DC MOTOR CONTROL (H-Bridge)

[WIRENUT]

Hi there,

I am using a basic H-Bridge to control a DC motor. I have attached the schematic of the circuit. I have been having trouble trying to control a pretty large DC motor. The specs are 24V 14A 200W. When I put a small DC motor to it I have no problems. The H-Bridge switches the polarity perfectly and the motor runs. The POT also make the speed adjustable. However, when ever I try to use this larger motor, the motor kind of ticks like a clock. It barely turns and seems like there is not enough current getting supplied. I measured about 0.2A going to this particular motor. But when I put a smaller motor I can successfully draw about 1.5A.

I am thinking that this has something to do with the induction load that is seen across large DC motors. I am thinkinking of putting a "kick back diode" across the output of the H-Bridge. Kind of like the same way you would put it across the coil of a relay. The only thing is that I am finding circuits online that put a Diode across the drain and source of each transistor. But my circuit uses two different mosfets. The IRF4905 and the IRFZ44.

Does anyone have any input on how to prevent kick back that applies to my particular circuit. Also does anyone think the problem might be something else?

Thanks
WIRENUT   

[johansen]

The body diode of the mosfets returns the motor's current to the dc supply when the fets turn off.. this is why the dc bus capacitor needs to be able to handle the same ripple current as does flow in the motor.

IT may help to put an RC snubber in series with each fet, or across the motor terminals as close to the mosfets as possible. use a resistor of about the same impedance as the motor.. so if you're switching 1 amp at 12 volts then use a 10 ohm resistor. increase the capacitor until your problems go away.. perhaps 10nF would be a good start, it really depends on the switching frequency and how much leakage inductance you have to deal with. the capacitor will be charged and discharged as fast as the pwm frequency and so you can easily calculate how much power the resistor needs to dissipate.

even if you have a good capacitor for the dc bus it really helps out to install a 1uF film cap locally across each mosfet pair.

[nuno]

You can, but usually don't need RC snubbers on controllers of this power. And the snubbers tend to dissipate quite some energy. The problems I see on the circuit are

1) lack of "DC Link" capacitor(s); for that kind of power it should have >500uF (for at least a few KHz bridge switching frequency) as close as possible to the FETs (suggestion: 2 x 330uF 50V+)

2) no current limitation; you'll end up blowing the bridge in some applications, such as electric mobility, in which the motor can be "forced" or stalled

If your motor has a low inductance and the bridge's switching frequency is comparativelly too low, you'll have problems driving the motor too, with high heating of both motor and controller and low torque, as the current through the motor will not build up to an almost constant value and instead will be spiky. For a typical 24V 14A motor I would want a bridge switching at more than 6KHz.


Warning: The FETs in this circuit usually fail shorted, so you should use a fuse on the supply.

[max_torque]

Do you have an oscilloscope?  For motor control, you'll spend a LOT of time guessing if you haven't got one!



There are a million possible reasons for your Bridge not working with the larger motor, and in fact, it probably doesn't work that well with the small one!  As mentioned, for typical small motors the series resistance can by high, and dominate the series impedance, but once you get up into big motors (or more acurately, low turns, high inductance ones) then the impedance dominates.  At this point, extremely high currents can flow in very short time periods, causing all kinds of issues.

We could all guess why your circuit isn't working for the next 20 years and still not be correct!


Sorry, if this sounds "negative" but i am always amazed by how many people start messing with motor controllers without basic knowledge and the necessary tools to be able to investigate the performance of their device.  I think because you can do electronics on the bench, people think it is easy.  I mean, no one would try to make say and internal combustion engine on their bench (well, only a few people) and yet lots of people try to make an electric one.......)

[jlmoon]

I would think with a motor of this magnitude you would need to PWM (pulse width modulate)  the drive of the H-Array and drive the pwm with the voltages derived from your op-amp circuitry.  This would provide protection in-case of stalling not to mention better low rpm control.