3

I am a big boy (~230lbs) living on a mountain. My 1000 watt electric hub powered by 78 18630 Lion batteries in 13 cells barely moves me level let alone up a hill (which was its intended purpose). I tested its current to find it only pulling ~6.7 amps on a wheel stand. Times 48v only gives me just under 322 watts. Did I get ripped off? Is there something I am missing?

Under the circumstances I'm considering removing the rim and spokes, attaching a cog and connecting it (centered in the frame) into the existing chain/de-railer system, directly.

Suggestions anyone?

  • 4
    Try exactly the same current test but with the wheel brakes dragging hard but not locking up the wheel. You want to simulate some resistance but not stall the motor completely. – Criggie Jun 2 '18 at 8:30
  • Are your batteries fully charged? Are they repurposed/reused ? – Criggie Jun 2 '18 at 8:33
  • Does your bike function properly other than the motor problems? Is it in good mechanical condition with wheels that spin freely and coast for a while? – Criggie Jun 2 '18 at 8:35
  • 2
    How, precisely, did you test the current draw? The 1000 watts would be max, with, as Criggie says, a load that is maybe half or 2/3rds of stall load. – Daniel R Hicks Jun 2 '18 at 11:48
  • 1
    There are many considerations here, and much more detail would be required to give you a comprehensive answer (details about all of: the batteries, the controller, the hub, how they are connected, used, and constructed (e.g. the actual electronics) matter). While this is related to bicycles, you will probably get more comprehensive answers on Electrical Engineering. – Makyen Jun 2 '18 at 20:09
6

Even if you've only got around 300W, that should shift you reasonably well on the level. The limit here is 250W and that can manage 15mph with your weight on the bike, though the acceleration isn't great. But the current draw with no load (on a stand) will be a lot less than the loaded current anyway, so you probably can get the rated power or close to it. I assume it's actually rated to 1kW at 48V

The motor controller is important - is that suitable for over 20A? Is your wiring?

  • The motor controller came with the hub, from China via eBay, without warranty and minimal documentation. I wired it with 10 gauge joined with terminals from an electric wheelchair. Unfortunately I am disabled and really struggle to ascend a 10% grade with pedal assistance. I need more power. – Jean-Paul McCoy Jun 2 '18 at 7:11
  • @Jean-PaulMcCoy just FYI, everyone struggles to go up 10% grades on a bike. The pro's make it look easy, but they're working hard like the rest of us. – Criggie Jun 2 '18 at 8:24
  • 2
    Sorry to say this, but you might be looking for a scooter, not a bicycle. – ojs Jun 2 '18 at 8:42
  • 9
    If your motor controller came from China via eBay without warranty and minimal documentation, then I wouldn't be sure it's actually 1000W. – immibis Jun 2 '18 at 8:55
5

On a wheel stand there is not much resistence so the motor doesn't have to use much torque (current) to spin the wheel, so the power draw is that low.

The motor outputs max power at half it's rpm range:

enter image description here

Your only way to test this is on a dyno or out in the field. Blast full power to the controller and mechanically restrict the motor to only go half the speed as it would go unrestricted - you are at max power then.


The other thing is the power rating itself. The power quoted is input power and not the mechanical motor output. This is because the motor controller is an integral part of the system and runs as an inverter.

the mechanical motor output + the heat the motor produces = input power

To see how much power you can get onto the ground you need to use dyno again or you can get a datasheet of the motor and look up the efficiency curve.

  • It appears that OP is measuring exactly the input power because he counts amps × battery voltage. – Agent_L Jun 2 '18 at 14:15
  • 1
    @Agent_L That is the first half of my answer, why he doesn't get the full input power read, I am optimistic that he could get close to it. The second part is just a leading statement to the efficiency argument that not all 1000W systems are equal and that the input power has only a partial indication of how well the powertrain will perform when buying one. – Jerryno Jun 2 '18 at 14:30
  • 1
    Even that's system input power, and the voltage drop in the wires or connections can be significant – Chris H Jun 2 '18 at 19:43
3

Cannibalising your wheel to make a chain drive will be significantly difficult.

  1. Getting a toothed wheel into the chainline and getting it to turn at a useful speed while maintaining chain tension.

  2. Depending on where you put the motor, you'll probably stop the front derailleur from switching between chainrings, because the top run of chain is under tension.

  3. Thinking about this, your hub motor will not work fixed to the frame without a complete redesign. That's because the power/control cables assume the axle is fixed and the body of the motor rotates with the wheel. In your case you want the body of the motor fixed to the frame, and have the axle turn.

If you want to go this route, you need a motor designed for fixed mounting that drives its output shaft only.

  • I'm glad you commented on my considered cannibalism. To be clear I am aware that the challenge would be great and cost/benefit, marginal. Yet as a compulsive gadgeteer with more time than money I must consider the challenge hypothetically at least. – Jean-Paul McCoy Jun 2 '18 at 20:21
  • ...I have a vision of a dynamic camber rear suspension to which the narrow form factor of the hub as a drive is desirable. Expect a link to dedicated post citing yourself. – Jean-Paul McCoy Jun 2 '18 at 21:36
  • Sorry if answer came across as condescending - its often difficult to know how technical a person is. Do please let us know how you get on - I'm genuinely curious to see your progress. – Criggie Jun 2 '18 at 22:38
  • I meant no disrespect and certainly perceived none. I do have a CS degree and am a decent welder. I am new to Lion battery integration, charging, balance management etc. I assembled my battery with a homebuilt spot welder, from an old microwave and an Arduino. (Good results). I did not mention that my primary power source is small trailer mounted gas turbine generator built from a small Subaru trurbocharger spinning a secondary 7" axial jet turbine mounted alternator (I intend to upgrade with a neodymium stator). It is coming along nicely. The recumbent frame is still in planning. – Jean-Paul McCoy Jun 3 '18 at 11:46
2

First, you need to measure the current at close-to-stall which is when electric motors draw the most current. Ideally, you’d have a dyno or rollers but with bikes wheel traction on the rollers can be a problem. So I’d try to stall the wheel with your brakes. Only do it in short spurts to not overheat the brake or hub or controller.

Scenarios (note by fake I mean re-labeled):

  1. Fake controller, real 1000 watt hub: You do have a good chance of burning out the MOSFETs of the controller if it’s a 350 watt controller labeled as a 1000 watt if your wheel can draw 20 amps. But you’d want to buy a new controller anyway.

  2. Real controller, fake hub : if your controller is rated for 1000 watts but your hub is only built for 350 then you’ll only get 350 Watts. Nothing will burn out. It’s the same as putting a 60 watt bulb on a switch rated for 20 amps, it won’t make the bulb any brighter

  3. Fake hub and fake controller: you’ll only get 350 watts but if the controller is dodgy you might burn out the controller before the hub.

Scenario #1 is your best case as a new controller would fix it. But if it were the case, the controller would have blown by now (or gone into thermal shutdown / self protect). Does your bike throw a code and die completely when it tries to take you up a hill? If not, then it’s incapable of doing so even with a new controller - it’s the hub.

The most likely scenario is #2 or #3 which are both expensive.

Note you can measure the winding resistance if you have a very good ohmmeter that can measure in the low ohm range (a crappy harbor freight won’t work). Your stator resistance should be around 2.4 ohms. If your stator are measuring 6 ohms then your motor can only do 380 stall watts. But measuring low ohm resistance really does take a very good meter as a cheap meter will just be measuring itself or it’s own lead resistance.

  • I'm using a Klein Tools CL2000 True RMS which set me back $130. I'm real happy with it actually. I did not get a chance to test the resistance today but definitely believe the controller is garbage. It has lots of power for about 20 seconds and then cuts out. I thought the battery was inadequate and doubled the cells to 13 blocks of 6 parallel. All new, highly rated SAMSUNG 25R 18650 2500mAh HIGH DRAIN 25A Rechargeable, spot welded with with a BMS personally. Performance is the same. Assuming the hub is good, could you recommend a decent controller, (ideally one I can integrate to an Arduino)? – Jean-Paul McCoy Jun 3 '18 at 10:59

Your Answer

By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

Not the answer you're looking for? Browse other questions tagged or ask your own question.