds, such as 318, are coupled to the rider platform 332 at connection points 334. The rider platform can be formed from a flexible material. When a rider stands on the platform and shifts their weight from quadrant to quadrant, the rider platform can flex. The flex can cause the pivot arm coupled to each of connection points 334 to move downwards which causes the hover engine coupled to each pivot arm to tilt. As described above, when the hover engine is tilted, a force can be generated which is approximately aligned with the tilt axis.

  • The rider can shift their weight and the amount of weight distributed to each pivot arm by changing their foot position on the rider platform 332 and the amount of weight distributed to each foot. Thus, the amount of force distributed to each pivot arm can be controlled and hence the amount of tilt to each hover engine can be varied. By varying the tilt, an amount of translational force output by each hover engine in a particular direction can be controlled. As described above, these forces can be used to control spin, such as starting or stopping a spin and controlling a rate of spin. The forces can also be used to steer the hoverboard.

  • In particular embodiments, different thickness platforms with greater or less stiffness can be used for different weight riders. For example, a stiffer platform may be used for heavier riders and a less stiff platform can be used for lighter riders. The support structure 302 can include a quick release mechanism which allows rider platforms with a different amount stiffness to be easily removed and then replaced. In another embodiment, adjustable stiffeners, such as springs which can be tightened loosened can be provided. The adjustable stiffeners can be used to increase or decrease the relative stiffness of the rider platform 332.

  • In one embodiment, the hoverboard can be configured to receive one or more extension arms. The extension arms can attach to the board on one end and can include an attachment on the other end, such as a skid pad, a wheel, a castor wheel. The one or more extension arms may be used to limit the amount of tilt of the hoverboard and provide additional stability. For example, during training, the extension arms might be used in a similar manner as training wheels are used on a bike,

  • figure. 11 is a side view 350 of a hoverboard. As can be seen in the figure. 11 all of the components need to operate the hover engines, such as the batteries and speed controls are suspended from the bottom of support structure 302 and packaged below a height of the bottom of the hover engine. As described above, the height of the hoverboard from the bottom of the hover engine to the top of the rider platform is about 4.5 inches Thinner designs are possible and this example is provided for the purposes of illustration only.

  • In this embodiment, the rider platform 332 is supported at the ends and coupled to the structure 302 via members 374 une et 374 b. This configuration allows the rider platform 332 to bend in the middle, such as when weight is applied at location 354 et 356 above the pivot arms, such as 310. In an alternate embodiment, the rider platform may be supported by a member, which bisects it lengthwise. Then, the rider platform 332 may be bent on either side of this central member when weight is applied.

  • In yet another embodiment, the rider platform 332 may be sectioned to allow portions to move independently of one another. The individual sections can be coupled to the hoverboard such that they may be flexed to actuate one of the tilt mechanisms. In another embodiment, the individual portions may be coupled to the hoverboard via a hinge mechanism. The individual portions can then be rotated about the hinge.

  • When a hinge mechanism is used, a stiffer material may be utilized for the individual section. However, a repositioning mechanism, such as one or more springs or flexible foam, may be used to return the individual portion to an original position after a force is removed. The repositioning mechanism, such as springs, can also be used to affect the amount of force required to move the individual section.

  • The hover engine shrouds are coupled to a hinge mechanism 372. The hinge mechanism 372 hangs from the support structure 302. The hinge mechanism provides for rotation about one axis. Some examples of hinge mechanisms which may be utilized include but are not limited to a butt hinge, a barrel hinge, a flush hinge, a continuous hinge, a pivot hinge, a coiled spring pin hinge and self-closing hinges. A gap is provided beneath the hinge mechanism, the gap allows wires 308 une from the speed controller 310 to reach the motor encircled by the shroud 318. In alternate embodiments, the shroud can include one or more apertures which allow wires to be passed to the motor.

  • In this example, the hinges allow each hover engine to rotate through some angle, such as 366 et 369, about one rotational axis. As described with respect to figure. 14, joints which allow for more rotational degrees of freedom are possible and this example is provided for the purposes of illustration only. The bottom of the shrouds, such as 318, when tilted is illustrated by the dashed line 362 et 364. The tilt angles 358 et 360 are defined as the angle between the shrouds are horizontal and the bottom of the shrouds when tilted as indicated by lines 362 et 364.

  • In one embodiment, the support structure 302 around the hinge mechanism 372 can be partially recessed. The recess may allow a portion of the hinge mechanism 372 to slide up into the recessed portion when a hover engine is tilted. This feature may allow the overall thickness of the hoverboard to be reduced since the clearance needed to rotate the hover engine may be reduced. In one embodiment, the support structure can be formed from materials, such as wood, metal, fiberglass (or other lightweight but strong composite materials), plastic and combinations thereof.

  • In one embodiment, the hover engines can be configured to tilt up to ten degrees in one direction. In operation, when the weight is removed from locations 354 et 356, the rider platform 332 may unbend and the shrouds may return to a first position. When weight is added, the rider platform may flex by some amount at each location and the shrouds may each tilt by some amount.

  • As described above, the amount of tilt associated with each hover engine may be constrained. Further, the amount of tilt doesn't have to be same for each hover engine. For example, one hover engine can be allowed to rotate up to ten degrees while a second hover engine can be allowed to rotate up to only five degrees. In particular embodiments, a hover engine can be configured to rotate through up to 10 degrees, up to 20 degrees or up to 30 degrees of total rotation. The rotation directions 366 et 368 are shown for each hover engine. In one embodiment, each hover engine is allowed to rotate in only one direction. In another embodiment, a hover engine may be allowed to rotate in two directions, such as angles of plus or minus ten degrees past the horizontal. In this embodiment, a different mechanical linkage would be needed, such as a second pivot arm coupled to the shroud (see e.g., figure. 15).

  • FIGUES. 12 and 13 are a perspective view and side view of a hoverboard 500 with four hover engines 506 coupled to struts 504. A rider platform 502 is coupled to the struts 504 and four pivot arms 508. The rider platform 502 is coupled to support member 516 via connector 514. The support member 518 is coupled to the struts 504 via connector 518.

  • In one embodiment, the rider platform 502 can be flexible, such that when a force is applied near its edge support member 508 is pushed down and the hover engine is also pushed down. A different mechanical linage can be provided such that a downward motion causes the hover engine to be tilted in the opposite direction.

  • FIGUES. 14, 15 et 16 show some embodiments involving manual STARM control. En particulier, FIGUES. 14, 15 et 16 are illustrations of a hover engine coupled, respectively, to a lever arm, foot pedals and reins. These controls might be used in combination with the previously described hoverboards.

  • Dans figure. 14, a lever arm 602 is coupled to a motor/STARM via a ball joint 606. When hovering, a movement of the lever arm 602 from side to side can cause the STARM 610, which includes an arrangement of magnets 612, to tilt relative to a conductive surface such that a vehicle including the hover engine moves forward and backward. The amount of side to side tilt can affect the speed at which a vehicle moves in these directions. A movement from front to back can cause the STARM 610 to tilt such that the vehicle moves either left or right. A combination of a left or right movement and a front or back movement of the lever 602 can tilt the STARM such that the vehicle moves in various directions along different lines. A change in the lever direction as a function of time can change the direction vector of the force which is generated as a function of time and hence the vehicle can move along an approximately curved path.

  • In various embodiments, a mechanical linkage can be used which causes one or more hover engines to be tilted in response to a movement of the lever arm 602. For example, two hover engines can be coupled to a common rotational member such that both hover engines are rotated in response to a torque applied to the rotational member (see e.g., figure. 20). In addition, as described above, digital controls can be used where a movement of the lever arm 602 is detected by one or more sensors. The sensor data can be received in an on-board processor. Based such factors, as an amount movement, a direction of movement and a rate of movement of the lever arm 602 and other factors, such as a current orientation and direction of motion of the vehicle, the on-board processor can generate one or more commands. The commands can be sent to one or more actuators via wired or wireless communications. The actuators can include logic devices (e.g., controllers) which enable communications with the on-board processor and interpreting of commands from on-board processor.

  • The one or more actuators can be coupled to a single hover engine or a plurality of different hover engines. In response to receiving the commands, the actuator controller can cause the actuator of output a force or a torque. The force or torque can cause the hover engine to change its position in some manner, such as but not limited to a tilt position.

  • In some embodiments, the on-board processor can send commands, which cause a rotation rate of a STARM associated with a hover engine to go to a particular RPM value. The motor commands, which can be received by motor 608, can be generated in conjunction with the actuator commands. The RPM value can affect the amount of force which is generated from the hover engine after its position has been changed. The motor 608 can include one or more controllers for 1) communicating with the on-board processor (wired or wirelessly), 2) processing the commands received from the on-board processor and 3) generating commands to control mechanisms associated with the motor which are used to implement the command, such as an amount of power delivered to the motor.

  • figure. 15 shows foot pedals, 652, which can be used to tilt hover engine including a motor 662 and a STARM 664. When one foot pedal, 652, is pressed downwards, the STARM 664 can generate a force, perpendicular to the page, which can cause the vehicle to move forward. When the other foot pedal is pressed downwards, the STARM 664 can generate a force, which can cause the vehicle to move backwards. The amount each pedal is depressed can be used to control a speed of the vehicle in a particular direction. When a first pedal is pressed to move the vehicle in one direction, removing pressure from the first pedal and applying pressure to the second pedal can act as a brake to slow the vehicle.

  • A mechanism is provided with each foot pedal which generates a restoring force. It can also be used to affect how much force needs to be applied to a pedal to move the pedal. Further, the mechanism can limit how far the pedal can move. Dans figure. 15, the mechanism is represented as a spring. The mechanism can generate a force which is approximately linear and/or non-linear with the amount of displacement of the foot pedal. In particular embodiments, one or more mechanisms which generate a restoring force can also be used with the lever arm shown in figure. 14. Again, as described above, one or more foot pedals can be used as part of a digital control system.

  • figure. 16 is an illustration of a hover engine coupled to reins 672 et 674. The reins, 672 et 674, are attached to a housing via connection points 680 et 682. A force, such as 676 et 678, can be applied one at a time or simultaneously to change a tilt angle of the STARM 664. When one or both reins are not in tension, then no force is applied to the hover engine including motor 662 and STARM 664.

  • In one embodiment, a restoring force mechanism can be provided, such as springs 654, which are configured to return the STARM to a neutral position (the neutral position can involve the STARM 664 being horizontal or in a tilted position). To maintain a particular tilt angle, a constant force is applied to one or both reins. When the force is removed, the STARM 664 returns to its neutral position.

  • In another embodiment, the mechanism can be resistant to movement such that the STARM 664 remains in place at a particular orientation after it is moved. For example, the ball joint 656 or other hinge mechanism can be formed with sufficient internal friction to hold the STARM 664 in position. In this embodiment, a first combination of forces can be applied to the reins, 672 et 674, to move the STARM 664 to a first orientation where it remains in place in the absence of an externally applied force. To move the STARM 664 to a second orientation, a second combination of forces can be applied via the reins 672 et 674.

  • FIGUES. 17A to 17E are illustrations of shroud configurations for a hover engine 706. The hover engines each include a motor 702 and a STARM 704 which rotates in direction 712. The motor 702 and STARM 704 are configured to rotate 708 about rotation axis 710.

  • Dans FIGUES. 9, dix et 11, the shroud extended around the sides of the STARM such that the bottom of the shroud was parallel to or extended slightly past the bottom of the STARM. In this embodiment, the bottom of the STARM is exposed. Dans figure. 17A, a shroud 714 extends beneath the STARM 704 and partially covers a bottom of the STARM. The lip portion which extends beneath the STARM can prevent debris from entering the side of the STARM. In addition, the bottom portion can provide a minimum stand-off from a substrate which supports the induced eddy current. During start-up, the minimum stand-off can reduce a maximum amount of drag torque which is generated because the drag torque decreases exponentially with a distance from the substrate (e.g., see figure. 4B).

  • Dans figure. 17B, a bottom portion 718 of the shroud 714 completely encloses a bottom of the STARM 704. In one embodiment, the shroud 714 can include apertures which allow air to enter and exit the shroud 714. The air flow may be used to provide a heat exchange mechanism for the STARM. The STARM may include fins to increase air flow within an enclosure. Further, aperture locations may be selected to increase air flow through the enclosure. The apertures may be covered with a mesh to prevent debris from entering the enclosure.

  • In another embodiment, the motor 702 and STARM 704 may be completely enclosed. The enclosure can be evacuated such that it operates at a pressure lower than the surrounding environment. Thus, the enclosure may include a pressure sensor and a valve for removing air. Alternatively, the enclosure may be filled with a gas different from the surrounding environment. For example, the enclosure may be filled with nitrogen or an inert gas, such as argon or xenon. In this embodiment, the pressure may to which the enclosure is filled may be lower, approximately equal to or above an ambient pressure surrounding the enclosure.

  • In yet another embodiment, an entire vehicle can be enclosed in a shell. For example, a flying saucer shaped vehicle can be built. The motor, batteries, STARMs, control circuitry and other vehicle components can be disposed within the shell. In this configuration, an enclosure and/or shroud may not be needed around the motor and STARM because the motor and STARM are disposed within the shell. Like enclosure in the previous paragraph, the shell can be configured to operate at a pressure which is different than the surround environment and/or can be filled with an atmosphere which is different than the surround environment.

  • Dans FIGUES. 17A and 17B, the motor 702, the STARM 704 and shroud may be tiltable as a unit. Dans figure. 11, the motor, STARM and shroud are also tiltable as a unit. In other embodiments, as described as follows with respect to FIGUES. 17C and 17D, the motor 702 and STARM 704 are tiltable as a unit while the shroud 714 remains in a fixed position.

  • Dans figure. 17C, the motor 702 and STARM 704 are configured to tilt about axis 710 while shroud 714 remains fixed. The bottom of the shroud 714 is open and doesn't include a lip portion as shown in figure. 17A to allow the STARM to tilt past the horizontal. The inner sides of the shroud 714 are curved accommodate the rotation of the STARM.

  • Dans figure. 17D, a bottom portion 720 of the shroud 714 is provided which doesn't rotate with the STARM in direction 712. To allow 720 not to rotate with STARM 704, it can be mounted to a static portion 718 which extends through STARM 704. However, the bottom portion 720 is configured to tilt as the motor 702 et 704 tilt.

  • In one embodiment, a flexible membrane 716 can be provided which couples the bottom portion 720 to the side of the shroud 714. The bottom portion can be formed from a rigid material. The membrane 716 can stretch as the motor 702, STARM 704 and bottom portion 720 of the shroud tilt. The tension in the membrane 716 can be used to restore the STARM and motor to a neutral position after it has been tilted.

  • Dans figure. 17E, a two part shroud is used. A first portion of the shroud 714 extends down the sides of the motor 702 and STARM 704. The first portion 714 is open and doesn't tilt with the motor 702 and STARM 704 or rotate with STARM 704. A second portion 724 of the shroud is mounted to bottom of STARM. The second portion 724 is configured to tilt with the motor 702 and STARM 704 and rotate with the STARM 704.

  • The second portion 724 of the shroud extends around the side of the STARM 704. In one embodiment, the portion which extends around the side of the STARM is selected so that it extends above a bottom of the first portion of the shroud 714 when the STARM is tilted at the maximum angle. Spacers 722 are provided which provide a stand-off distance between the STARM 704 and the second portion of the shroud. In alternate embodiments, the second portion of the shroud 724 can be mounted directly to the STARM 704 without any standoff distance.

  • Dans FIGUES. 17C, 17D and 17E, an actuator can be used to tilt the motor 702 and STARM 704. In one embodiment, the actuator may be located outside the shroud. The shroud can include one or more apertures which allow a mechanical linkage to extend from actuator and into the shroud. The mechanical linkage can be used to transfer a torque from the actuator to rotate the motor and the STARM.

  • For example, a rigid L-shaped lever arm can expend from the shroud where the arm is pushed up or down to rotate the motor and the STARM. The lever arm can be pushed by an actuator or via a force from a person's body, such as their foot. In another embodiment, a straight rotational member can extend from the actuator which is coupled to the motor and the STARM. The motor and the STARM can receive a torque from the rotational member.

  • In yet another embodiment, one or more actuators can be located inside the shroud. An actuator can be coupled to a static portion of the motor and a static portion of the shroud. The actuator can be then configured to generate a force which tilts the STARM and the motor.

  • Next, shroud configurations are described where the shroud is tiltable and also forms a portion of the rider platform. FIGUES. 18A to 18D are illustrations of examples of shroud configurations for a hover engine where a top of the shroud forms a portion of the rider platform. A rider may be able to directly step on a portion of the shroud to tilt the shroud, motor and STARM as a unit which causes a translational force to be generated.

  • Dans figure. 18A, a rider plate platform including three portions 800, 802 et 804 is provided. The second portion 804 is part of a shroud which is coupled to a motor and STARM (not shown). A structural member 810 extends from the first portion 800 and through the shroud 804. The third portion 802 of the rider platform 802 is coupled to the structural member 810. In this embodiment, the first portion 800 and the third portion 802 are configured to move as unit. In alternate embodiments, structure can extend around one or both of sides of second portion 804 such that first portion 800 and the second portion 802 are joined with one another to extend around all or a portion of the circumference of second portion 804.

  • The second portion 804 can be configured to rotate 806 around axis 808 which is through the center of member 810. For example, portion 804 can be formed with a hollow tube through which member 810 extends. In this example, member 810 may be cylindrically shaped. Other types of hinge mechanisms can be provided which allow rotation between the member 810 and the second portion 804. Thus, the tube structure is provided for the purposes of illustration only and is not meant to be limiting. In particular, member 810 doesn't necessarily have to be cylindrically shaped.

  • In one embodiment, in a first position, the top surface of 804 is substantially parallel to the adjacent top surfaces of 800 et 804 across gap 816 such that a continuous surface is formed. In another embodiment, in the first position, the top surface of 804 can be slightly above the adjacent top surfaces of 800 et 802. The discontinuity between the surfaces may allow a rider to detect a position of their foot relative to the top surface 804 and portions 802 et 804.

  • A force can be applied on the top surface of 804 to cause it to tilt about axis 808. The direction of tilt depends on which side of tilt axis 808 a net force is provided as forces from a source, such as a rider's foot, can be distributed across the top surface. Upon tilting, a portion of the top surface of 804 rises above an adjacent top surface of 800 et 802 and a portion of the top surface of 804 sinks below an adjacent top surface of 800 et 802. This tilt can cause a translational force to be generated where the magnitude of the force depends on the magnitude of the tilt angle, the direction of the tilt, as well as the overall orientation of the hoverboard relative to the substrate which supports the eddy currents. The direction of the tilt can affect the direction in which a translation force is generated.

  • A structure can be provided which provides a maximum limit to the tilt angle in either direction. As described above, the maximum tilt angle does not have to be the same in each direction. Further, the tilt potential can be limited such that 804 can tilt in only one direction.

  • The tilt mechanism, such as a hinge, can be configured to resist to tilting such that it can maintain a tilt angle. For example, the tilt mechanism can be designed with a certain amount of friction between parts which may need to be overcome to tilt portion 804 about axis 806. Thus, an external force, such as from a rider, is applied each time the tilt mechanism is moved from one position to another position.

  • In another embodiment, the tilt mechanism can include restoring force mechanisms which provide a restoring force, such as springs, membranes or actuators. After portion 804 is tilted in a first direction to a first position, a restoring force mechanism can be configured to generate a moment which acts to tilt the portion 804 opposite the first direction. The restoring force mechanisms can be configured to generate a restoring force which helps to move surface 804 in a particular direction along or conjunction with an externally provided force.

  • As an example, a user can press on portion 804 with their foot to tilt portion 804 from a first position to a second position. When the user removes their foot, the restoring force mechanism can be configured to tilt portion 804 from the second position to the first position. As another example, a user can press on portion 804 with their foot to tilt portion 804 from the first position to the second position. When the user removes their foot, the portion 804 may maintain the first position. However, if the user applies a small amount of force in conjunction with the force output from the restoring force mechanism, the portion 804 is tilted in the opposite direction.

  • In one embodiment, a restoring force mechanism can be an actuator which is powered from a source other than a user. In another embodiment, the restoring force mechanism can be a device, such as but not limited to springs or flexible membranes, which stores energy applied from the user. The stored energy can be released to supply the restoring force.

  • In another embodiment, a flap can be utilized. The flap can be formed a flexible but rigid material to allow it to be anchored on one side and then bent. The flap can be designed to return to its unbent shape which generates a restoring force.

  • As an example, flexible membranes 812 et 814 can be attached to surfaces 800 et 802. The membranes are shown over the top of portion 804. In other embodiments, membranes can be located on the sides of portion 804 or run beneath portion 804. The attachment points for the membranes can also be on the sides of portion 800 et 802 or on a bottom surface of 800 et 802.

  • In one embodiment, membranes 812 et 814 may be only attached to 800 et 802 such that when portion 804 is tilted in a particular direction only one of membranes 812 et 814 is stretched. The membrane which is not stretched doesn't generate a restoring force. In another embodiment, one or both of membranes 812 et 814 are attached, such as glued or fastened, to portion 804. For the membranes attached to portion 804, the membranes are stretched when the top portion is rotated past the horizontal in either direction.

  • In various embodiments, to accommodate rider preferences, the width or thickness of membranes 812 et 814 can be adjusted to increase the amount of force required to stretch. Further, membranes don't have to be of equal thickness or width (e.g., see figure. 21). Thus, more energy may be required to tilt 804 in one direction versus the opposite direction.

  • In another embodiment, a single membrane can be used. When an equal amount of membrane is placed on either side of axis 808, the amount of energy required to tilt 804 from the horizontal in either direction can be equal. When more membrane is placed on one side of axis 808 than the other, more energy can be required to tilt 804 in one direction versus the opposite direction. The membranes don't have to stretch across the entire top of 804 de 800 à 802. For example, membranes can be attached which span gap 816 and are anchored to portions 800 et 804 and/or are anchored to portions 802 et 804.

  • The top surfaces of 800, 802 and/or 804 can be recessed where a membrane is attached and/or spans the portion. The recess can allow the membrane to be attached in a manner where it is flush with the adjacent portions. Thus, the depth of the recess can depend on the thickness of the membrane and bonding agent which is used.

  • Dans figure. 18B, a rider platform is formed from the top surfaces of structure 820, member 824 and hover engine 822. The member 824 is anchored and can be integrated with board 820. Hover engine 822 is coupled to member 824 with a gap 812 between board 820 and hover engine 822. The gap 812 enables the board 820 and the hover engine rotate relative to one another.

  • In one embodiment, member 824 can be formed from two pieces which can rotate relative to one another, such as two tubular pieces. A housing or shroud associated with hover engine 822 can be coupled to one of the pieces. One or more springs, such as a spring disposed in the interior of member 824, can be used as a restoring force mechanism.

  • Dans FIGUES. 18A and 18B, a single hover engine is shown. In other embodiments, the top surfaces from a plurality of hover engines can form a portion of a rider platform (see, e.g. figure. 20). For example, in figure. 18B, hover engine 822 may be reduced in diameter relative to the size of the board 820 and moved closer to side 826 of board 820 to allow two hover engines to be placed side by side. In this example, the tilt axes of each hover engine associated with the tilt mechanisms can be parallel to one another. In another embodiment, each of the two hover engines can be coupled to board 820 such that the tilt axes are non-parallel to one another. In this example, the two hover engines can be configured to have the same or a different range of tilt angle.

  • Dans FIGUES. 18C and 18D, boards 830 et 850 surround hover engines 832 et 852, respectivement. Dans figure. 18C, a rider platform is formed from a top surface of the hover engine 832 and a top surface of board 830. Dans figure. 18D, a rider platform is formed from a top surface of the hover engine 852 and the top surface of board 850.

  • Dans figure. 18C, two tilt mechanisms, such as 838, are coupled to a shroud associated with hover engine. The tilt mechanisms allow the shroud to rotate 836 about axis 834. The gap 840 between the board 830 and the hover engine 832 allows the hover engine 832 to tilt relative to the board 830.

  • The board 830 is provided with slots, such 842. The rotation mechanism is coupled to the slot to allow the hover engine 832 to be moved in accordance with the path of the slot 842, which is curved in this example. A shift in position of the hover engine 832 along the slots, such as 842, changes the orientation of axis 834 relative to the board 830. A change in orientation of axis 834 relative to the board changes a direction of a translation force which is output from the hover engine 832 when its STARM is angled relative to a surface which supports the eddy currents. The direction of the translational force output from the STARM can affect a magnitude of a rotational moment which generated about a center of mass of the STARM.

  • In one embodiment, a fastener can be provided which fixes the position of the tilt mechanisms, such as 838, in the slots, such as 842. The position can be selected to suit a rider preference. During operation, the fastener is configured to hold the STARM in the selected position. I

  • In another embodiment, the tilt mechanism 838 can be coupled to the board 830 in a manner such that its position along the slot 842 can be adjusted during operation of the hoverboard. This feature allows the hover engine to be tilted and a direction of the associated tilt axis to be adjusted in flight. In various embodiments, the force to change a position of the tilt mechanism in the slots, such as 842, can be provided from an actuator coupled to the hoverboard or from a force provided by the rider, such as from a rider's foot.

  • Dans figure. 18D, two hover engines, 852 et 864, are coupled to board 850. The, first hover engine 852 is coupled to board 850 via tilt mechanisms 862. A thin piece 860 of board 850 extends around the hover engine 852. A gap 858 between the board 850 and the hover engine allows the hover engine 852 to rotate 856 about axis 854.

  • In one embodiment, the tilt mechanisms 862 can be coupled to a track, such as a slot in board 850 (not visible) formed below the top surface. The length of track can extend around all or a portion of the circumference of the hover engine. The tilt mechanisms can be rotated along the track to change an orientation of the tilt axis 854 relative to the board 850. In one embodiment, a top of the track can be substantially parallel to a top surface of board 850. In another embodiment, the track can be angled relative to a top surface of board 850 such that when the top surface of board 850 is parallel to the ground a STARM associated with the hover engine 852 is angled relative to the ground.

  • The second hover engine 864 is coupled to board 850 via tilt mechanism 866. A gap between the board 850 and the second hover engine 864 allows the hover engine to rotate around axis 868. In this example, the orientation of the tilt axis 868 is fixed relative to board 850.

  • Dans FIGUES. 18A to 18D, when a top surface of the hover engine is parallel to the top surfaces of the adjacent structures which form the rider platform, a bottom of the STARM associated with the hover engine may be also parallel to the top surface of the hover engine. In this position, if the STARM is also parallel to the surface which supports the induced eddy currents, the translational forces output from the hover engine approach zero. In another embodiment, when a top surface of the hover engine is parallel to the top surfaces of the adjacent structures which form the rider platform, a bottom of the STARM associated with the hover engine may be angled to the top surface of the hover engine. In this position, if the top surface of the hover engine and rider platform is parallel to the surface which supports the induced eddy currents and the STARM is angled to the surface which supports the induced eddy currents, some amount of translation force can be output from the hover engine.

  • FIGUES. 19A to 19C are illustrations of side and top views of tiltable hover engines. Dans figure. 19A, a hover engine 870 is shown. The hover engine includes a shroud 872 and STARM 874 underneath the shroud. A tilt mechanism 876 extends through the shroud 872, such that a middle portion of the tilt mechanism is enclosed by the shroud. A motor may or may not be enclosed in the shroud. For example, the shroud can include an aperture which allows a belt or some other mechanism which provides a torque to turn the STARM to extend through the shroud.

  • Dans figure. 19B, a hover engine 880 is shown. The engine 880 includes a shroud 882 and STARM 874. In this example, tilt mechanisms 884 are coupled to the side of shroud 882 and extend from the shroud 884. In various embodiments, the shroud, such as 882, can be formed with a slot for receiving a tilt mechanism, such as 884.

  • Dans figure. 19C, a hover engine 890 is shown. The hover engine 890 includes a shroud 892 enclosing a STARM 874. In this embodiment, the shroud includes a recess configured to receive a tilt mechanism 894. The shroud 892 is coupled in some manner to the tilt mechanism 894.

  • figure. 20 is an illustration of a hoverboard with four hover engines 902, 904, 906 et 908. A payload platform (or rider platform) is formed from the top surfaces of hover engines 902, 904, 906 et 908 and board 900. Gaps are provided between the board 900 and the hover engines to allow the hover engines to rotate relative to the board 900. The hover engines are shown to be about the same size. In one embodiment, the four hover engines can be identical to one another except for the manner in which they are coupled to the hoverboard.

  • In this example, the two hover engines are coupled to a single tilt mechanism 910. The single tilt mechanism is coupled to board 900 and is configured such that hover engines 902 et 904 tilt in unison. The hover engines are shown to both rotate around a single tilt axis. In alternate embodiments, the ends of rotational member 910 can be angled to one another such that hover engines, 902 et 904, rotate about different tilt axes.

  • The hover engines 906 et 908 are coupled to a tilt mechanism 912 et 914 respectivement. These hover engines may be tilted independently of one another and hover engines 902 et 904 respectivement. The hover engines 902, 904, 906 et 908 can be configured to be tilted in response to an external force provided by a person and/or an actuator. For example, the hover engines can all be tilted in response to a force supplied by the rider. As another example, hover engines 902 et 904 can be tilted in response to a force supplied by a rider whereas hover engines 906 et 908 can be tilted in response to a force provide by an actuator. In yet another example, all the hover engines can be tilted in response to an actuator supplied force.

  • figure. 21 is an illustration of a hoverboard with two tiltable hover engines 922 et 924. In this example hover engine 922 is coupled to board 920 via rotational mechanisms 925 such that it can rotate through the board and a top surface of the hover engine forms a top surface of board 920. Hover engine 924 is coupled to rotation mechanism 926 on a bottom surface of board 920. Thus, it doesn't rotate through board or form a portion of the rider platform.

  • In this example, two flexible membranes 928 et 930 are provided to generate a restoring the force. The membranes are of different widths, equal thicknesses and formed from the same material. Thus, more energy is required to tilt hover engine in one direction versus another direction. In other embodiments, the membranes can be formed from different materials and/or have different thicknesses.

  • The hover engines 922 et 924 are of different sizes. In one embodiment, hover engine 922 can be configured to generate more lift than hover engine 924, such as via a greater volume of magnets located on its STARM. In one embodiment, the hoverboard may be designed to hover using only the lift generated from hover engine 922. Hover engine 924 may configured to provide primarily control forces such as to steer the hoverboard left or right and keep the front portion of the hoverboard from hitting the ground.

  • FIGUES. 22A to 22D et 23 are illustrations of a hoverboard 1000 with two hover engines. The rider platform 1002 is formed from wood. The rider platform 1002 includes end pieces, 1004 et 1006, which are angled upwards. Between the end pieces, the rider platform is substantially flat. When a user applies pressure to the end pieces a moment can be generated which causes one end of the board to rise relative to the other end of the board. Since the forces output by each hover engine are a function of their distance from the surface, raising one end can change the magnitude of forces associated with each hover engine and a distribution of forces across the hoverboard.

  • The two hover engines are each enclosed in a shroud 1008 et 1010. The shrouds 1008 et 1010 extend beyond the width of the rider platform 1002. The shrouds can be configured to support a rider's weight without impinging on a STARM rotating within the shroud. In one embodiment, the maximum diameter of the shrouds is about twelve inches.

  • The hover engines are coupled to the rider platform via interface with fasteners, such as 1014. A light, such as 1016, is provided at the bottom of each hover engine. An enclosure 1012 is suspended beneath the rider platform and between the hover engines. The enclosure can include such components as but not limited to a power source (e.g., battery backs), engine speed controllers, sensors, wireless communication devices, an on-board processor, wiring bundles, etc.

  • figure. 23 is a cross section of the hoverboard 1000. The length from end piece to end piece is about thirty one inches. The thickness 1026 of the rider platform is about one half inch. The maximum width 1030 of half of the rider platform 1002 is about seven and one half inches. The height 1028 of the hover board 1000 from the bottom of enclosure 1024 to the top of the rider platform is about 3 inches.

  • Each hover engine includes a motor 1022 and STARM 1020. In this example, the hover engines are identical. In one embodiment, the engine speed controller (not shown) can be integrated into the hover engine. The shroud 1008 can include apertures to allow wire bundles to pass from enclosure 1024 into the interior of shroud 1008. In one embodiment, the bottom of the rider platform 1002 can include grooves for routing wire bundles from enclosure 1024 into the shroud 1008.

  • The maximum diameter 1032 of the shroud is about 12 inches. The diameter of the STARM is about eleven inches. The STARM is configured to accommodate magnets with a height from 1 to 1.5 inches. In one embodiment, the magnets can be cubic shaped.

  • figure. 24 is an illustration of a hoverboard 1100 with eight hover engines, 1104 une, 1104 b, 1104 c, 1104 , 1106 une, 1106 b, 1106 c et 1106 . The eight hover engines are each coupled to a structure 1102. A potential direction of movement of the hoverboard 1100 is shown via arrow 1108. Various force distributions on structure 1102 which can lead to a particular direction of movement, such as 1108, are described in more detail with respect to FIGUES. 25A to 25E. In particular embodiments, structure 1102 can be formed with flexible portions to allow bending and/or twisting. The bending and/or twisting can be used as part of a control strategy associated with the hoverboard 1100.

  • A direction of rotation, an axis of rotation and a direction of a translational force which is output from the hover engine when structure 1102 is proximately parallel to a substrate which supports eddy currents induced from the hover engines is shown for each hover engine. For example, hover engine 1104 c is configured to rotate in direction 1110 about axis 1116 and generate a translation force in direction 1122. As another example, hover engine 1106 is configured to rotate in direction 1118 about axis 1124 and generate a translation force in direction 1120.

  • The hover engines 1104 une, 1104 b, 1104 c et 1104 can be coupled to platform 1102 at an angled orientation which is fixed. A single angle of rotation is shown for each of the hover engines. For example, axis 1112 can be the axis of rotation of the motor and STARM for hover engine 1104 c and axis 1116 can be perpendicular to a top surface of 1102. Angle 1114 can be ninety degrees plus the angle between axis 1112 and axis 1116. In general, the orientation of axis 1112 relative to an axis perpendicular to surface 1102 can be represented as rotation amounts about two axes, such as 1115 et 1117.

  • In this example, the axes of rotation of hover engines 1106 une, 1106 b, 1106 c, 1106 are shown as parallel to surface 1102. In alternate embodiments, the axes of rotation of these hover engines can also be angled relative to a top of surface 1102. The orientation of these axes of rotation relative to surface 1102 can be also represented as rotation amounts about one or two axes.

  • In particular embodiments, in hover engines, 1104 une, 1104 b, 1104 c et 1104 , the North-South poles associated with the STARMs can be substantially parallel to the axis of rotation, such as 1112. Some examples of magnet arrangements were described above with respect to FIGUES. 2 and 3. In hover engines 1106 une, 1106 b, 1106 c et 1106 , the North-South poles associated with the STARMs can be approximately perpendicular to the axes of rotation. In this configuration, hover engines 1106 une, 1106 b, 1106 c et 1106 may generate some lift. However, in some embodiments, the hover engines 1104 une, 1104 b, 1104 c et 1104 may generate a majority of the lift and hover engines 1106 une, 1106 b, 1106 c et 1106 may be used primarily to output translational forces.

  • In various embodiments, in hover engines 1106 une, 1106 b, 1106 c et 1106 , the magnets on the STARMs can be arranged such that all the poles are in the same direction, such as all north poles facing outwards. In another example, the poles can alternate, such as alternating north and south poles. In yet another example, magnet polarity arrangements such as those described with respect to FIGUES. 2 and 3 can be used except with the poles rotated some amount, such as ninety degrees.

  • Dans figure. 24, the four pairs of hover engines: i) 1106 une et 1106 b, ii) 1106 c et 1106 , iii) 1104 une et 1104 b and iv) 1104 c et 1104 c are each configured to output a translation force which oppose one another. The hoverboard 11100 can be configured such that at some orientation all the forces and moments generated from the eight hover engines balance one another to allow the hoverboard to hover in place without rotating. For example, this condition may occur when surface 1102 is parallel to the surface which supports the induced eddy currents. When the board 1100 is in other orientations, a particular direction of movement can be generated. Some examples of these orientations are described in more detail with respect to FIGUES. 25A to 25E.

  • FIGUES. 25A to 25E are illustrations of movement directions in response to a force inputs at different locations on hoverboard 1100. Dans FIGUES. 25A to 25E, the eight hover engines and the direction of their output force previously described with respect to figure. 24 are shown. In addition, a location where a force can be applied and a resulting direction of movement of the hoverboard are discussed.

  • In one embodiment, the two hover engines, such as 1106 une, 1106 et 1106 b et 1106 c can be replaced and a single cylinder with a single motor, like a rolling pin, can be used. Magnets can be arranged on the face of the cylinder in rows and columns which span all or a portion of the length of the cylinder and all or a portion the circumference around the cylinder. When arranged in this manner, the diameter of the cylinder may be smaller than the diameter of the hover engines, 1104 une, 1104 b, 1104 c et 1104 . The rotating portion of the cylinder can be covered in a shroud to prevent the rider from stepping on it. The shroud may form a side rail on the board.

  • Dans figure. 25A, a force can be applied within area 1122, which is indicated by the rectangle. For example, the force may be applied by a rider placing their foot in the area 1122 and then pushing down and shifting their weight towards it. This motion can cause a fore of the hoverboard to rise relative to a substrate which supports the eddy current induced by the hover engines. When the fore end of the hoverboard 1100 rises, due to the increased distance of the fore hover engines from the substrate, the translational forces output from the fore hover engines decrease and the translation forces from the aft hover engines increase. Thus, a net translation force 1120 can be generated which moves the hoverboard in the aft direction.

  • Dans figure. 25B, a force can be applied within area 1124, which is indicated by the rectangle. For example, the force may be applied by a rider placing their foot in the area 1124 and then pushing down and shifting their weight. This motion can cause an aft end of the hoverboard to rise relative to a substrate which supports the eddy current induced by the hover engines. When the fore end of the hoverboard 1100 sinks and the aft end rises, due to the increased distance of the aft hover engines from the substrate, the translational forces output from the aft hover engines decrease and the translation forces from the fore hover engines increase. Thus, a net translation force 1126 can be generated which moves the hoverboard in the fore direction.

  • Dans figure. 25C, a force can be applied within area 1128, which is near the center of mass of the hoverboard. For example, the force may be applied by a rider placing their foot in the area 1128 and then pushing down and shifting their weight. This motion can cause the port side of the hoverboard to rise relative to a substrate which supports the eddy current induced by the hover engines. When the starboard side of the hoverboard 1100 sinks and the port side rises, due to the increased distance of the port hover engines from the substrate, the translational forces output from the port hover engines decrease and the translation forces output from the starboard hover engines increase. Thus, a net translation force 1130 can be generated which moves the hoverboard in the starboard direction.

  • Dans figure. 25D, a force can be applied within area 1132, which is near the center of mass of the hoverboard. For example, the force may be applied by a rider placing their foot in the area 1132 and then pushing down and shifting their weight. This motion can cause the starboard side of the hoverboard to rise relative to a substrate which supports the eddy current induced by the hover engines. When the port side of the hoverboard 1100 sinks and the starboard side rises, due to the decreased distance of the port hover engines from the substrate, the translational forces output from the port hover engines increase and the translation forces output from the starboard hover engines decrease. Thus, a net translation force 1134 can be generated which moves the hoverboard in the port direction.

  • Dans figure. 25E, a force is applied in area 1138, which may be some distance from the center of mass of the hoverboard and rider. The port and forward end of the hoverboard can rise. This orientation can cause at least a net force 1140 to be generated. The net force can produce a moment about the center of mass of the hoverboard and rider which causes the hoverboard to rotate. The area where the force is applied can be moved to the port side to generate a moment in the opposite direction and cause the board to rotate in the opposite direction.

  • In the examples above, the hoverboard was described as moving forward, backwards, side-to-side or rotating. These motions can be combined to generate more complex movements. For example, the hoverboard can be actuated to both turn and rotate at the same time, like a ski or a snowboard carving a turn. Thus, the examples provided above are for the purposes of illustration and are not meant to be limiting.

  • In another embodiment, the force directions which are output from each of the STARMs can be reversed. In this embodiment, the board can respond oppositely to the manner described above with respect to FIGUES. 25A à 25E. For example, if the rider presses on fore end, the board will move in the aft direction. As another example, if the rider presses on the aft end the board will move in the forward direction. Combinations between these two examples are also possible. For example, the board can be configured to move in the port direction and the starboard direction when the board is pressed on the port side and the starboard side but move in aft direction when the board is pressed on the fore side and the fore direction when the board is pressed on the aft side.

  • Embodiments of the present invention further relate to computer readable media that include executable program instructions for controlling a magnetic lift system. The media and program instructions may be those specially designed and constructed for the purposes of the present invention, or any kind well known and available to those having skill in the computer software arts. When executed by a processor, these program instructions are suitable to implement any of the methods and techniques, and components thereof, described above. Examples of computer-readable media include, but are not limited to, magnetic media such as hard disks, semiconductor memory, optical media such as CD-ROM disks; magneto-optical media such as optical disks; and hardware devices that are specially configured to store program instructions, such as read-only memory devices (ROM), flash memory devices, EEPROMs, EPROMs, etc. and random access memory (RAM). Examples of program instructions include both machine code, such as produced by a compiler, and files containing higher-level code that may be executed by the computer using an interpreter.

  • The foregoing description, for purposes of explanation, used specific nomenclature to provide a thorough understanding of the invention. However, it will be apparent to one skilled in the art that the specific details are not required in order to practice the invention. Thus, the foregoing descriptions of specific embodiments of the present invention are presented for purposes of illustration and description. They are not intended to be exhaustive or to limit the invention to the precise forms disclosed. It will be apparent to one of ordinary skill in the art that many modifications and variations are possible in view of the above teachings.

  • While the embodiments have been described in terms of several particular embodiments, there are alterations, permutations, and equivalents, which fall within the scope of these general concepts. It should also be noted that there are many alternative ways of implementing the methods and apparatuses of the present embodiments. It is therefore intended that the following appended claims be interpreted as including all such alterations, permutations, and equivalents as fall within the true spirit and scope of the described embodiments.

  • US20150175031A1 – Hoverboard – Google Brevets
    4.9 (98%) 32 votes
     

    Laisser un commentaire