Is this motor overvolting scheme safe?

[dknguyen]

Hi all. Suppose I had a robot motor rated for 12V that put out whatever amount of torque and speed. This is just fine when the robot is climbing something like stairs or a rock pile, but leaves speed to be desired when the robot is just cruising on relatively flat terrain.

I cannot use a bigger motor, and I cannot use a faster, higher power motor of the same size since I cannot easily add any more gearing (and I don't want those extra transmission losses). I was thinking that I could run the motor off of an 18V battery. During climbing when load (and current) are highest, I would limit the PWM duty cycle so that the motor effectively sees rated voltage. But during cruising when current draw is less, I would run the motor on the full overvolted battery voltage to get more speed.

THis would be accomplished by monitoring the current and always keeping it below the rated motor current (at 12V) so when current increases during climbing, it's limited so the motor won't burnout, but when the load is low, the current is increased (via overvolt increase) so that the speed would be higher.

Does anyone see anything wrong with this? Running a motor at higher voltage, but limiting the PWM duty cycle so it "sees" rated voltage when it is most likely to burn out (I'm thinking there may be problems that arise from the fact that a during PWM overvolt pulses are still applied to the motor so the voltage at the motor is actually above rated)? These are RC truck motors so I am not sure what they rate the insulation breakdown as (would it be higher than several times rated voltage? Or would it just be 2 or 3? or even less?). Or do I not even have to worry about breakdown voltage since the voltage is so low and they are probably using standard wire which is rated at 100V or higher? I've also read that smaller motors (like these ones) are sensitive to overvolting (if it's even possible, I do not know).

Does the life a 12V motor decrease when you run it from a 24V supply but you are only running it at a PWM duty cycle of 50% max? The current the motor gets is the same as if a 12V battery was used, but during PWM 24V pulses are is still being applied to the motor.

THanks.

 

[Pommie]

I think the only problem you will have is mechanical. You will effectively be running at 150% of rated RPM and the main problem I see with that is increased friction at the commutator and possible mechanical failure of the same. However, if you stay below the no load speed I think you should be OK.

Also, running at 24V and 50% duty cycle is not the same as 12V. For the same RPM, doubling the voltage may increase the current many fold.

For example, for a motor that produces 1000 RPM/Volt.
If you apply 12V with no load it will accelerate until the back EMF balances the applied voltage - this will occur at 12000RPM.
The same motor with a load may slow down to 8000RPM, the current draw will now be (12-8)/R. (12V applied - 8V back EMF)/DC resistance of the motor.
That same motor with 24V applied running at the same speed will draw (24-8)/R, which is 4 times the current. That is 8 times the power!!
The above does not take inductance or any snubber circuitry into account.

I think you can do what you propose but you should limit the applied power to the motor.

Mike.

 

[Nigel Goodwin]

Check the robot wars sites, it's normal to run the motors at higher voltages - at least double, and sometimes treble their ratings. Bear in mind these are used under high load, but only operate for 5 minutes or so.

One of the top robot wars guys builds effects for movies, and models for museums etc. - he's been using 12V wiper motors for decades, and running them on 24V for more speed and power, he's never had one fail!.

 

[Hero999]

Watch the power dissipation or even the internal temperature.

Opening the motor up an installing a temperature sensor is probably a good idea providing you can mount it to the hottest spot (probably near the brushes on the armature). You don't even need to cut the power if it overheats, just reduce the power dissipation by ruducing the duty cycle.

Don't forget that using a larger motor at the rated power will be more efficient than a smaller moter at more than the rated power due to the copper losse being higher.

 

[dknguyen]

Ive decided to just go with a brushless motor which handles higher voltages anyways and outputs a little bit more than the power I need. Add an extra electronics aspect of building the driver onto my project.

 

[Oznog]

50% bigger wheels would do the same thing, if you could fit them. That'll do much better on rough terrain.

Wait- "stairs" and "rock piles"? What sort of monster-bot is this?

 

[dknguyen]

I am using 50% bigger wheels (7.5" from 5"). I did this since larger wheels climb better. But they reduce force so I had to gear it down more to still be able to climb the same inclines. But on flat surfaces I still want speed to take advantage of one of the real advantages wheels have over legs, so I need a motor that can put out a lot of torque at low speed, but a lot of speed at low torque while gearing tries to trade one for the other.

I just want to make a robot that's not limited in the areas it can go. Slow enough for indoors, fast enough for outdoors, and won't get be confined by silly stairs or curbs. It should be able to go on snow if I get snowchains for the tires. I'm trying to think of a way to waterproof the motors the motors and electronics, especially since the motors are right next to the wheels.

It's basically going to be like an automated RC rock crawler truck.

 

[Hero999]

I take it that building some sort of multi ratio gear box is out of the question?

Can't you simply use a high voltage on flat then reduce the voltage when going up hill to protect the motor?

 

[Sceadwian]

Use a constant variable transmission (variable belt drive) the transmission becomes an active device at that point but this isn't a car, the torque limitations don't apply the same way. It's like a pulley system attach between the motor and drive wheels, adjusting one motor in a precise way results in a variable transfer of torque/RPM allowing you to keep the motor in highest efficiency while optimizing outside torque/speed

 

[Hero999]

Good idea.

CVTs are normally used in scooters and other small vehicles.

I don't know if they are any suitable for a small robot though.

 

[dknguyen]

I take it that building some sort of multi ratio gear box is out of the question?

Can't you simply use a high voltage on flat then reduce the voltage when going up hill to protect the motor?

Isn't that what we have been talking about this whole time? The problem is that the (brushed) motor that fits into the space available cannot reach optimal cruising speed at rated voltage.

A transmission or belt drive will not work since this is not a shaft-driven car. It is solid axle where the front and rear each have their own motor bolted right beside the wheel:

http://rcrock.com/product/rc4wd/clod/rockbull/ridgeBlade/g1/2.jpg

I am wrestling with the idea of using a differential+ transmission or using solid axle motor mount. Lower COG for one with more space on the chassis for everything else, plus nicer parts. Or the transmission which has much higher COG, and less tighter turning radius, less space on the chassis since the transmission is there, but variable gear reduction.

 

[Hero999]

Then what's wrong with overvoltaging the motor (like you origionally said) it shouldn't do any harm providing you watch the curent and thereofre the power dissipation?

 

[dknguyen]

Then what's wrong with overvoltaging the motor (like you origionally said) it shouldn't do any harm providing you watch the curent and thereofre the power dissipation?

That's what I've been asking.

 

[Hero999]

And we've already answered your question, overvoltaging your motor won't do any harm proving you watch the power dissipation.

 

[dknguyen]

I know. At some point the thread started going backwards in time.

 

[Hero999]

So what did you decide to do in the end?

Keep the brushed motor or go for brushless?

 

[dknguyen]

Brushless outrunner.