PWM Motor Controller/Amplifier

[AllAngles]

I'm working on an application where I need to intercept an existing PWM signal (~120Hz, 0-12.0V) - which is limited to 20A max current - and output an identical signal with a capacity of up to 60A. The signal controls a single brushless DC motor but my application will add another (identical) motor in parallel with the first (and possibly 2 more motors).

Based on my limited understanding (never worked with any kind of transistors) and what I've been able to gather researching online, I thought N-channel MOSFET's might be just the thing and I've come up with the attached circuit that I believe will meet my needs.

I'm considering using a pair of IXYS HiPerFET IXFE48N50Q in parallel to give plenty of overhead. Am I on the right track with the circuit design? If so, are the IXYS units a good fit (if a little expensive)?

Link to IXYS specs:

http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=IXFE48N50Q-ND

Appreciate any insight/feedback.

Mikah B.

http://www.allanglesdesign.com

 

[IRstuff]

You need to get a EE to do this.

TTFN

FAQ731-376
Chinese prisoner wins Nobel Peace Prize

 

[AllAngles]

We will turn it over to an EE if this direction looks promising as a solution to this need.

Is the concept sound or are we barking up the wrong tree? There are other more "brute force" simple solutions, but this seems like it could be elegant and cost-effective for our need.

Mikah B.

http://www.allanglesdesign.com

 

[IRstuff]

It's a lot of current to be running around. I thik have separate PWM drivers controlled by the same signal would be preferrable, particularly since you have a working example.

TTFN

FAQ731-376
Chinese prisoner wins Nobel Peace Prize

 

[AllAngles]

The module that currently generates the PWM signal also implements the control logic and is a CAN bus node - neither simple nor inexpensive to reproduce. We have verified through experimentation that the control logic is able to compensate (it's closed loop) for the additional motor. All we need is more amperage capacity to drive an additional motor or two.

You're right, it is a fair amount of current. The FET I spec'd is rated for 41A max continuous drain. I expect the system to consume about 30A during normal continuous operation which is why I chose a pair of FET's in parallel. My thought was that by running two FET's in saturation but well below their capacity, I could avoid excessive heat buildup. Demand could spike as high as 60A for short periods of time (4-8 seconds).

Mikah B.

http://www.allanglesdesign.com

 

[MacGyverS2000]

Will the loop closure include the other motors with the new design? If not, you'll be controlling one motor nicely and the other two, well...

Dan - Owner

http://www.Hi-TecDesigns.com

 

[AllAngles]

Yes, Dan - both motors will be included in the loop. I've added a 2nd motor to the diagram to indicate how it will fall in place. Physically, the flyback diode will be built in to the new device across the output connectors. But, both driven motors will essentially share the same signal. We've tried it with two motors pulling on the original (signal-generating) module and it worked fantastically, but I cannot in good conscience let that go because it is only a matter of time before the additional current draw burns up something expensive.

Mikah B.
www.allanglesdesign.com

 

[MacGyverS2000]

My forte is definitely not power (or motors), but placing motor coils in parallel and expecting them to work identically doesn't sound wise... one will invariably work harder than the other.

Dan - Owner

http://www.Hi-TecDesigns.com

 

[IRstuff]

Saturation is relative. Your datasheet shows curves that put Vds@30A = 3V. That'll be 90W dissipated. 2 in parallel would be 45W each. Still a bit, although the package could probably hack it, maybe...

Again, you should get an EE to do this.

TTFN

FAQ731-376
Chinese prisoner wins Nobel Peace Prize

 

[AllAngles]

Dan - no need for the motors to perform identically, in fact it would be impossible as they will probably never see identical loads - though it should be close. If the circuit was prone to one motor ALWAYS doing the heavier lifting, that could be a concern. My expectation, though, is that the circuit doesn't care much and the load balance will come down to the mechanical design.

IRstuff: Like I said, I will get an EE to look at it if the concept has merit. I just want to have my homework well and truly in order first.

Mikah B.

http://www.allanglesdesign.com