Motor Current Limiting: More Power and Less Heat!!!

[cadstarsucks]

Hmmm... It might well be that they take short cuts never expecting to see the speed. Or perhalps the torque...was that free wheeling or under load?

Dan

 

[xyster]

Hmmm... It might well be that they take short cuts never expecting to see the speed. Or perhalps the torque...was that free wheeling or under load?

Dan

Here's one of the threads about 60 volt kollmorgen destruction:
http://endless-sphere.com/forums/viewtopic.php?t=1434&highlight=kollmorgen
Sounds like he was riding it at the time. Perhaps outrunner BLDC motors will do better from this standpoint.

 

[Miles]

Hmmm... It might well be that they take short cuts never expecting to see the speed. Or perhalps the torque...was that free wheeling or under load?

Or perhaps it's simply not designed to "run safely and reliably at 10-50K RPM".....

 

[cadstarsucks]

Here's one of the threads about 60 volt kollmorgen destruction:
http://endless-sphere.com/forums/viewtopic.php?t=1434&highlight=kollmorgen
Sounds like he was riding it at the time. Perhaps outrunner BLDC motors will do better from this standpoint.

Interesting... It was a crappy assembly job they did on it. I am usually thinking in terms of "hub" or "wheel" motors myself at any rate.

Dan

 

[safe]

My two cents worth...

I suspect the bigger problem is not high rpms (except in those cases like the Kollmorgen where the design itself seems not well suited for high speed with nothing but glue holding it together) but the heat that normally develops with higher voltage and the use of a BCL (battery current limited) controller. The real killer is the heat and not the bearings.

Let me point this out clearly with a set of charts.

The first chart shows how a small 24 volt motor might be given a higher voltage, but still keep the same controller type. (BCL)

The second chart shows how that same small 24 volt motor could be given the 48 volts, but this time you restrict the controller using the MCL (motor current limiting) idea.

It's pretty clear just by looking at the heat production of these two situations that the MCL approach creates a motor that is not stressing itself a great deal. One wonders if the case of the Kollmorgen was really one of overheating that made the glue fail and then it flew apart.

 

[cadstarsucks]

My two cents worth...

I suspect the bigger problem is not high rpms (except in those cases like the Kollmorgen where the design itself seems not well suited for high speed with nothing but glue holding it together) but the heat that normally develops with higher voltage and the use of a BCL (battery current limited) controller. The real killer is the heat and not the bearings.

Safe, in the case of most of the stuff it seems you've done you will have terrible motor life due to accelerated brush wear from speed (if you are actually going faster) and arcing.

Industrial motors have replaceable brushes just so they can be serviced readily.

Dan

 

[BiGH]

Most BLDC motors can run safely and reliably at 10-50K RPM - the only limitation is the bearings.

...

Basic enamel insulation is good for 500V or more.

As reported by Knoxie and others here, the popular brushless kollmorgen motors fly apart at around 60 volts. The magnets separate from the rotor. Don't know the RPMs, but doubt it's near 10k.
Outside of tiny RC motors, I've not heard of ebike-sized BLDC motors surviving those kind of RPMs and voltages. Can you site some examples? Do you think my 5304, spinning 700 rpm at 80 volts, can handle 500 volts with just better bearings?

I could go 250mph....

actually makes me wonder what the real voltage limits would be for the 4 and 5 series.

 

[fechter]

I could go 250mph....

actually makes me wonder what the real voltage limits would be for the 4 and 5 series.[/quote]

Several hundered volts I bet.

Like those land speed record bikers that ride behind a truck, if wind resistance wasn't the limiting factor, and the balance was perfect, you could easily top 250mph

 

[Jozzer]

So what determines the volt/amp limit for these motors? Insulation of wire on windings for voltage? How many amps before wire blows apart?
Seems thermal managment is the biggest limiting factor?

 

[fechter]

So what determines the volt/amp limit for these motors? Insulation of wire on windings for voltage? How many amps before wire blows apart?
Seems thermal managment is the biggest limiting factor?

I think you got it. :wink:

The maximum rpm the thing can spin before something flies apart is another limitation.

If you rewound one with smaller wire and lots more turns, you could easily run it on 500v.

 

[safe]

Seems thermal managment is the biggest limiting factor?

That's the area I've been focusing on. Most of the really damaging heat comes from low rpms and that's a strange artifact of the controllers we're using. If instead of allowing the extra current down at low rpms you restrict how much current gets through (the MCL concept) you effectively filter out some behavior that is harmful to the motors health. When the motor is running at higher rpms then it makes sense to allow it current to do it's job, but at low rpms it's not a good idea. Assuming you have gears you can sidestep the heat problems by filtering out inefficient motor situations entirely.

People here continually complain about the MCL idea because it robs the low end of that "surging" feeling and instead would place the "surge" up into the higher rpms like in a really "peaky" sports car or motorcycle.

 

[xyster]

People here continually complain about the MCL idea because it robs the low end of that "surging" feeling and instead would place the "surge" up into the higher rpms like in a really "peaky" sports car or motorcycle.

Exactly. Your designs could single-handedly castrate the electric motor's biggest advantage over the internal combustion engine. If you succeed, we'd all be stuck dependent on fossil fuels until the global ecosystem and everyone in it collapses in a heap of octane-fueled pollution, war, and misery. Then how would you feel about what you had done ... huh? HUH? Just who do you work for, anyway...BP? Texaco? The GOP? Who!?

:wink:

 

[BiGH]

People here continually complain about the MCL idea because it robs the low end of that "surging" feeling and instead would place the "surge" up into the higher rpms like in a really "peaky" sports car or motorcycle.

Exactly. Your designs could single-handedly castrate the electric motor's biggest advantage over the internal combustion engine. If you succeed, we'd all be stuck dependent on fossil fuels until the global ecosystem and everyone in it collapses in a heap of octane-fueled pollution, war, and misery. Then how would you feel about what you had done ... huh? HUH? Just who do you work for, anyway...BP? Texaco? The GOP? Who!?

:wink:

hehe sounds like the debate of high torque vs high power welcome to fast and the furious 1

whats also important here, is that with motor current limiting you could run much higher power levels than you could normally run though.

I for one like the idea.

 

[safe]

whats also important here, is that with motor current limiting you could run much higher power levels than you could normally run though.

I for one like the idea.

Heat is the only constant... the motor can only handle so much of it before it burns up. Power is all about what gear you are running. If you want to increase your torque you can simply gear down to increase torque while increasing the rpms. It's final horsepower that really matters and the best horsepower comes while the motor is in it's lower heat rpms. (which is high in the powerband)

So MCL just allows a good motor to run even better...

I forget about "The Fast and the Furious", did they talk about torque and power in that movie? I do suspect it's the same discussion that people have been having about motors for a long time. The Indy Cars run at about 12,000 rpm for the very reason that you get more peak power that way... but if they aren't careful in the pits they can stall on exit because the torque is very low at low rpms. So this issue plays itself out in real life already...

 

[Jozzer]

Hmm, this is exactly what my last 2 hubmotors died of. Still, given a choice, I'd rather have the stonking torque at low end, UNTILL things get warm, and only then power gets limited :lol:

 

[cadstarsucks]

Hmm, this is exactly what my last 2 hubmotors died of. Still, given a choice, I'd rather have the stonking torque at low end, UNTILL things get warm, and only then power gets limited :lol:

That is what we do with treadmills...but we doing it for the sake of the transistor - we are only sending 25A into a 30A 130V motor.

Dan

 

[BiGH]

nah was an underlying theme of the movie towards the end - with the big black detroit muscle and the hipo jap turbo car.

 

[safe]

Hmm, this is exactly what my last 2 hubmotors died of.

Hub motors don't have gears. The MCL solution is specific to geared bikes (one's that channel the motor power through the gearing) and does not apply to hub motors. The hub motor is best suited with the conventional BCL controller.

Gears change everything because you can shift your powerband around. The Hub motor is a fixed gear and that fixes the powerband.

The short answer:

You're stuck... the only way to escape is with more voltage and even there you can overheat your hub motor. The X5 hub motors survive because of sheer mass (a whopping 25 lbs of metal) and so that large mass dissippates heat across a larger volume of metal.

I would like to see someone take a hub motor and severely overvolt it and then prevent overheating with MCL. It would be really weird to ride because it would have very little power at low speeds, but the faster you went the more power you would have. (and you wouldn't have the overheating problems you would normally have)

Still... the MCL solution is really a geared bike solution... hub motors are more or less fixed in what they can do...

 

[cadstarsucks]

Hmm, this is exactly what my last 2 hubmotors died of.

I would like to see someone take a hub motor and severely overvolt it and then prevent overheating with MCL. It would be really weird to ride because it would have very little power at low speeds, but the faster you went the more power you would have. (and you wouldn't have the overheating problems you would normally have)

Still... the MCL solution is really a geared bike solution... hub motors are more or less fixed in what they can do...
[/b][/i][/color]

I disagree. A low speed hub on a small rim would give you low speed torque while rerating ( I refuse to call it overvolting since the elevated voltage in at a higher speed in an AC machine simply brings the current back to where it belongs ) gives you high speed torque.

Dan

 

[Jozzer]

Whats more, since the efficiency is lower at lower rpm's, it means that when deciding the maximum continuous power you have to fix it at the worst scenario. If current were limited in relationship to motor temperature then you would always enjoy the highest amount of power saftly possible.
Besides, i'd argue that its YOU who shouldn't need current limiting at low RPM, since if your gears are properly organised you shouldn't need to run out of the motors sweetspot. :lol:
And i'm not stuck as such, I just need to limit the power to safer levels. Though I regret the loss of the motors, they died learning thier true limits. (at pretty high voltages). running a 4011 @ 90v at the moment whilst the Puma is worked on, it doesn't feel very nice surprisingly, mainly BECAUSE it's power decreases so drasticly as it accelerates (check out a 4011 in a 24" wheel @90v 40A on the simulator at ebikes.ca). Cant wait to get the puma back in there tbh!

 

[cadstarsucks]

And i'm not stuck as such, I just need to limit the power to safer levels. Though I regret the loss of the motors, they died learning thier true limits. (at pretty high voltages). running a 4011 @ 90v at the moment whilst the Puma is worked on, it doesn't feel very nice surprisingly, mainly BECAUSE it's power decreases so drasticly as it accelerates (check out a 4011 in a 24" wheel @90v 40A on the simulator at ebikes.ca). Cant wait to get the puma back in there tbh!

It seems as though that one is a BDC instead of a BLDC. While a bit more complicated BLDC is more flexible.

Dan

 

[fechter]

If current were limited in relationship to motor temperature then you would always enjoy the highest amount of power saftly possible....

There we go! That's how I'd like to see it done.

Motor Heat Limiting controller: Maximum power, limited heat.

Ride like a madman or an idiot and NEVER burn up your motor. Of couse, you could always adjust the maximum current limit as well, to extend range or improve performance as desired.

It would require a temperature sensor of some type, which could be a challenge on a brushed motor. The windings could heat up faster than the piece of metal the temp sensor is mounted to. Indirect infrared would work. On a brushless motor it's much more straightforward. The biggest challenge is running the wires, which could be very small.

It would also take a circuit to limit the current as a function of temperature. Something like the CA could be programmed for this. I'd go for an analog circuit.

 

[Jozzer]

Excellent Fetcher, now what might that circuit look like? :lol:
(http://endless-sphere.com/forums/viewtopic.php?t=278&start=30 hint hint)

 

[safe]

Motor Heat Limiting controller: Maximum power, limited heat.

The problem I'd see is that in the beginning when your motor was cool this type of controller would allow a lot of current to get through in what are often the least efficient moments. Then later when the motor begins to get too hot the controller would be forced to restrict very heavily all the time in order to allow the motor to cool. The rider would get in one or two fast laps and then all of a sudden he would be underpowered for the rest of his ride. All his shift points are going to be off and he's going to be struggling to adapt to a motor that now seems sluggish.

If the MCL approach is used you never allow the motor to overheat and so the rider is given a "temperamental" (very peaky powerband) but consistent motor that allows the rider to keep his exact same shifting points all throughout his ride.

That's the biggest problem with a heat limited controller, the motor will seem to really sag as time goes on. (but at least it can no longer be destroyed by heat, so it does solve that problem)

 

[fechter]

If you know the limitations of your motor, you can prevent it from getting too hot in the first place by controlling the throttle and avoiding the inefficient operating zone.

The amount of power the motor can handle without overheating depends a lot on the outside air temperature and air flow. On cold days, you can push it much harder than on hot days.

I'll have to dig into the available thermometer circuits. I seem to recall a transistor sized device that outputs a voltage that's porportional to temperature. One of these could go into a comparator similar to my current limiter circuit. In fact, it might be good to combine the current limiter and the temperature limiter in one circuit.