DIY ACA mini

you must clean well the mosfet leg
1675888837533.png
 
What temp is your soldering iron? It may be a bit low. Adding a bit of flux to the rework joint can also help it flow better. Using NP's artwork from Page 3 of his writeup (attached), you can print it and test for continuity along each PCB trace to confirm you have a good connection.

I tried to circle the pins to show where you should have continuity.
Continuity test points.jpg

If you don't have continuity, that will be a good clue to make note of. Hang in there--troubleshooting can be frustrating.
 

Attachments

Orange, no continuity, getting a resistance reading?
  1. How much resistance did you read? That's the variable resistor, so some is to be expected.
  2. How does the continuity on the rest of the board check out?
  3. Can you adjust the temperature on your soldering iron?
  4. What kind of solder do you use?
  5. What kind of tip is on your soldering iron?
  6. On the side that pulsed earlier, did you plug in the power supply and immediately turn it on? Mine does best if I wait for a 5 seconds, or so. If I flip the switch too quick, it will pulse and I just have to turn it back off, wait, and turn it back on.
It's hard to tell from the photos, but the blobbiness (did I make up a word?) could be hiding a cold joint somewhere.
 
  1. How much resistance did you read? That's the variable resistor, so some is to be expected. 113ohms and starts climbing.
  2. How does the continuity on the rest of the board check out? I have not traced the whole board yet, but I am having a problem with it going into protection, if I had an open circuit somewhere would it actually do that, or would it just not work ?
  3. Can you adjust the temperature on your soldering iron? Yes, it's all the way up. Model is here: https://www.parts-express.com/Stahl...perature-Soldering-Station-374-100?quantity=1
  4. What kind of solder do you use? It's Cardas Quad Eutectic with rosin core, on the heatsink, I used a more traditional old leaded
  5. What kind of tip is on your soldering iron? the tip is pretty thin, it is tinned well, shining bright when cleaned
  6. On the side that pulsed earlier, did you plug in the power supply and immediately turn it on? I did, Mine does best if I wait for a 5 seconds, or so. If I flip the switch too quick, it will pulse and I just have to turn it back off, wait, and turn it back on.hmm
It's hard to tell from the photos, but the blobbiness (did I make up a word?) could be hiding a cold joint somewhere. I understand, before I hooked this up it was running great for burn in/warm up/biasing. Got all my levels set right over the couple hours I had it one, and left it for the next day. When I put it in, I was short on speaker wire, also had it plugged in and ON when I hooked up the speaker cable. I believe I shorted something in the process with the bare speaker wire. Possibly even on the bottom of the board.

Man, I was pretty happy with my work, until I blew it up. This was the board before biasing. I am just trying to regain a bit of credibiltiy. I have been soldering for 30 years on and off haha!
 

Attachments

  • IMG_1493.jpg
    IMG_1493.jpg
    507.5 KB · Views: 102
Just to make sure...

When checking continuity here, the unit is unplugged from its PS....

You have just your unit, without its SMPS, and a multimeter set on reading R automaticaly (or so) and you use 2 pins to measure R between the points you want to measure... right?

Sorry, is quite obvious, but reading the above I just wanted to make sure...
 
Hey, @St. Peter(sburg): for the side where the SMPS entered protection mode, do you want to try it again? This time, before turning it on, plug in the SMPS and wait 5-10 seconds to allow the caps to charge. If this works, will it play music on that side? I'll keep my fingers crossed that it does!

If that doesn't work and you have good solder joints, that may help the smarter people determine where the damage is located.

Note for after your amp is fixed: when the SMPS is plugged in and the switch is off, the amp will still thump if you plug/unplug the input cables. To avoid this, I unplug the SMPS and wait for it to fully discharge before messing with the connections.
 
I know I am somewhat persistant on that, but...

Shouldn't "orange" check be 0R (or negligeable resistance) if aiming at the very next legs of the variable resistor directly?

If so, and it is showing 113R, wouldn't that indicate that possibly one of the ciruit trace from tha transistor is damaged... and a simple fix be just to solder a wire in parallel between said transistor leg and the variable resistor leg?

Not saying it would be the only damage nor the transistor would be intact, but step by step perhaos...

Just thinking out loud, and of course assuming all care had been taken re "swich on" procedure to avoid SMPS hi-cup mode...

Claude
 
Update. Last night, after some continuity checks and solder pad reworks and cleaning. I decided to solder back in the R10 leg. I put it down for the night. this am I connected the smps, but did not turn it on for a minute or two based on Chris's recommendation. Well, when I did the whole thing came on, at 9:02 and is still on now. I am reading voltages at the bias points that make sense and are in range. Thinking I leave it one foe a while, tweak the bias and test in my system. Right now, my VB is .200 and my VO is 15.5 on bolth sides give or take. Thanks for the support so far all.
 
  • Like
Reactions: Chris Jones KS