Electric Moped Update #4: The Death of a Beautiful Thing

The Straw That Broke the Camels Back

I killed her, the moped’s dead… after making the big investment and buying a new power system I ran her a little too hard and the motor went up in smoke. So a little explanation, I’d been riding the moped with the new power upgrade for about a week. The power increase alone was amazing and it wouldn’t even blink when I pointed it up the steepest hills. I could run the new motor all over campus without the motor every getting warmer than a Canadian summer. Alright, you may be asking “how the hell did you manage to kill it then?”.  Well here’s the thing, the motor controller I was running is only rated for 80 amps (far below what the motor is rated for) so I wasn’t worried about pushing the system too hard because I figured the motor controller would be the first to go and I’d just drop the new bike speed controller in to replace it. Well, it turns out I drastically underestimated what the little motor controller could handle (when I figured it would be the weak link). I took the moped out with my roommate Colin to tour an apartment that we’re looking at renting this summer. Of course, both of us were squished onto the little bike and from the prior week’s stress testing, it had shown it can handle the load. The little guy handled the trip like a champ easily taking us both halfway across town and up way too many hills. It didn’t even blink when we got lost (twice). I checked the motor when we got to the apartment and it was barely warm, well below the 80c operating temperature. What killed it was the ride back to the dorm. We decided to take a more direct route back home, one that included a massive hill…

When I saw the hill I was worried right away, but going back to my earlier assumption of the motor controller I figured the worst that could happen is a ruined controller that I’ll be replacing anyway. Well, the moped put up a good fight, it hauled both colin and me all the way up the huge half-mile hill screaming bloody murder all the way. At the very crest, it met its end having sacrificed itself to get us up that godforsaken incline. At the time I thought it was just the motor controller that had gone out. It wasn’t until several days later when I sat down to install the new bike speed controller that I realized the motor was wrecked.

Motor Heating/Cooling

Now the biggest enemy to electric motors is heat. When an electric motor gets too hot the rare earth magnets begin to degrade and eventually the insulation coating the motor windings will melt causing an electrical short and a ruined motor. The heat generated by a motor is directly related to the amps flowing through its coils.

Joule’s first law, also known as the Joule–Lenz law,[1] states that the power of heating generated by an electrical conductor is proportional to the product of its resistance and the square of the current. (Wikipedia)

This means that the amperage you let flow through an ohmic system (any motor), causes an exponential increase in heat generation. This is why most electric vehicles run high voltage because you can get away with a high wattage system that draws fewer amps and thus produces drastically less heat. The power of a motor can be increased by increasing the motors’ cooling. Yes, this decreases the efficiency of the motor as more energy is being converted to heat than to motion (an optimized system that produces little heat in the first place is ideal) but who doesn’t want more power out of a smaller cheaper motor? (*Cough Cough* every Honda civic owner ever). Now the hard part is how to get the heat out of the motor, this part’s a little difficult. The motor I’m using is what’s called an outrunner motor.

 

What that means is the stator (the copper wirey bit that electricity flows through) is surrounded by the rotor (the magnetic bit that spins).  This produces some pretty interesting heat issues. All the coils are on the inside and this is where the heat will be generated. The rotor (the outer bit) doesn’t actually come in contact with the stator so heat can’t be transferred through conduction (the most efficient way to transport heat) to the rotor which has a larger surface area and can dissipate heat to the outside air easily. Because of this, the heat is mostly trapped inside the motor. The only heat that escapes from the stator is from the convection of air through the motor or through heat radiated to the rotor. Now there are two main ways to increase heat dissipation from the stator; increase the airflow through the motor and thermally connect the stator to the rotor. The second option helps the most because thermal energy is transferred more readily through conduction than any other form of heat transfer, a heatsink can then be attached to the rotor drastically increasing surface area and thus heat dissipation (Fourier’s Law explains why this works so well). This thermal connection can be accomplished by sealing the motor and filling it with oil (its non-conductive, non-corrosive, and can act as a lubricant). Heat from the stator is transferred to the oil and the oil transfers the heat to the rotor. Now as ideal as this is it’s hard to seal a motor that’s been designed to be air-cooled and is covered in holes like the RC plane motor I’m using.

Alright, so increasing airflow it is then. These RC airplane motors are designed to be strapped to the front of a plane rocketing through the air with a big ol’ prop blowing back on them, not fixed inside a slow-moving moped. In this case, to increase airflow I could add intakes that channel air from the outside and forces it across the motor like the ones you see on supercars.

Or even simpler, could attach an impeller to the end of the motor that would suck air through the center. An impeller works by using centrifugal force to throw air out of the motor. This air creates a vacuum that sucks in more air producing airflow through the motor.

Now here’s the thing. I didn’t have any extra cooling on the moped motor. All of my experience with these Turnigy SK3 motors is from the one I have on the electric bike which has both an impeller and is exposed to elements getting plenty of airflows. I realize now that my concept of how much heat these motors can take is skewed because I didn’t factor in this extra cooling that the e-bike motor has. My misunderstanding of the robustness of the 80 amp airplane motor controller and of the ability of my motor to dissipate heat are the two largest factors that aided in the untimely death of this poor motor. May she rest in pieces…

Autopsy

You thought I was going to ruin a motor and not take her apart? HA! After a striped sprocket grub screw, several glued in and striped Philips head screws (why not hex?!), a little heat gun work, and a whole lot of finesse I finally got the rotor off. Upon close inspection of the coils, I noticed visible bubbling in the wire enamel coating and a complete melt through in one of the wire bundles that is most likely what killed it. This melt through shorted the wires in the bundle making the motor unusable. The bundle that brunt through is suspended in air and does not come in thermal contact with anything. This lack 0f contact is most likely what caused the group to overheat. Because the group is suspended it can’t conduct heat anywhere easily so it became the thermal weak point of the motor. All other wires look good most likely because they were able to dissipate heat into the steel core.

Did I mention the motor heat completely warped and broke the 3D printed motor mount? Alright yeah, I printed it in PLA but still! Probably going to have to weld a new one to handle the heat.

Can We Re-build Her?

That would be a no… at first I thought I could simply pull out the wire and rewrap the stator but after a couple of hours of struggling to pull/cut out the wire and far too many little wire cuts to count I gave in to the fact that I can’t revive the poor motor.

Summary

So she’s dead for now. I lost quite of bit of money that I can’t afford to lose in this “learning experience”. It’s going to be a while till I can afford to order more parts to get the moped up and running again. She was awesome, absolutely awesome for getting around campus. With how small and nimble she was yet still having plenty of power to get 2 people mostly up hills. After this failure, I am beginning to doubt my knowledge and skills. I’ve been working on the electric bike for almost a year now and it’s still not running consistently (sitting in the shop now with blown MOSFETs). Now I can’t even get this project working that was going to be pretty straightforward. It’s like I just can’t do anything right. Either that or the universe REALLY doesn’t want me to have an electric vehicle. I started writing this post out of sheer anguish and depression because I thought maybe I could prove that it didn’t break just because I made it. Definitely feeling a little better and optimistic now that I’ve typed this out. Hopefully, I can find a way to make enough money to get her going soon.

This is Connor, Signing out!

One thought to “Electric Moped Update #4: The Death of a Beautiful Thing”

  1. Connor, that’s a great explanation/description as to why motors fail. It’s when you start getting to this level of understanding that you designs and projects get more robust. Keep up the good work!
    Remember WD-40 got its name because they had 39 failures befor it.

Leave a Reply

Your email address will not be published.