l297 7amp unipolar driver

[ljd10]

Is it possible to make the 7 amp unipolar board and run it with a 65 volt power supply at a full 7amps without any modification to the now basic design?

[NC Cams]

It is not uncommon to use low voltage circuits to drive much, much higher ones. Most mosfets are driven with either logic levels (5-10v Vgs) or conventional vets (10-18v Vgs). The trick is to use the proper pass elements (fets or transistors) and, with fets, the proper fet driver IC"s.

Fets are rated at various voltages. Check the manufacturers ratings but the IRFZ44 or IRLZ44's are rated for something like 65 volts, perhaps even 100. These readliy switch 7 amps with 10-18v or 5-10v gate drives respectively. If you're going to drive something at 65v, you'd be better off finding fets that are rated at 100v.

You may have to regulate the 65v down to 15 or lower to operate the control ciruit. There are not too many IC's the run at 65 volts. It is better to regulate the voltage in multiple stages (IE: 65 to 30 or so and then 30 to 15 than from 65 to 15 - much less power dissipation and subsequently heat.)

THen again, if you use a regulated 12v wall bug (or whatever you want/need to do the control voltage) and tie the grounds of the 12 v and 65 together, you don't have to make a regulator circuit which is much simpler. Just make sure the wall bug and source of the power for the 65v are the same souce - again, makes things simpler.

[kreutz]

Originally Posted by ljd10 Is it possible to make the 7 amp unipolar board and run it with a 65 volt power supply at a full 7amps without any modification to the now basic design?

Unipolar drives stress the Vds of the switching Mosfets during the interval between the turn-off time of the mosfet and the reverse recovery time of the anti-parallel diode.

Mosfets must be rated Vds >> Vmotor (minimum 2 times) and must have a high "Repetitive Avalanche Energy" because the reverse avalanche voltage will become a clamping voltage for this high voltage transient. Higher voltage mosfets have higher die size and higher "on" resistance, so losses are higher.

Bigger die mosfets will have higher equivalent "gate total charge" requiring higher gate currents to provide short switching times.

Parasitic inductances and capacitances in the gate-source circuit (including current sense resistors) have to be controlled to a minimum on the PCB design, otherwise ringing transients will induce a lot of EMI.

[ljd10]

Please excuse my ignorance but are you saying thats its possible to operate the control circuit at a lessor voltage (5-15v) and power the steppers at 65v?

[kreutz]

Originally Posted by ljd10 Please excuse my ignorance but are you saying thats its possible to operate the control circuit at a lessor voltage (5-15v) and power the steppers at 65v?

I would like to see the basic schematics you are referring to. It all depends on the components and drive design. Basically you need at least two different voltages besides Vmotor, at least 5Volts dc for the logic Ics and at least 12 Volts dc(or 15 volts) for the mosfet drivers. Then you will have to replace the current sense resistors too.

[ljd10]

The particular board I'm referring can be found at http://pminmo.com/.
It is the 7 amp unipolar drive found on the design page. basically I want retrovert an existing machine with 7amp motors and 65 volt power supply and an extra 5-15 volt power supply to power the control or logic board and the stepper motors.

[NC Cams]

"...are you saying thats its possible to operate the control circuit at a lessor voltage (5-15v) and power the steppers at 65v?..." In a word, yes. It is done all the time PROVIDING you follow the design guidlines as partially outlined in post #3.

You might want to do some research in/at the International Rectifier site. THey hava a lot of tech manuals on how to design mosfet switching circuits.

You could also drop a PM to Phil at PMINMO as he's a regular poster/contribuitn member here and posts quite often. Perhaps he'll give you some insight into how best to use/modify his board.

[kreutz]

Originally Posted by ljd10 The particular board I'm referring can be found at http://pminmo.com/. It is the 7 amp unipolar drive found on the design page. basically I want retrovert an existing machine with 7amp motors and 65 volt power supply and an extra 5-15 volt power supply to power the control or logic board and the stepper motors.

I don't like to put the negative note, but you will need to change the gate drive circuit to include mosfet driver Ics, the logic gates will not provide enough current to keep short the mosfet switching times at higher voltage due to the additional gate capacitance added by the Miller effect. You will also need to use mosfets with more than 130 Volts Vds max.

[riko]

Is there anyway to clamp the inductance spike casued by imperfect magnetic coupling of the coils?

Possible ways are to use zeners and snubber clamps.

I recall this is similar to a flyback power supply circuit. See page 9-11 of
http://focus.ti.com/lit/ml/slup112/slup112.pdf which uses a dissipative RC clamp.



But I was looking for a way to recirculate the energy back to the power supply via a lossless type of snubber.

Possible routes are a passive component design based on a small transformer.

Or building a reverse switcher type of power supply to take the high voltage spike down to the power supply level.

Any power supply gurus out there.

[kreutz]

Originally Posted by riko Is there anyway to clamp the inductance spike casued by imperfect magnetic coupling of the coils? Possible ways are to use zeners and snubber clamps. I recall this is similar to a flyback power supply circuit. See page 9-11 of http://focus.ti.com/lit/ml/slup112/slup112.pdf which uses a dissipative RC clamp. But I was looking for a way to recirculate the energy back to the power supply via a lossless type of snubber. Possible routes are a passive component design based on a small transformer. Or building a reverse switcher type of power supply to take the high voltage spike down to the power supply level. Any power supply gurus out there.

There are a few circuits usable for that task, but nothing as simple as dv/dt controlled by gate drive resistor value and the use if the intrinsic antiparallel diode of the power mosfet itself as a clamp. i.e, no extra components.

A unipolar stepper drive is supposed to be simple, the use of dissipative or not dissipative snubber will make the circuit highly motor and power supply voltage specific to be of any general use. I already traveled that road when designing the Mardus-Kreutz boards.

The magnetic field stored energy is automatically delivered back to the power supply after the diode recovery time interval, is only during the time interval between mosfet switching off to the end of the recovery time that needs to be controlled (approx 200nS-300 nS depending on the diode characteristics).

[riko]

Isn't the energy in the spike basically from the leakage inductance (imperfect magnetic coupling) from coil to coil in the unipolar motor.

This energy is avalanched by the Mosfet.

I was thinking of some type of passive transformer clamp back to the power supply.

[kreutz]

Originally Posted by riko Isn't the energy in the spike basically from the leakage inductance (imperfect magnetic coupling) from coil to coil in the unipolar motor. This energy is avalanched by the Mosfet. I was thinking of some type of passive transformer clamp back to the power supply.

Hello;

The energy on the spike is due to the sudden interruption of the effective(including the leakage inductance) coil inductance's current flow during turn off. Adding more complexity to the unipolar drive is going to eliminate its main advantage: cost.

If the cost of adding some energy recovery is going to be approximately equal or higher than adding 4 extra mosfets, it is more cost effective to design a bipolar drive. If you are doing it because of the learning experience, then it is totally justified.

The clamping properties of the internal Mosfet's diode can be used as an advantage, by helping to get a faster decay (behaving as a zener diode connected to the drain). The repetitive clamping energy has to be controlled by limiting the Power supply voltage under 1/2 Vclamp, and choosing rugged mosfets with higher repetitive pulse clamping energy.