Can you solve this puzzle?

[Meatball]

This is an ideal setup:

I have a 'friend' that wants to power one 9 mm Nichia at 2.6A with a flexmodP3.

Nice. Right?

Vsupply: 12VDC ATX PSU <= 15A
Mod Input: 3.3V GPIO on microcontroller (GND tied to ATX PSU GND)

My friend has setup many flexmods before, so he goes about his usual business, dials in a bias at 200mA, and a gain for 2.6A @ 3.3V on MOD. This essentially maxes out the gain ----> (2.6A/3.3V) ~= (4A/5V)

With the driver dialed in, he uses thermal epoxy to secure the thing to a large aluminum heatsink, and proceeds to solder up the LD after shorting the output momentarily.

Textbook. Right?

So after all this, no current. No red LED, no bias current, nothing.

My friend connects another P3 to the same PSU and LD, works like a charm. He then tries the original unit again, but nothing is happening.

He then reviews the board level for cold solder joints, shorts and retests conductivity through all his wires and joints.

LED lights up when he floats the interlock as it should, but closing the interlock doesn't enable current flow.

What is the problem?

My friend is feeling down about this, because it is delaying his group project laser engraver. Should he consider swapping boards? Attaching the working P3 board to the FET of the lifeless one?

I can't solve this one because I have yet to read of any problems without the LED turning on.

Anyone else experience this?

[planters]

Not sure what you mean by shorting the LD and soldering, but I am guessing this is to protect it during the physical process of attaching the wires. For what it is worth, I have soldered over 100 LDs at my ESD grounded station without doing this and without a single failure.

In any case this seems to indicate a bad flexmod and I would let it go at that. Replacing the P3 only as a component in a system that works fine with its clone is strong reason to suspect the original and I would not waste the time to try to diagnose the problem. Just replace it.

[Meatball]

I was afraid of that.

Aren't these things hiding under rocks these days?

I wonder if anyone is keeping a stockpile somewhere?

[planters]

If he was able to run this with another P3 then he should be good for now. I think Dave at LSP and Lasertack both sell good drivers and the 2A limit can be doubled to 4A by running two channels in parallel. If he is willing to spend a couple of hundred dollars, Wavelength Electronics has some really nice drivers that go up to 12A.

[dnar]

I have found the connection pads often require a scrape before soldering. I have had what look like good connections fail or go intermittent with these drivers. I have a stock pile, but am holding on to them as they are hard to source when you do need them...

[Draco]

sorry for off topic but what are the flex mods? i have seen them but have never played with them? I have a power supply that has a current and voltage limit when i want to test a high current diode, like my class 4 infared

[Lazerjock]

sorry for off topic but what are the flex mods?

Here you go.
https://innolasers.com/shop/index.ph...roller=product

[planters]

Mixedgas will probably be able to give you a more detailed explanation, but other than the very best (expensive) power supplies from say HP, current limiting and voltage limiting is not tight enough with most power supplies to prevent brief transients that can destroy diodes. As a general rule, you always want to use a driver designed to drive diodes as your power source. The flexmods in my mind have one main advantage, 4 amps.