# Summer Ebike - Calculations for running CIMs at 24v

I’m working on an electric 125cc-equivalent motorcycle (>8 ft-lbs), powered by overvolted CIMs and Practice-quality batteries stacked to make 24v nominal.

I know from persistent rumors and empirical experience that CIMs will happily run at 24V (@RoboChair kept managing to not burn up his scooter at MadTown '19…), but now that I’m trying to run some calcs for a summer project… I don’t actually know how to model it.

This thread is the best I could find 24V Motors for 2010?

Ether touches on the theory, and it sounds like I can literally double the free speed, double the stall torque, and double the stall current and I’ll be aight for first order calcs…that’s 4x the power output? Seems too easy.

Unrelated to the motor calcs, anyone have a P/N for bearings that can replace the bushings on the CIM? I’d like to get to 30 minute runtimes with good efficiency. I accept I’ll still need to change the battery every ten minutes though, at least until I move to a lithium chemistry.

1 Like

What is your motivation for a 24V operating point?

More torque for less weight.

The explicit goal of this vehicle is to stick its nose in the air for ten minutes at a time between battery changes, not to be practical transportation.

(I might use a similar powertrain for a PowerWheelsRacing entry, but I need to have calcs (ideally backed up by measurments) that I’m not exceeding their 2kW limit.)

1 Like

Generally, increasing torque requires increased current.
Also, increasing Voltage increases speed.

My opinion is that you are likely to see a speed increase - due to the Voltage - but probably not a lot of torque increase.
You can certainly operate the motors at higher current with higher Voltage, but the losses you will see in the rotor are I^2*R losses - double the current and quadruple the losses.
In this case the losses will produce heat.

Changing the bushings into bearing will help with the mechanical losses - you will get benefit there.

If you can leverage a suitable gear ratio, you might see some benefit from operating the CIMs at a higher Voltage.

My take, anyway.

1 Like

This is where I got worried because it seemed too easy

My logic is:
-Winding resistance is constant
-Input voltage is doubled
-Stall current is doubled
-doubled stall current -> doubled stall torque

Yes, I agree with that point for the large gear ratio - between that, and the current draw limitations of the batteries, I’m getting a top speed on the order of 4mph right now… seems too low, like I should be able to get away with something faster.

What speed are you trying to reach?
What size wheels?
Do you have any idea what sort of acceleration you are looking for?

I don’t care for v1, probably need a shifting transmission to get anywhere useful in v3 or v4

12" rims with 130s should be about 17" dia after the tires get added but I’d like to confirm this.
Benchmarking against rear wheel torque anyway though

Break the rear loose or put the nose in the air, whichever comes first.
Trying to benchmark against a stock Grom, which I’m seeing as 8 ft lbs * ~6:1 in first = 48 ftlb at the rear wheel? Which I’m certainly exceeding right now, but my stall current situation isn’t great.

I suspect I need to tweak my “weight on driven wheel” value to get something closer to real expectations.

Not sure if this is at all helpful for what you’re trying to do, but I made a custom version of JVN a while ago which includes various force calculations. https://docs.google.com/spreadsheets/d/1Q2vFfh0Y2dJz4JQFbDL9l3cI6SAiu9pkZ7GoUsa0ytg/edit?usp=sharing

2 Likes

I like the way you do the current limiting to show the actual performance!

On you robot, are you programming it in on the supply side or on the stator? I saw that the default was 40A on the calculator, and it’s implemented as though it is programmed to affect the stator… if it’s implemented supply-side, your stator current values can be higher and you’ll get resulting higher torque.

(We ran 4x falcon drive, and eventually implemented a ~40A cap at the supply side but haven’t done anything on the stator)

1 Like

On our robot we didn’t wind up needing to use any current limiting on our drivetrain. One of the things I spent quite a while figuring out was calculating the “ideal” gear ratio, meaning the minimum ratio required to be traction limited (i.e. your wheels start to slip before your motors brown out). We wound up with a 12.92:1 reduction, which has worked very well for us this season.

Edit: Also, the ~40 amp limit is actually coincidental because of the battery resistance I had entered at the time. The spreadsheet will calculate the maximum current draw without going below a threshold voltage, and set the current limit accordingly. It’s left as an option so that you can manually override it if you wanted to.

2 Likes

I wonder if this could be an application for the Playing With Fusion Venom motor/controller, but I couldn’t find a maximum voltage rating.

They’re where I got the bearing idea from, but I don’t really want to deal with a CAN controller. Going with TalonSRs and Victor 883s for now…

@s-neff the PwF Venom will do PWM. You can drive it just like a Victor or a servo.

2 Likes

Good point, that’s fair.
I also don’t want to pay \$99 ea when my first guess from no documented limit is the caps would pop at 16.1v…
If I move back to a 12v system (ie, make a “practical” vehicle for “practical” things) the venoms are VERY high on my list. We probably would have used them this season if the Falcon didnt’ get delivered.

Taking 6 CIMs on a single drive,
30 Amp current limit per motor,
~30:1 gear reduction - motor shaft to wheel shaft,
17 inch wheel diameter,
350 pound system weight,

I get 16.6 mph with a peak wheel shaft torque of 70.4 FtLbs.
System reaches steady-state speed at ~4 seconds.

At that operating point, each CIM will draw ~10 Amps continuous and will be spinning at ~10krpm.

1 Like

The big thing you have to watch out for is heat - the CIM only has so much heat capacity, and that isn’t going to increase because you double the voltage.

The heat on a CIM is generated in the windings, and doesn’t work its way out too efficiently. Using a brushless motor (with a metal motor plate/block on your gearbox) or air cooled motor would be better for heat dissipation purposes.

Is there a cheap way I can do that implemented in a PWM controller?

I’m thinking a potentiometer and an arduino for V1 controls of the TalonSRs, with a ramp rate. I could add an inductive current sensor but I don’t know if my controller would be anywhere near fast enough for it to take advantage of it.

Otherwise I’m stuck adding motors and gearing until stall isn’t completely unreasonable…

At 24 Volts, I suspect you will need to pursue a more commercial motor drive - something like Advanced Motion Control, or perhaps Maxon. A lot of these units will operate from analog references - not sure about PWM inputs or CAN. Typically these units implement current limits.

My simulation indicates currents in excess of 75 Amps will result in wheel slip. On the other hand, my simulation assumes that all wheels are driven and the weight is equally distributed.

The SRs are rated to not blow up at 24v, which is all I need from them for V1. Not interested in spending money on commercial controllers.

75A x 6 motors = 450A, which is over double what I’d expect the typical FRC lead acid to deliver successfully. I could potentially keep the 30:1 gearing with a 2S2P battery setup; with an even number of motors, I could divide the wiring so that I have two parallel electrical powertrains and not actually wire the batteries into parallel, which would be nice from a wire sizing perspective.

For your system, you may want to seriously look into how to add significant capacitance on your distribution system. It’s all about energy transfer potential->kinetic->potential.
Starting will tend to brown-out your system with Voltage droops, but stopping will raise the system Voltage!

It may not be as simple as just adding capacitance; you may need to put a pre-charge circuit in front of the capacitance, you may need a discharge path for the capacitance as well.

As a final measure, you may want to look into an over-Voltage protection circuit - a.k.a a dump circuit. This sort of device will dump current into a resistor whenever the Voltage exceeds one threshold until the Voltage is less than a lower threshold.

2 Likes