Driver issues

Dibs · 2018-05-07T15:02:43.000Z

HI,

I’ve built a driver following the thread at 17mm/22mm MTN-MAXlp - Low-Profile 1A-4A Programmable Buck Driver + 22mm MTN-MAX (4-5s) but wiring up a 12v supply to the outer ring (0v) and 12v to where the spring would be, I’ve checked with a multimeter across the LED +/- pads but I’m getting 12v, I was expecting far less.

I’ve checked with a microscope and nothing is bridged (that I can see) and things look soldered properly on the pads.

Any advice\etc.?

Thanks

Dibs

Phlogiston · 2018-05-07T17:12:28.000Z

Do I understand correctly that you haven’t actually connected the LED?

I don’t see a schematic on that thread, but I looked at the MAX16820 data sheet:

https://datasheets.maximintegrated.com/en/ds/MAX16819-MAX16820.pdf

Looking at the typical operating circuit on the first page, not having an LED in circuit means that no current can flow through that branch of the circuit, so I’d expect to see the full supply voltage across the gap where the LED should be.

On that basis, you’d need to connect the LED to make the circuit behave as expected. Make sure your test setup can sink enough heat from the LED to stop it frying itself.

If you don’t want to risk the LED in testing, you could try a resistor sized to drop the correct voltage at the desired current. Bear in mind that the resistor will need to be rated for the expected output power. It will also get very hot at any substantial current.

Dibs · 2018-05-07T17:18:50.000Z

You are right - I haven’t connected the LED yet.

I thought that the pads would show a voltage less than 12v and wanted to check it before I fried a Cree XML2.

Dibs · 2018-05-07T19:40:32.000Z

A quick calc suggested I needed a resistance of around 1 Ohms and rated at 10W to drop 3V @ 3.4A. I don’t have anything to hand for that but I do have some automotive tail light bulbs and measuring one it had a resistance of 0.8Ohm and is rated around 20W, so tried that.

The measured voltage at the pads with the bulb lit was around 3.5V and switching the power off & on repeatedly did step thru the brightness levels - I flashed it with DrJones MiniDrv with 3 equally spaced brightness levels.

I’ve got some aluminium round bar offcuts ranging in diameter from 15mm to 25mm so with the mcpcb (with some thermal paste underneath) sat on the end of one - it’s alive. Tried a TIR orgthog lense and got the required beam pattern on the ceiling.

Well chuffed. Now on with machining the host.

Greatly appreciate the assistance Phlogiston.

Thanks

Dibs