PDA

View Full Version : Overclocking controllers and motors



4dollarjoe
01 September 2016, 0615
Hey all I've switched the gear ratio on my bike from a 4:1 to a 2.67:1 to achieve higher top end.

Currently I am using an me0709 motor and an alltrax spm 72300


Now I can hit about 70mph top speed (confirmed with gps speedometer), but only for less than a minute. Beyond that there is a noticeable loss of toque. Riding on the highway I slowly hit a sluggish 45mph. I believe this was because a lack of torque from the continuous 100 amperage

With the 4:1 I was able to keep riding 55 mph continuous, although after the peak time period it goes from feeling like a crotch rocket to a scooter.


It got me thinking, with the amount of copper in there, I don't think that the size of the conductive material is a limiter, after all motenergy sells some motors around the same diameter that can do 450 amps peak. The me1003 motor is essentially 1 inch longer than mine with a double brush holder and it can do 400 peak and 200 continuous, maybe that extra inch adds some thermal mass that allows it to reach these levels. considering the diameter of the rotor is no different than mine.

If you could cool it enough, do you think you could have a 400 amps continuous?


Do you guys know how the alltrax controllers determine what is the limit on the peak? Does it just have a simple timer to switch to continuous mode?


I feel like temperature is the real limiter to the true peak power of the motor and controller.


If you look at the air intake of the motor, the metal bracket that holds the bearing for the rotor and the rubberized backing of the brush holder take up the majority of the space, blocking air intake.

Beyond that the air intake is on the side, reducing the flowrate of air into the motor, and wasting the useful air that you fight against as drag.




In summary I think adding a temperature sensor that could interface with the controller and adding in some type of hoodscoop and heatsink could conceivably 'overclock' the cheap dc motors that are available today.


Any ECE people or people knowlegeable of these systems that can add their input it is much appreciated.

dcb
01 September 2016, 1911
https://evmc2.files.wordpress.com/2011/09/72v-me0709-cw-2011-9-1.pdf
6 KW cont, 13.4 KW peak

just a swag, but you were probably only doing 1670 rpm @ 45 w/2.67, the power required vs speed graph went exponential while the power vs rpm remained linear, and you stalled out at 45mph (@ ~4kw instead of the rated 6kw continuous).

with some more swags it looks like your top mph continuous (regardless of gearing) should be about 55mph, on that bike with that drag. And top speed at 13.4kw is ~74mph.

so 4:1 looks like a fairly optimal gear ratio, considering the continuous demands and continuous power rating. (The temporary extra current capacity is designed to get you up to speed, not keep you there).

and 2.67:1 is close to optimal for top speed when you have 13.4kw on tap.

When you talk of overclocking, I think of spinning faster, i.e. a 96 volt pack at 4:1 would get you close to 70mph. Don't know if the motor can handle it (brush advance might help). I also think of liquid cooling. If heat is your problem, throwing more current at it won't help, though a faster spinning motor will spin its fan faster too. continuous might be 60mph at 96v. Of course it means a different controller, so you can experiment with more amps for longer too, but I think more volts should be on the menu too.

Hugues
02 September 2016, 0404
Hi,

2.67:1 seems to be very low. I had over heating problem on my AC-20 at 4:1, now running at 7.04: 1

Even at that ratio i have to ventilate my motor, i have 3 fans:
- the original rear end fan of the AC-20 that runs only when the motor spins
- i've added a blower in the air intake of the motor to force fresh air in and hot air out, especially for the rotor
- and this year i've added a third fan to ventilate the stator envelope from the outside.

From what i can see, this is now sufficient.

4dollarjoe
02 September 2016, 0759
Well noted, To be fair the (about) 70 figure I gave was going down a very slight decline; the actual max I read was 67. Going back up that same incline the max clocked at 63.


In addition I am running peak amp mode. I changed my torque curve to be a very aggressive vertical curve while the speed curve is linear. I definitely feel that i am getting less mpc on this set up, so I think you are right about the 4:1 being better for the continuous demands; making a two speed transmission good for saving range so you can have the energy hogging top end only when you need it.

I talked to one of my ece friends and I think I might try adding two more modules on a saddle bag so I can hit 76v nominal.


I feel like air cooling isn't enough for these motors, and If you look at AMR's crazy offerings the oil cooling must definitely help. Maybe I will see if I can get a sealed motor and attatch a 12v computer radiator or something.


In addition, the impeller specifically on my motor could be improved. They have straight fins (probably easier manufacturing, and also allows for good cooling no matter which way the motor is spining). Designing a better impeller to only take in mind CCW rotation might be much better and increase air flow rate while reducing loss.

Hugues
02 September 2016, 0806
... Designing a better impeller to only take in mind CCW rotation might be much better and increase air flow rate while reducing loss.

I tried that but there were minimal gains. I gained much more with my first blower, which runs even if the motor is at rest. You should try that first, much cheaper and easier than liquid cooling.

podolefsky
02 September 2016, 1235
The motor won't limit you to the continuous amperage. If you exceed it for long enough the motor will simply burn up. If you didn't burn up the motor, then it's probably the controller cutting back current. Some motors like HPEVS have a temperature sensor that feeds back to the controller, so you won't burn it up. But the Alltrax/ME0709 combo doesn't. The controller is blind to what the motor is doing.

The controller doesn't have a timer (as far as I know). It monitors it's own temperature and cuts back current if the temp gets too high. My best guess is that's what you're experiencing. At that RPM and power draw, the motor current is going to be pretty high, much higher than battery current, and the controller will heat up fairly quickly.

There are two main reasons you get higher continuous power at higher RPM. One is that the fan runs faster, so you get slightly better cooling. The other reason, and probably the larger effect, is that at higher RPM you need less motor current for the same power. So you get less I^2*R losses and less heating. That goes for both the motor and the controller.

So - if you are going 55mph, at 4:1 the motor might be just fine. If you switch to 2.7:1, then the motor needs more current to maintain the same power. Both the motor and controller will heat up more. On the battery side, the power is (roughly) the same, so at 55mph you will see the same battery current regardless of gear ratio.

dcb
02 September 2016, 1312
So many cool setups!

re blower: spinning no load looks like it can be effective at cooling quickly, with the right considerations. *could* be done in reverse with just a free-wheel, *might* be worth the cost of adding a reversing relay or doing without regen in the right circumstance.
https://youtu.be/h6-G8Zn0CrE?t=45

nedfunnell
09 September 2016, 0714
Maybe I missed something, but I am surprised nobody has asked yet: What battery are you using? The controller is only going to limit you if the battery can source enough current. Are you sure your battery voltage was not sagging?

podolefsky
09 September 2016, 0849
Maybe I missed something, but I am surprised nobody has asked yet: What battery are you using? The controller is only going to limit you if the battery can source enough current. Are you sure your battery voltage was not sagging?

If you change gearing, then the current and voltage change *motor side*. If you're at the same speed you're going to see basically the same current, voltage, and sag battery side. Since he could go 55mph before, the batteries will provide enough power to go 55 regardless of gearing. The controller is the bottleneck.

nedfunnell
09 September 2016, 0948
If you change gearing, then the current and voltage change *motor side*. If you're at the same speed you're going to see basically the same current, voltage, and sag battery side. Since he could go 55mph before, the batteries will provide enough power to go 55 regardless of gearing. The controller is the bottleneck.

Hmm. Is that true for all kinds of controllers? I know brushless/AC controllers do their own thing with phase currents, but that's new to me for PMDC motor controller. I've got an Alltrax that is a duty cycle/voltage control- would it be the same for a torque-control style controller?

podolefsky
09 September 2016, 1049
Hmm. Is that true for all kinds of controllers? I know brushless/AC controllers do their own thing with phase currents, but that's new to me for PMDC motor controller. I've got an Alltrax that is a duty cycle/voltage control- would it be the same for a torque-control style controller?

It's the same for all controllers.

To go 55 you need, say, 5kW. The power required is just from rolling and air resistance, so it's independent of gearing (ignoring any small difference in efficiency).

To get 5kW you need 70A @ 72V. Since sag depends on current, there's only going to be one current/voltage combination that gives you 5kW for a given battery.

The relationship between battery current and motor current is actually simplest for PMDC controllers. Pin = Pout, so Vin*Iin = Vout*Iout. Vout ~ RPM (at steady state), so for a given Pout, as motor speed goes down motor current goes up proportionally.

AC systems follow roughly the same rule, although the voltage/speed relationship isn't as straight forward as for PMDC.

4dollarjoe
09 September 2016, 1122
I'm using nissan leaf batteries that can do the continous and max currents, so I think that isn't the problem. However voltage sag is prevalent. But it's within the C rate limits listed on hybrid auto.
I am also next door to a battery company that has a bunch of academic papers on their computer. They were nice enough to let me look up the same cells and if I remember correctly they could handle 300 amps per cell so since the nissan leaf modules are 2s2p it should handle it.

Definitely outdoor temps are affecting it to. Today is really humid and my controller fan started right up as soon as I turned on the ignition. The other day when I got slogged back to 45mph was really hot.