Stepper induction

[marklewry]

Hi all,

I have recently built a drive board for a stepper motor. This is a unipolar drive and simply has 4 mosfets, 4 fast recovery diodes and 4 1 ohm 3watt resistors (2 in parallel = 0.5 ohm sense resistors). All was going well until I encountered a problem using a different motor. The current surges were very large and everything started getting very hot. The motor was not reacting very well either. The drive is controlled by a pwm chip L297 from ST. This has a comparator that looks for the voltage across the sense resistor and compares it to the preset voltage (Vref). When the voltages are equal the output turns off. I found that there is a small delay before it 'looks' ate the voltage. This can be changed by adjusting the value of the capacitor for the oscillator to the chip. It seems that if the delay is too long the current is allowed to build up too rapidly without control, so making this delay smaller helps.

Here is the question:
Why does the current build up in such a small period of time 2micro seconds when the theoretical build up should be much smaller L/R time constant. The motors inductance is 4.5mH, resistance is 1.5ohms and 3V. My chopper circuit uses a 40V supply. According to the L/R time in 5micro seconds it should only be 50mA? approx.

If anyone has any info on how to correctly work out he oscillator frequency and delay period for a particular type of motor I would appreciate the input.

Many Thanks
Mark Lewry

 

[melone]

Could the motor inductance have increased with the new motor? If so, more energy is stored in the coil when you turn it off, and therefore the voltage spike will be higher for the same amount of time.

Also, it sounds like something is a little screwy with the monitoring circuit. Using the comparator as a monitor should be fairly fast. How much of a delay are you seeing? Is the voltage being clamped at the same level (with respect to the other motor)? Could you be poluting your reference voltage with a sneak current path associated with your motor?

Don't forget, if you do not have a pull-down resistor on the gate of your FET, it will take a long time for the charge to evacuate from the device. Remember, the gate-to-source on FET's are basically capacitors. This might be the source of your delay.

Let us know what you find.

 

[jclough]

I don't know if this will help much but for what its worth we sell a stepper drive using the L297 and L298 chips. I remember having some involvement in the design aspects of the drive project. We refer customers to anothe rdrive maker now. Let me know if you're interested.

Some things to note:

The chopper frequency was set at 20Khz in a effort to make the drive quieter. This wasn't a performance descision.

Lower resistance motors would cause the drive to run hotter because the power semiconductors spent more time in the high resistance region between the on and off states. I remember that the L298 chip required a rather large heat sink to survive long.

The manufacturers of the L298 have a schematic for a bipolar stepper drive. I've built it, used it and found it pretty reliable. Just make sure the enable line keeps the drive off during power up. You can also run VR steppers in a unipolar mode. PM steppers blow the drive when hooked up unipolar. I never did find out why.

Hope this helps


Jon

 

[marklewry]

Thanks Jon and Melone,
Jon - I used the L298 with the L297 before I used this new drive. I kept blowing the L298 up! I now have the feeling that this is something to do with it.
Melone - this delay is there on purpose and is set by the L297 so that the comparator ignores any transisents while the device switches on. You set this delay by altering the capacitor in the RC network. I think my biggest problem at the moment is trying to understand why the current in the inductor builds up so quickly when the books say that it should not. I have seen that with my oscilloscope it builds up to 2 amps within 2 micro seconds. If you do the L/R time constant equation you get only 50mA for the same time?
I can't understand it!!!!!!

Thanks for your help guys.
Mark Lewry

 

[melone]

Don't forget about your line inductance! You may have more inductance than you think especially if you have long lines run to your motor. Although wire is thought of as a perfect connection, it actually has line impedance, inductance and capacitance associated with it.

Have you thought about putting a clamping / reciruclation diode on the motor? If you are only rotating it in one direction, the coil can be discharged with a diode.

Just a thought.

 

[artlewry]

Mark,
I'm a retired Ford and Chrysler electrical engineer, just wondering if and how we are related. There aren't many Lewrys out there.

 

[Rupesh]

We are having problems using two L298 in parallel to in case of higerr current demand. Eg 3.3 amps per phase. If any one can suggest a complete ciruit of two l298 and 297. plese do so.
We have tried connecting as show in the datasheet but to no advantage.

liladhar1thoke@rediffmail.com

 

[jbartos]

Suggestion: Some applications do not accept fast switching rises or rates caused by very fast switching devices.