Stepper Motor/Mosfet Problems

The limiting resistor voltage drop should NOT have any effect on the the mosfet gate voltage as is happening with the cirquitry you describe. Each gate voltage should be applied referenced to respective mosfet's source, and MUST turn the mosfet fully ON, meaning Vgs about 10V and ALWAYS safely below the max mosfet gate voltage (~20V). If your gate driving cirquitry doesn't take care of all that, don't expect any reliability. And another thing. With the 2 Ohm resistor plus winding 1.8 Ohm, when one phase is driven by 24V on a static stepper, you feed it with more than 3 times the motor current rating (!). You propably have already ruined their magnetic system, critical for stepper produced torque, even if the windings are not fried yet. S.M.

You could "split" the signal of the gates to turn on extra MOSFETS to drive M2.

Or you could use kickback diodes to supress the tremendous spikes produced every time your MOSFETS switch off; you have two coils in parallel and your fets are only 100V. Warning: kickback diodes cause the remnant currents within the windings to drain a bit slowly, it may affect the response of your motors to high speed pulse trains.

Regards

1. Use H-bridge configuration to drive the motors (bi-polar). You'll get more torque at higher speeds.

2. IMHO you're asking for trouble trying to drive them in parallel. Separate them out. The drive transistors are cheap enough - or use H-bridge ICs (L298 is good for 2A/phase, up to 46V, and unit price is about £5. L297 driver is not much more; feed them with the same clock pulse and the motors'll stay in sync as if driven in parallel).

Unless this is a very cost-sensitive application, and is going into mass production, you'll save on developement cost/time by using proven circuitry.

Use quality chips with the major part of the logic included. There are plenty of chips around for up to 2 amps. They also handle over current and overheating in the correct manner with no damage.

All you need to supply is direction and frequency....

Look at the chips from Allegro, especially the SLA7078 and later, no external resistors needed for current control.....

Google it for fuller infos but start here:-

http://www.alldatasheet.com/datasheet-pdf/pdf/176062/ALLEGRO/SLA7078MPR.html

In Fig.3 you definitely mess up the gate drives of the MosFETs - Vgs is dynamically reduced by the voltage drop on R1.

Is it implied that in Fig.2 adding the 2 Ohm resistor prevents burn-ups? If so, have you tried a 1 or 0.5 Ohm to improve the torque?

That said, I agree that you need to use protected drivers instead of MosFETs, as has been already mentioned.