loading
2Instructables48,259Views15CommentsZurich, SwitzerlandJoined June 8th, 2014
I was born in Zurich, Switzerland, grew up nearby. After finishing my studies in Electrical Engineering at the ETH in Zurich with the masters degree, I was working as a software developer - initially in semiconductor industries, and later many years in life science. Designing and creating innovative software, interacting with customers, and managing a small team of software developers was my big pleasure. As a hobby I always designed and experimented with model airplanes. When 3d-printing bec... Read More »

Achievements

10K+ Views Earned a bronze medal
Contest Winner First Prize in the Design Now: In Motion Contest
Show 1 More »
  • Create a Parametric 3d-printable Slew Bearing With Fusion 360

    I've done that. Friction doesn't feel much different. Main advantage is the better control over the bearing diameter and the gaps between the rollers. https://www.thingiverse.com/thing:2381833

    View Instructable »
    • Create a Parametric 3d-printable Slew Bearing With Fusion 360
      3,996 views
      84 favorites
      3 comments
  • Create a Parametric 3d-printable Slew Bearing With Fusion 360

    I had a robot in mind, or maybe a turn-table ...

    View Instructable »
  • 600 Watt, 3d-printed, Halbach Array, Brushless DC Electric Motor

    No. I think the Halbach arrangement partially replaces the flux-ring of the rotor. The weaker side has a less strong and a less extended magnetic field. The rotor has some space to embed continous fibers in order to make it printable with a Mark 2 printer from Markforged (not yet tested).

    I've contacted makeSEA. They are working on it.

    View Instructable »
    • 600 Watt, 3d-printed, Halbach Array, Brushless DC Electric Motor
      33,601 views
      683 favorites
      51 comments
  • 600 Watt, 3d-printed, Halbach Array, Brushless DC Electric Motor

    Axial flux is an interesting arragmenent. I'm a bit scared to get the centrifugal forces under control. Maybe it would be worth to just do an experiment, and get familiar with the potential issues ... many thanks for you great ideas.

    Yes, that works. For higher voltage at low rotation speed, only the number of copper wire turns needs to be increased. 10 times higher voltage should be simple. However, it won't be possible to get 600W at a lower rotation speed.

    View Instructable »
  • 600 Watt, 3d-printed, Halbach Array, Brushless DC Electric Motor

    The ratio of power and weight is really inferior compared to a traditional motor. Efficiency isn't as bad - a traditional motor with the same power at 90% efficiency is already a top-quality motor. Costs are difficult to compare, since quite some manual effort needs to be invested. I think the most interesting aspect is, that it's possible to fabricate a good-quality motor from scratch at home. It's not necessary to have a lathe or a cnc-router in your workshop - just a 3d-printer and a drill-press.

    Your hints for improving the performance are absolutely correct.My motivation to build this motor was driven by the 3d-printing technology: how far can I get, if I'm creating a proper design? And where is the limit when omitting the "traditional" laminations?Consider the two major factors of losses relevant for efficiency: 1) Copper losses are proportional with the square of the current. 2) Iron losses are caused by eddy-currents, which are (linearly?) related with the rotation speed.In order to avoid the copper losses, a superconductor would be required. That's difficult to make. For eliminating the eddy-currents it's only needed to leave away the laminations.Note that this motor has zero iron losses, because there is basically no iron. The magnetic PLA which I'm using for th...see more »Your hints for improving the performance are absolutely correct.My motivation to build this motor was driven by the 3d-printing technology: how far can I get, if I'm creating a proper design? And where is the limit when omitting the "traditional" laminations?Consider the two major factors of losses relevant for efficiency: 1) Copper losses are proportional with the square of the current. 2) Iron losses are caused by eddy-currents, which are (linearly?) related with the rotation speed.In order to avoid the copper losses, a superconductor would be required. That's difficult to make. For eliminating the eddy-currents it's only needed to leave away the laminations.Note that this motor has zero iron losses, because there is basically no iron. The magnetic PLA which I'm using for the stator behaves more like a ferrite. It's iron powder in plastique, not conducting. There are only copper losses.Increasing the voltage (and hence RPM), while keeping the current fixed, will result in in more power and even more efficiency.Example:- Nominal values: 30V, 20A, 600W input, 120W copper losses, 0W iron losses => 80% efficiency- Increased values: 60V, 20A, 1200W input, 120W copper losses, 0W iron losses => 90% efficiencyUnfortunately it's not possible to increase the voltage for this motor. Like you're saying, the fast spinning rotor is the limitation.But there is the Mark 2 printer (from Markforged), which can 3d-print continuous carbon fibres, and the material is 10-20 times stronger than the material I've used. With 100V and proportional higher RPM (hence better cooling) a current of 30A could be possible. Resulting theoretically in 3kW with 90% efficiency.Of course there are other factors (e.g. electronics: fast switching FETs), but these are my thoughts to increase the performance ...

    View Instructable »
  • 600 Watt, 3d-printed, Halbach Array, Brushless DC Electric Motor

    For conventional machining I'd probably change the design: in particular the rotor would be too heavy (and unnecessary strong) it it's all in metal. But I guess a full-metal-version with the same arrangement of magnets and copper-wires could perform at 5kW.

    I'm selling the STL-files for $10. If you have a 3d-printer and some spools of filament available, the used material will cost approx. $15. Any ESC will probably work, as long as it supports 30V and 20A. Recommending a specific ESC is difficult, because it depends on the application. E.g. an ESC for a car has a reverse-function, while this is not useful for a plane.

    This video shows the motor with 600W input power (30V, 20A): With 80% efficiency there is 480W pure shaft power. Torque is 0.7Nm and speed is 6650rpm. The ventilation is good enough to dissipate the lost 120W - the motor only gets a bit warm (cool enough to not have a problem with the plastic). I think this motor can perform up to 1kW (40V, 25A, 8500rpm). The rotor will probably be able to sustain the centrifugal forces. However it needs to be dynamically balanced in order to avoid big vibrations.

    You're right, a gearing is needed to get the 600W from this machine. If only a higher voltage is needed at a very low rpm, it is possible to use thinner wires with more turns per coil. I've done this for an earlier project: https://www.makesea.com/windpowerWriter

    View Instructable »