chopping motor voltage 70% via pwm, safe or not?

rwgast_logicdesign

So basically I just want to PWM an h-bridge connected to a 7.2v motor using a battery the supplys about 22v at peak.

Now I realize there have been a bunch of posts about this, or similar topics. Most of the posts about this are a bit more sane though, maybe a 12v battery pulsing a 7 volt motor or something like that. The real issue is all these threads have conflicting opinions.

Some say it is just fine, others say since the peak to peak voltage is higher than the nominal voltage of the motor the brushes will arc. Right now im using an 11 volt battery and just not pushing the duty cycle over 72% (the extra 7% is to make up for voltage drop across the bridge). So if i move to 22v and stay around 35% max am I safe?

Ive also read that people get huge increases in torque at hi voltages with low duty cycle is this true? And finally will pishing the carrier freq to hi burn the motor up since that means it would be seeing peak voltage more often? I would like to shoot for a freq of 20khz plus but logically this seems like a bad idea. I do plan on adding current feedback at some point too which would add another layer of saftey, and allow the max duty cycle to be calculated on the fly

stamptrol

There is, of course, some voltage at which the commutator will flash over. Every DC motor with a comm will do that.

Your logic with the 11 volt source makes sense. But, you'd probably be OK at 12v and 100% duty as well. If the 22 volt source doesn't flash across the comm, your 22 volt calculation is probably OK as well.

From my experience, 20 KHz is way too high. Its not buying you anything other than you won't be able to hear the motor responding to the PWM. The motor is made to work at DC so a lower frequency (1 KHz) is closer to what it expects. Also much easier on the H-Bridge transistors because they spend less time in transition and more time fully conducting or fully off.

From a practical point, its a 7.2 volt motor, its not very expensive and the consequences of failure during testing are probably not too high. Sometimes you have to break a few eggs to make an omelet.

Cheers,

LoopyByteloose

PWM doesn't really reduce voltage, the insulation and other voltage sensitive components still see 22 volt peaks, just for less sustained time.

The power is less and the motor doesn't run as hot, because Power is averaged over the duty-cycle.

If you actually want to reduce voltage, you have to use a transformer and either pulse DC or AC through the windings.

shimniok

This thread has some good info on considerations of frequencies, tradeoffs, etc.

http://www.electro-tech-online.com/robotics-mechatronics/19246-suitable-pwm-frequency-motor-control.html

It has to do with power dissipation, inductance, etc. HTH!

RickB

Fundamentally, the current through the motor needs to reach its rated value by the time the pulse ends. Otherwise you will not reach the rated torque. Higher voltage gets you there faster. The higher the motor inductance, the higher the voltage must be. 3 volt steppers can be run at 40 to 50 volts, It's the current that matters. Most likely you will need to be in the very low KHz range to get the torque you need in the average small brush type motor. By measuring the motor inductance, looking at the current pulse through a sense resistor with a scope and changing the frequency/pulse width, you can find the minimum pulse width and maximum freq for your motor at a given voltage.

LoopyByteloose

Steppers don't have brushes, so it is rather easy to run them at over-voltages if they don't get hot enough to burn off the insulation from the copper wire.

On the other hand, small brushed motor have to be managed more carefully, as the brushes may be the weakest link in the wiring, not the copper enamelled wire (Copper enamelled wire will often handle 100 volts if it does not get hot, even if it is in a 6 volt motor)

It was mentioned that over-arc might be caused by a higher voltage jumping across the tiny rotor. That is a physical possiblity. Another hazard is that we tend to use cheap DC motor intended for rotation in one specific direction and actually alternate the rotation. This really is hurting the brushes, especially if more voltage is applied.

I guess what I am trying to say is that with brushed motors, much depends on buying good quality motors. Expecting excellent performance outside the specs from a so-so motor is just never going to happen.

But if you have a 24VDC brushed motor that is properly rated for the directions of spin you want, chopping the voltage 79% may do nothing harmful.

But the truth is that it is always better to run DC motor toward the high end of their rating as the H-bridge and control logic waste less of electricity provided. Trying to make 6VDC brushed motor think it is seeing less than the 22Volts may just be an experiment in destructive testing.

RickB

A brushed motor is still a current operated device. If the pwm pulse ends when the rated operating current is reached, then the motor is not being operated over its ratings. This assumes that reasonable voltages are being used. 24 volts for a 6 volt motor is not unreasonable.

al1970

Hi:

Make sure your bridge can take over 22 volts when being plused!

I have found that about 200 Hz is in a good range for the freq. to use in a small motor.
DO NOT use 20k Hz as many places tell you. You will have just about no torque.
This is a good place where you can play and see what happens for yourself.

"Some say it is just fine, others say since the peak to peak voltage is higher than the nominal voltage of the motor the brushes will arc. Right now im using an 11 volt battery and just not pushing the duty cycle over 72% (the extra 7% is to make up for voltage drop across the bridge). So if i move to 22v and stay around 35% max am I safe?"

Maybe yes, maybe no. Most likely it will be fine. Make sure you add the 3 caps to your motor before you start. One across the leads and each lead to motor metal case. Try with 7.2 volts 1st. See what the arc look like if any. Listen to the way the motor sounds. Also smell the motor for O3 gas. You get O3 when things arc. Then do the same with your setup to see if the motor acts the same.

Al

rwgast_logicdesign

Well I have read the thread shimniok posted, it was fairly informative and it got me to testing. I have also found the 200hz range is pretty good, it doesnt have an annoying squeal and the wheels run pretty smooth, not stutter. The motor does have much more torque at 10% duty when im running a 200hz carrier compared to an 8000hz carrier. Although the higher frequency carrier signal run much smoother, I think the lesson here may be to adjust the frequency on the fly as part of the motor feedback loop. I have a 32khz set up I will test when I have the time, but its on Arduino, so far I havent found any code canned motor control code that can do very hi frequency PWM on the propeller. Since I still have yet to learn PASM, the code I wrote for Arduino is about the only way I can do hi frequency PWM for test.

I have been thinking about using a TI tl5940 chip to do my PWM it delivers a 12 bit resolution, its meant for LEDs but ive seen it used with motors. I just cant figure out what what the carrier frequency of this chip is, from reading the data sheet I think it is set by one of the two clock lines and can reach up to 30mhz. Maybe someone could help me out and take a quick look at the sheet for me

http://www.ti.com/lit/ds/symlink/tlc5940.pdf

There is a driver in obex, but it looks as if the PWM frequency is done via the chips set up method. Im not sure if you could can actually dynamically change the chips carrier on the fly. Im acually pretty happy with the 200HZ pwm im running at the moment, but im about to build my motor control stuff and I really wonder if this chip is the way to go, as I can control 16 pwm lines with two cogs, if I can easily adjust the carrier from 60hz to 50Khz im definitely going to use it.

As far as the 22v thing, I do have 3 .1 filter caps connected as you described . Is there any other way to detect arc's other than an O3 sniffer, that is something I do not have at my disposal?

You guys keep referring to the quality of the motors im using, well they are just the sting ray motors. The info found here isnt all to detailed

http://www.parallax.com/Store/Accessories/MotorServoControllers/tabid/160/List/0/ProductID/587/Default.aspx?SortField=ProductName%2cProductName

Im not sure how quality the motors are but I have popped open the gearhead and found it has metal gearing opposed to the nylon gears shown in the PDFs on Parallax's product page.

LoopyByteloose

I think the Stingray motors are good quality. I was more concerned with you trying to exploit some motors that were salvaged from some oddball gadget. I have quite a few motors take from DVD and CD drives, and they definetly have the wrong kind of brushes for reversing direction. And other motors I have are very cheap 3v to 6v hobby motors that don't hold up well.

Metal gearing will last longer than nylon. So that is a good thing.