FEELER: High Current PSU PCBs

I'm working on a high current project here. Its essentially a constant current supply which will be fully adjustable from 160 mA - 54 Amps, at voltages from .5V up to 5.5V.

It can be configured for driving purely resistive loads, delivering up to 250W through a .1 Ohm resistor, or can be configured for driving power diodes, high power LEDs and laser diodes. There IS a soft start current ramp on the output.

Multiple PSUs could be connected in parallel for even higher currents with maximum values appearing in multiples of 50A.

The design has coarse and fine current adjustment, multi - turn pots.

I would provide you a list of part names and numbers you can find from mouser, digikey and allied to completely populate the board, and install it into a 1" thick 10" X 6" enclosure, fitting the circuit, a heatsink and active cooling.

Mounting the parts into the aluminum enclosure would be your responsibility, and cutting vent holes in the enclosure would be as well.

This is a high power supply, so you would need to be comfortable soldering 50A connections on 10 gauge wire, and capable of safely building and testing the thing for yourself.

The circuit's input power comes from a 12V, 8.5A laptop supply just for some perspective. A typical atx PSU modified for stand alone use can power the circuit as well.

Expect to spend around $150 - $200 on the parts to get it running and installed in the enclosure. Cost also depends on if you decide to go with an atx psu or the laptop psu I've found. Two of the main components besides the PCB account for around $130 of the cost alone, so a lot of the remaining parts are discrete components.

I'm feeling out if there is a demand for high current PSUs. Anyone out there want to experiment with 40W laser diode bars?

The feeler is essentially for the PCB and rights to the official parts list.

Some prototype boards have been ordered, and the first PCB based version will be assembled and tested when the rest of the parts come in.

If there is interest, I can put together a "user manual / datasheet" for the circuit.

If it seems there is demand, but I'm going about this entirely wrong, I would consider going open source.

Its a high current PSU which can't otherwise be found or simply purchased for any less than $900 from the industry, so let me know what you guys think! I've put a hell of a lot of work into it, and I want to be able to share it if I can see it would be of some use.

Best Regards,

Tyler

I have an F-mount 10W and a 20W bar that I have been looking for options to power for a while. F mount will be used to pump a home made green so adjustable power is really nice. can the board be modified to allow a 6V breakout to power a circulation fan for the enclosure? My main goal is to incorporate the power supply and water circulation pump/radiator fan /TEC all of of one power plug. So basically a self contained PSU + cooling in a box.

The thing runs off a 12V psu, so if you stick to an ATX PSU, there should be plenty of current for the supply, the TEC, a fan and a pump.

You could certainly use an LM317 set for 6V to get the 6V you need.

I would love the curcuit and part numbers to try to build one. Only thi g slowing me down from my dpss build is powering the pump above 3A

What the most current you think you'll need?

There is a current limit that can be set to limit the max adjustable current you can set the dial to.

My F-mount requires 9-10A and my bar requires 20A. All at 2V.

Ok. I'll send you some options via PM. Lots of details depending on what you want to do. It doesn't seem like anyone else needs a high current supply

I may have some interest in this as well. A nice adjustable High Current PSU would be great for some of the projects i'v done. My 40W 808nm comes to mind for sure!

Perhaps you would be willing to send me some options via PM as well.:beer:

Ok.

Id love to see this open source!

I don't have any particular use(right now), but I would like to see the schematic. Sounds like an interesting circuit.

^ me too. could come in handy in the future.

I'm in! How much will the PCB cost me?

I'll just keep posting updates here as they come.

If there is interest, I'm hoping I can get it down to $30/board. The copper weight is set at 3 oz, so there would be some extra cost. But to be honest, the quotes are coming back for $60 ea. ATM - if we ordered 5. Total job price does not usually increase much (if at all) with more boards ordered, so price would totally depend on how many we have to order.

On board components are adding up to $80 (The main IC is $52) plus cost of panel mounted switches, two potentiometers, knobs, panel jacks for input/out, and whatever your 12V, 8A PSU is. (ATX recommended) The on-board components are what first picked, and then made custom footprints for in Eagle, so it would be recommended to stick with those parts.

Is there any need for anyone to drive purely resistive loads? Because overall cost of parts per unit can be greatly reduced if there is no need for a series power diode to drop the minimum output voltage for the driver to regulate.

This would mean it would be limited to power diodes, LEDs, LDs as the load.

These can also be handy for driving high power shunts and heating elements. Is there any need for this at all? Speak now or it will be removed from the design!

Prototype PCBs have been ordered just today. There was a 10 day turn around time for the job, so between now and then the design can be "tweaked" to the popular set of needs. The prototypes will be tested as soon as the boards arrive.

Any need for modulation (1kHz max)? Extra limit switches? Knobs? Status indicators? Readouts?

Adding features can be as easy as adding a few extra pads here and there for a current readout, or an extra limit switch, or an RCA jack for modulation. I would assume panel mounted control features (switches, pots, meters) to save money on board surface area.

Any more votes for open source? I'm open to it. This has been so much work though.

I think we'll need 5-6 to continue with this. 10 would be ideal.

Speedy78, if you'd prefer someone else builds it for you, I could put the circuit together for you. Enclosures would be your option though - I can make suggestions for them, but you'll end up going with whatever you want. Its a surprisingly small footprint though.

All comments, suggestions and questions are welcome.

Yeah. I would like the ability to wire up a readout for output. If you can buuld it ill throw a little your way for your effort. Either way this is way cheaper than my alternative to drive these. Also that two way limit switch for two option if able. Not sure how easy that is to add in but woyld be nice to have a switch for 10A or 20A max so I dont pop a 200$ diode

Definitely, a readout would be super handy.

I'd love to have it Open Source as well

I think you can keep the series diode in the circuit. Just add an option for a jumper so if it is not needed, you can simply add a link on the PCB.

If you don't mind, can you share what controller IC you used?