Can someone identify this component/help me repair this laser power supply.

[AlessandroAU]

Hi all, I bought a DPPS laser module from ebay quite awhile ago. However it was DOA and the seller was most unhelpful (I didn't actually get a chance to power it up until months after the sale)

I did some probing with a DMM around the circuit and managed to have it power up when I probed the solder pads of the component listed as U1 so I figured it was merely a faulting solder joint. However I managed to foolishly short the component whilst probing around it and promptly exploded.

I was wondering if anyone else had used these modules or similar and could help me identify the component so that I could replace it. Unfortunately when the component failed it blew off the entire front and any indication of a part number. Looks like a TO-220 but with 4 pins?

[ConKbot]

Hmm. Looks like a basic non-PFC power supply, however that looks like the only power device on the primary side.  the 'U1' designation makes me suspect some sort of all-in one IC smps controller/power transistor.  The only other thing that is on the primary thats a semiconductor other than the diodes and bridge rectifier.   I havent seen an all in one mains switcher before, hopefully its not some oddball chinese brand IC, you'd really be up a creek 

[Nerull]

It looks like its connected between the transformer and one of the big caps - is that the case? Also isolation slots between the pins on one side. Could it be a bridge rectifier? I've never seen one in that package, though. Looks like there is one in a more traditional package next to the smaller transformer on the right.

[Nerull]

So the package looks to TO-220F-4L and the only device that seems to exist in that packaging as far as google can tell is this: http://dalincom.ru/datasheet/KA1L0380R.pdf

[xygor]

U1 could be a MOSFET with a built-in current sense resistor.  I can't recall any examples at the moment.
Edit: That fairchild integrated controller posted by Nerull looks like an even better candidate.

[AlessandroAU]

This what the top copper looks like, the pins seems to match that fairchild device?, I don't know enough to confirm for sure though

[xygor]

The pinout looks right.  The voltage and current rating looks right.  If you can carefully remove the epoxy of the part, I think there are probably two die: one is the FET and the other, the controller.  That would add further confirmation.

[AlessandroAU]

The pinout looks right.  The voltage and current rating looks right.  If you can carefully remove the epoxy of the part, I think there are probably two die: one is the FET and the other, the controller.  That would add further confirmation.

Best I could do, looks like there are two 2 dies on the copper backing?

[xygor]

What's the line voltage you are using?
What is the voltage rating of "Main Cap"?

If line = 120 vac rms, Vdc = 120*1.414 = 170V
If line = 240 vac, Vdc = 340V

Is the bridge good?  If so, with U1 removed, power up and measure the voltage of the main cap.  Or measure the voltage at pins 1 & 2 of U1.

If that's all good, you can try the fairchild part if you can find one.  I don't know what frequency to get though.  Can you measure the primary and secondary inductances of the transformer?

What is the voltage and current of the laser module?

I like your photos.  What are you using?

[AlessandroAU]

I'm on 240ac, I get 340v ish at the main power cap. Cap is rated for 400v. So the bridge seems good.

I looked on ebay and I've found a few of the fairchild parts. No idea how to determine the operating frequency. I don't have an inductance meter.

What would be the consequences of running it under/over the operating frequency? The laser is just some generic module, claimed to be 250mw, but who knows what it really is. It has a fan and a TEC however so not very easy to drive without knowing any operating parameters.

As for photos, nothing special at all, just the camera on my note 3 smartphone.

[xygor]

It might be possible to patch in something like a TOP225YN.
http://www.digikey.com/product-detail/en/TOP225YN/596-1161-5-ND

Too high or too low frequency might work not quite right, or it could smoke again.

To get the frequency, it might be possible to estimate it from the core volume.  We need to know more.
I am guessing it is a flyback convertor.

Can you find out how much power it is supposed to supply?
The transformer core material?
Core dimensions?
The values if the snubber circuit components? (an R in parallel with a C,  then in series with a diode;  that circuit is in parallel with the primary of the transformer; the diode points toward the + of the main capacitor;  if the snubber is not that topology, then state what it is.)

The more info the better.

[AlessandroAU]

Hey, I'd just like to say thanks for all your help so far I don't have my multimeter on me right now, I'll confirm tomorrow about the values. Here's some more information

The core has EI28 stamped on it, there is a EI28 series of transformers https://www.sumida.com/products/pdf/EI-28.pdf which are good to 300kHz apparently?

The low side of the transformer feeds into two schottky diodes in a -VE-0-VE+ configuration.  http://www.datasheetcatalog.com/datasheets_pdf/S/T/P/S/STPS2045CT.shtml

the only other interesting components on the low side are a lm324n op amp and some power transistors

http://www.alldatasheet.com/datasheet-pdf/pdf/170798/STMICROELECTRONICS/D882.html
http://www.elite-ent.com.hk/php/pdf/To-220/D880(TO-220).pdf

[xygor]

You're welcome.  I've been in your shoes and I know how it feels.

EI-28 is the core size.  That is good to know.  The operating frequency is going to depend on, among other things, the core material.  The sumida does not state what the material is.  Even then, just because a material is rated to 300 kHz, does not necessarily mean that they chose to run it there.

The next most important thing is the total load power.  Is the 250 mW the beam power or the total power?  The TEC is probably going to suck down a lot.

Is there any shim or gap visible between the E part of the core and the I part?

The others in this thread should feel free to jump in too, but I guess they went to bed.

If you have access to an x-ray machine, it would make it easier to see the gap if it is in the center leg.

The secondary side stuff isn't too useful unless it gives some clue about the voltage and current output.  Really what we're after is the power the transformer needs to pass.

Edit:
What you said "The low side of the transformer feeds into two schottky diodes in a -VE-0-VE+ configuration." I'm not too clear on.  Can you clarify?  I am starting to think this is much more than 50 W and that it might therefore be a forward converter.  Can you identify any ratings or values on the secondary side inductors?

Edit:
There are some other parts it could be, that while still obsolete, seem to be more available.
http://www.fairchildsemi.com/ds/KA/KA5H0280R.pdf
http://www.fairchildsemi.com/ds/KA/KA5H0365R.pdf
And here are some app notes on power supply design using the parts:
http://www.fairchildsemi.com/an/AN/AN-4141.pdf
http://www.fairchildsemi.com/an/AN/AN-4140.pdf
http://www.fairchildsemi.com/an/AN/AN-4137.pdf

It could still be a flyback with that diode configuration.  It just threw me off for a moment.

Edit:
More stuff on DPSS laser repair:
http://repairfaq.ece.drexel.edu/sam/laserscl.htm

[AlessandroAU]

Hey sorry I went to bed. What I meant was that the secondary side has a split rail configuration, I assume to generate a negative voltage to power the op amp. I can't see a shim in the transformer core, looks like it's all ferrite.

I don't think the power consumption is would be that great. 250mw of green light would need about 2w of pump diode for the laser cavity. So maybe 5w total electrical input for the laser. and another 10w or so for the TEC to cool everything? The traces on the low voltage side look like they've been designed with high current in mind though.

I couldn't find my model for sale any more, but the power supply 'box' of this laser http://www.ebay.com/itm/DPSS-100mW-130mw-532nm-Green-Laser-Diode-Module-TTL-Analog-110VAC-220VAC-Stage-/120964892851?pt=US_DJ_Lighting_Single_Units&hash=item1c2a11c8b3 looks VERY similar to mine.

What's interesting that that they state 5v at 1a for 100mw of laser output, so for 250mw that would be 10-15w maybe? I think that seems about right?

Edit: I took a look at the application notes, the primary configuration looks very much like that!

the values for Rsn and Csn look to be 5.6k and 1uf

Also now that I think about the laser power supply has a 0-5v TTL input to enable/disable the beam so it would make sense if the board was running at -5v,0,5v

[xygor]

Based on this being ~20W supply, my wild guess is that it is supposed to run at 50 kHz.  To be on the safe side, use the 800V part (KA5L0380RTU) as opposed to the 650V.  Maybe try it with a 100W light bulb in series with the line input to start testing.  Use AN-4141 for troubleshooting, especially the section on subharmonic oscillation.  If its doing that, the frequency may be too low.  Check the temperature of the transformer and U1 for overheating.

Edit: I will add that there might be a current regulator in the laser, so that the external supply needs to supply more power to accommodate that.  The modulation goes to the laser module and not to the power supply in question, right?  It might get routed through the power supply without connecting to any of the power supply circuitry.  I'll still stick with my original answer though.

[Psi]

There are people on the photonlexicon forums with crazy detailed knowledge of DPSS lasers and their drivers.

Could post there and ask if anyone has that same laser.

[AlessandroAU]

Got some good news, I didn't manage to repair the mains side, but after from more examination I was pretty much convinced that the circuit just ran off 5v, It doesn't even have a negative rail. So I wired in 5v and it worked .

I think I'll still fix it properly, but its good to know it'll run off 5v .

[xygor]

Measure the current while it's working.

[AlessandroAU]

Pretty close to 1a at 5volts, without the TEC. Turns out that several components that power the TEC have no been populated :\ Cheap bastards.