ESC issues with large vehicles and brushless motors

[bart_dood]

Hi folks,

After my success with my ebike (900 miles and counting!) I want to move on to something larger. I've been doing some background reading and formulating some ideas but I have a question about electronic speed controllers for brushless motors I hope someone can answer.
As I understand it, if a vehicle is stopped and there is no clutch between the drivetrain and motor if the throttle is floored the motor(s) will pull lots and lots of current (spike) which overloads the controller. This is why folks simply can't use a model motor with a model controller without being super careful.

I've been thinking of ways around this; couldn't a resistor be placed on the DC portion of the controller (battery side) to limit the current spike to a managable level? sure it would dull the acceleration from a standstill but it would seem to solve this problem.

Incidentally I'm not going for massive acceleration so I don't care if the performance from a dead stop is not as good as it could be with a infinite current ESC.

Thanks!

Edit: I'm toying with now with 3 or 4 turnegy motors (the 6500 or 7000 watt versions) combined with 3 or 4 180 amp (360 amp peak) marine controllers (with water cooling enabled). I'll hook these up to 4-5x lead acid batteries in series. It just occured to me that each battery will have internal resistance plus all the wires and connectors, so this might actually limit the peak current by itself....hmmmm

 

[bart_dood]

I guess one of the problems with adding a resistor is the power loss and hence loss of efficiency..some kind of variable resistor that would change to limit the peaks but drop to zero when not needed would be better...

 

[vanilla ice]

What kind of vehicle?

 

[bart_dood]

What kind of vehicle?

A bug!

 

[bart_dood]

I've done some calculations on cable resistance and adding in cell resistance I'm very close to have a current limit based just off the entire resistance of the system.
However I thought of another way around this, I could have a microcontroller on the throttle that would monitor the rpm of the motors and limit absolute full throttle amount based off rpm, it would limit most at a full stop and as the rpm would rise so would the max throttle setting.

Would this limit the current effectively?

 

[Gordo]

There are a couple of things you might want to look at for cave man ideas? Golf carts and trolley busses. In the end, money spent on a modern controller is probably well spent.

 

[bart_dood]

There are a couple of things you might want to look at for cave man ideas? Golf carts and trolley busses. In the end, money spent on a modern controller is probably well spent.

I guess I wasn't aware there were other controller options open for this kind of motor? do you have any links?

 

[Gordo]

No Links, but I have 2 disassembled golf carts in the back yard with the "controller" in a box in the shop. The controller is just a large pot with the throttle pedal connected to the wiper and large wire wound resistors on the back of a bakelite board. The wiper is like the hand of a clock and there are large copper studs where the numbers would be on a clock face.
I always wondered why golfers seem to quickly accelerate to full throttle? This is most efficient method of using a golfcart. Anything less than full throttle and you are just making heat from the shunts. THIS APPEARS TO BE WHAT YOU ARE THINKING OF BUT THE SHUNTS ARE ALL IN PARALLEL WITH THE MOTOR TO GROUND. AT FULL THROTTLE, THERE IS NO SHUNT IN THE CIRCUIT. JUST WHAT YOU ASKED FOR, ZERO SERIES RESISTANCE AT FULL THROTTLE.
To be very clear; I am not suggesting you use this style of "controller". I would not use one even if it was free. The standard e-bike controller using PWM is far more efficient and well worth the cost.

 

[amberwolf]

Since he's talking about a brushless motor, the type of controller you're describing can't work, as it doesn't do any commutation (which a controller must do for a brushless motor, but which is done by the motor itself in a brushed motor, for which your described controller *can* work).

@bard_dood: I really can't recommend a resistor inline with the battery, even if it's switched out of circuit for full speed/etc., because you're going to have a lot of waste power from it, meaning a lot of waste heat. Sure you could do it, but unless you need a vehicle heater I'd pass on that.

Now, for a controller for a brushless motor in a regular EV like a Bug, there are options. Curtis, Alltrax, and I forget who else make some that I've heard of over on DIY Electric Car forums and seen here and there on EVAlbum. DIYEC also has at least one thread for a build using some unknown Chinese brushless motor and controller bought as a pair.

If you can find them in a junkyard, some city buses use a Siemens brushless controller and motor (there's a build on DIYEC about that, too, IIRC).

 

[Gordo]

Yes, you are correct. The cart controllers only work for the brushed motors and not very efficiently at part throttle.

 

[bart_dood]

Since he's talking about a brushless motor, the type of controller you're describing can't work, as it doesn't do any commutation (which a controller must do for a brushless motor, but which is done by the motor itself in a brushed motor, for which your described controller *can* work).

@bard_dood: I really can't recommend a resistor inline with the battery, even if it's switched out of circuit for full speed/etc., because you're going to have a lot of waste power from it, meaning a lot of waste heat. Sure you could do it, but unless you need a vehicle heater I'd pass on that.

Now, for a controller for a brushless motor in a regular EV like a Bug, there are options. Curtis, Alltrax, and I forget who else make some that I've heard of over on DIY Electric Car forums and seen here and there on EVAlbum. DIYEC also has at least one thread for a build using some unknown Chinese brushless motor and controller bought as a pair.

If you can find them in a junkyard, some city buses use a Siemens brushless controller and motor (there's a build on DIYEC about that, too, IIRC).

Thanks! I'm actually looking into using the hybrid motor and controller from a honda automobile to drive my bug. Its designed for this kind of thing, the only problem is the reverse engineering required to make it work. The module that controls the 3 phases is quite 'dumb' in that it doesn't have the pwm and various curves programmed into it (its controlled by another module that won't work without a lot of drive signals which look to be difficult to reproduce). So I'm looking at developing a microcontroller system to drive it.

I'll post further information and images when I have them on the forum.

 

[liveforphysics]

If I'm not mistaken, the controller for an Insight IMA only looks at a simple PWM signal from the ECU to operate. It's something like 50% duty cycle means do nothing, below 50% means start regening, and above 50% means start cranking out some torque.

 

[bart_dood]

If I'm not mistaken, the controller for an Insight IMA only looks at a simple PWM signal from the ECU to operate. It's something like 50% duty cycle means do nothing, below 50% means start regening, and above 50% means start cranking out some torque.

Partially correct, unfortunately honda split the functions of the motor controller up. There are two modules required to drive the 3 phase motor, one is called the MCM and one the MDM, the MDM is a low level dumb IGBT module system that is driven by pulses to tell it to fire the invididual U, V, W etc. The MCM had a microcontroller in it that takes in signals from the position sensors and current sensors on each phase leg, its pre-programmed with the various PWM for loads and speeds etc and sends signals to the MDM. The MCM does just what you mention and runs off a simple PWM control signal for assist and regen etc.

Unfortunately the MCM requires lots of input signals to work, so reproducing them to convince it there is still a gasoline engine with a full transmission attached is rather difficult, its even more difficult as I don't have a car to measure it from (sold my insight years ago).
There is a bit of a community over on the insight forum trying to work on this, there is also a paper written by the department of energy where they drove a honda MDM with their own electronics, I'm trying to get some info out of them currently.

Any info you could pass along on 3 phase controllers would be much appreciated live for physics, I've seen your various posts on the forum and yu da man when it comes to high power electronics!!

P.S. I picked up a decent used bug tranny tonight to design around, I just finished degreasing it. I'm about to pull the trigger on buying some IMA motors too...watch this space...