A12 and PS3 problems

[daveotero]

I started out thinking I had a PS3 problem but now I'm not sure where it is. I posted this in another forum but I'll copy it here.

--------------------------------
Crap!

OK, I'll start with my only known error so far. The very first time I powered up my A12 module for testing I had the molex connector wired backwards, as in pin6 to pin1, pin5 to pin 2, and so on. The PS3 health indicator lights flashed around a little and after about 10 sec. the main power light started flashing... Power down, realise the problem, curse a lot, fix the problem.

Now for my second try. connect everything and power up unit. Only the first 2 health indicator LEDs on the PS3 light up (+30V and -30V on, +48V off) and the LED on the A12 lights up VERY faintly. So much so I almost couldn't tell it was on. At this point a see and smell a little bit of smoke rising from somewhere under the molex connector on the PS3. Power down again, much more cursing, disconnect the A12.

I removed the front cover to try and scope where the smoke was coming from but this time there was no smoke. I still got the first two LEDs only and my voltages on the A12 we're way off (something like 43V at p4/p3 and 31V at p4/p5.

When i power on the PS3 with no modules connected it seams fine and all 3 indicator LEDs light up. Please help me figure out what I screwed up when I reversed the molex connection or is this something unrelated?
-------------------------------------
Update...

Checking over my work again and low and behold the very last component I installed (and most obvious ) C13 was backwards. I flipped it around and tried again.

This time I powered up and got all 3 health indicator LEDs. After about 15 seconds R32 went up in smoke! It's completely toasted. I'm hoping it was just damaged already and I could replace it and everything will work fine. Am i just dreaming? Do I need to replace C13 as well?
----------------------------------------
I replaced the resistor at R32 and tried again. I think I must have burnt out a part on the PS that's causing this. I think the problem might be the large resistor (R12) below one of the heat sinks under the molex connector on the PS3. Can anyone confirm this?


Sorry for the long read, any help would be much appreciated.

[tpryan]

I'd suspect a shorted C16 on the A12. Pull it out and see if the PS03 indicators all light.

[daveotero]

OK, I removed C16 and R32 is still removed. Powered up and got all 3 light on the PS3, and nothing went up in smoke. I think thats a first for me .

Anyways, so should I just replace C16 and R32? Do I need to worry about the cap I reversed at C13? I know I saw smoke from the PS3 at one point but I can't visualy see anything thats burnt. Could there be something wrong even if I'm still getting all 3 LEDs?

[tpryan]

Anyways, so should I just replace C16 and R32?

Yes.

Do I need to worry about the cap I reversed at C13?

Replacing C13 won't hurt.

Could there be something wrong even if I'm still getting all 3 LEDs?

Yes, but if the output voltages are OK, there's not much to do at this point. The prolonged dead short across the +48V output probably got R13 hot enough to smoke a bit, but if it had opened you wouldn't be getting any +48V at all. If you plan to keep shorting it out, you could replace it with a 5K6 or so.

[daveotero]

I replaced C16 and R32 and tried again. I get all 3 LEDs but R32 still burns out. I did not replace C13 (couldn't find the right part around here) could that be screwing things up?

[daveotero]

Alright, I found a 1500/50V cap for C13. Its not the snap in type and its a little tall but I just wanted it for testing purposes. I installed it and tried again. This time I switched off +48V on the module before powering up.

Powered up and R32 was fine and everything looked good. Before I tried the +48V switch I checked the voltages at U1 4/3 and 4/5. Pins 4/3 showed 78V and 4/5 showed 64V. I tried the +48V switch and R23 started smoking again. I'm still totally lost here.

I have double checked every component on the A12 and everything is right and I don't think there are any shorts unless there is a solder short under one of the larger components.

Is there anyway to check the current coming from the PS3 with my DMM?

This is getting really frustrating but on the plus side I'm learning a lot from trying to figure it out and I'm getting pretty good at desoldering. The worst part about it is that I just had a client in the studio with a brand new $5000 Pearl Masters Reference Series kit and I would have killed to get some tom samples with the A12.

[daveotero]

OK, I just cant stop screwing with this thing.

I started looking around at the PS3, testing resistors and so forth. I found two things that seem strange to me.

First off, the schematics show R11 being a 1K resistor and its labeled that way on the board. My PS3 doesnt have anything at R11, is this alright?

Second, I was testing all the resistors in the area I first saw smoke. Although it looks fine on the outside I can't get a reading from R14. Its supposed to be 220R but I can only get it to flash a reading around 180. If i test it for continuity i get a very short click when I make contact. None of the other resistors do that.

Could either of these explain whats going on with my A12?

[tpryan]

I think you oughta just send it in, we'll get it going.

[daveotero]

Well, I finally got it working right. To make a long story short, for the last 2 days I've had the molex harness wired in reverse.

When I started the kit a few weeks ago I soldered the molex connector in backward by accident. Rather than deal with desoldering the multi pin I figured i would just wire the harness by the pin1s. Then a few days ago when I first had problems I figured the harness was soldered correctly (forgetting my mistake) and re wired according to that, when in fact my problem was the backwards C13. After I fixed C13 I never thought about checking the molex connection again. So lots of soldering and a few 150ohm resistors later I finally have a working A12. At least working so far, the real test will be tomorrow when I actually run signal through it.

Thanks for all the help and I'm sure the next one will go a lot smoother!

[tpryan]

OK, mystery solved!