Reduce Physical Size of Old Chopper Circuit

[Tom S]

I have an old, tried and true chopper circuit design that I wish to refit into a smaller package. It is a proven, robust design and I would obviously like the new to be just as reliable. It is a DSP micro-controlled current regulator for a large solenoid type load (coil). It is able to deliver up to 8 amps. The control circuit is not the issue; it is the power electronics components and the large heatsink that create the the problem with size reduction. Is there a better topology I should be considering? I can provide schematics of my current design if needed. The package limitations are about 5" x 5" x 1".

 

[davenn]

I can provide schematics of my current design if needed.

yup that and photos would probably help some one give some suggestions

 

[Tom S]

Here you go...

 

[KrisBlueNZ]

Nice! I would be interested to see the rest of the schematic.

Do you have a description of operation written up?

Have you looked into synchronous rectification to improve efficiency?

That's probably the only suggestion I can make.

 

[Tom S]

Here is a brief description of operation.

The first page is the the gate drive power supply, pretty straightforward. The second page is the chopper circuit. The load is connected to CON10, 6 & 7. U10 provides current feedback to the DSP, which sets the operating current (preprogrammed). Q11 is simply an on or off switch, turning it off disconnects the load. Q12 is the PWM switcher used to draw current through the load from the DC bus created by the AC signal at CON10, 1 & 3 (single phase, but 3-phase can also be used at higher current levels).

I have not looked at synchronous rectification, but I am skeptical using it would greatly reduce the size of the components.

If this is your only suggestion, it is still good feedback because it can bring me to the realization that this much power throughput is not possible in this size and move in a different direction altogether.

 

[KrisBlueNZ]

I don't understand what your circuit does because I don't have a general overview of its purpose and how it fits into the system.

But don't worry because I don't think I can suggest anything anyway.

I suggested synchronous rectification to increase efficiency so you might be able to use smaller heatsinks. But it looks like the actual rectification section is only producing four separately isolated low-voltage low-current rails anyway.

Please don't assume that just because I don't have any other suggestions, there are no major design changes you could make. My experience with switching supplies is only the conventional types, and you are obviously doing some extra stuff that's outside my experience and knowledge.

There are a few other people here who might be able to help. If they don't join in soon, I'll PM them and ask them.

 

[hevans1944]

I have an old, tried and true chopper circuit design that I wish to refit into a smaller package. It is a proven, robust design and I would obviously like the new to be just as reliable. It is a DSP micro-controlled current regulator for a large solenoid type load (coil). It is able to deliver up to 8 amps. The control circuit is not the issue; it is the power electronics components and the large heatsink that create the the problem with size reduction. Is there a better topology I should be considering? I can provide schematics of my current design if needed. The package limitations are about 5" x 5" x 1".

After looking at your photos and the partial schematic, I have to ask myself, "If it's not broken, why try to fix it?" This controller may be old, but as you said it is a "proven, robust design." I doubt you are going to improve on the IG-BJT chopper with more "modern" components that would decrease the dissipation and allow for a smaller heatsink. This appears to be a high-power circuit driving a highly inductive load that perhaps stores a lot of energy in its magnetic field that somehow ends up being dissipated in the chopper transistors.

I have an engineer son who works on such things, used to control a solenoid gate valve on a high-speed hot-glue dispenser, but he regards the circuitry as company proprietary and so we do not correspond about it. It would require an on-site visit and much more background information to even begin to re-engineer your solenoid driver.

Unfortunately, this is a hobbyist forum and what you need is an experienced professional engineer consultant who is willing to sign a non-disclosure agreement and work for a specified fee with some sort of guaranteed performance objective. In my younger days, I might have considered offering a "best effort" cost plus fixed-fee contract, but that doesn't fly well in a profit-oriented manufacturing environment... I doubt you will find much help here, but thanks for the visit. It is a real pleasure to see electronics built like the proverbial brick sh*t house that is still working years after the original designer went on to other things.