DIY JE-990 DOA

Shop deals on ebay
shop deals on amazon
Advertise with us
GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
Wow EVM that's the kind of analysis I need to get me started!

I just took some rubbing alcohol and a toothbrush and cleaned up the bottom of those two boards enough to see them. I actually used freon spray on the last one which is why it looked so gunked up.

I'll post updated pics of the bottoms as soon as I can.

Yes its Kester 63/37 lead solder. How can you tell from looking at it? I have two 1 lb spools. Enough to last a couple more projects - maybe more if I learn to use the right amount!

I have looked and looked at the bottom of the boards. The top of the boards I haven't really tried too much. With the silkscreening its tough to follow the traces. I'll bet the problem is on top of the board somewhere.

The only obvious difference between the two that do not work and the one that does is that it looks like I used less solder on the working one. That may have resulted in less chance of a solder bridge.

On the third one, I was thinking maybe I should use MORE solder to ensure a good connection - WRONG! The plated through holes should give a good connect anyway I guess and that ground plane seems to leave little room for error. More and more I am convinced that my problem lies on top of the board like you suggest.

Many many thanks. I'll update if I figure anything out.

Wanting to finish an API 312 and post some pics but need to have two working DOA's to say its really complete.
 
Also check the notes given in www.groupdiy.com/index.php?topic=10868 (last pages) about the overlay print causing shorts to the ground plane. I think these boards came from the same factory.

Samuel
 
Thanks Samuel for the tip. I too have a board that fails and appears to be a short to ground. (1 out of 8) I'll look for the silkscreen problem and see if it is there. I do have 8 more unstuffed 990 boards that I can inspect as well.

--Ethan
 
I was able to get the output transistors swapped the other night. 7 or the 8 I built were working. The non working 990 had the trace R6-q5-q6 near ground. Hooking up an ohm meter it registered 1.9 ohms.

I'm looking all around the traces on the bottem and top of the board and do not see anysoder bridges. So next step is to remove the resistor and transistors one at a time to see if that will show me the short to ground.

I did scrape away the silk screen just to eliminate wicking. All 3 parts out and still a short to ground. Looking closely through a magnifing glass shows the problem...

Take a close look at the emitter pad of Q5 in this photo (pad to the right):

q5-short.jpg


A little bit of photoresist in the wrong place and we have a 1.9 ohm resistor to ground.

Now they all work.

Here are 2 links to the test jig that I built (1 MB each photo):

http://www.arafel.org/audio/990/dcp_6065.jpg
http://www.arafel.org/audio/990/dcp_6066.jpg

You should note that I'm using a breadboarded JLM powersupply at +-12 volts and that I have current limiting resistors installed for testing. Just a simple non-inverting configuration. But it does have socket pins.

Regards
 
OK so I finally got one of the non-working doa's to work. It was the one that had a smoked R7. After careful study, I realized that there really are only two ground points in the circuit. Nothing else should go to ground. After checking continuity all over the place, nothing else was shorted to ground. So I started checking to see that I had continuity everywhere in the circuit I was supposed to and yes it seems I had.

Finally, I checked for shorts anywhere I might find them. Found that I had a short in the MJE172 power transistor! Fortunately, I had a working one to compare with. Also checked it with a new part from the kits. Replaced that MJE172 and WOW. I gotz another working doa!

Not sure if that would cause R7 to smoke but I don't think I did anything else to it.

I will see if I can troubleshoot the other doa and get it working but so far, the MJE172 does not show the same short that the bad one did.
 
Finally got the third doa working. Everything really seemed OK so I just took some excess solder off the bottom of the board. Plugged it in and still had horrible noise. Pushed around on the inductor and whoosh it went quiet (in a good way). Pushed the inductor around and it got noisy and then quiet again. Just decided to push the inductor into a good place and leave it. None of the wires of the inductor appeared to be touching anything but that inductor sure seems touchy! Sounds like a turntable when you scratch the record. I can see why it would be a good idea to use a little stronger wire than 30 guage so it will stay in a good place once you get it there. The 30 guage seems to allow it to move all over the place. Anyone else experience this with their inductors?

Anyway, SUCCESS! Three working doa's and I can finish my API 312 project. Yay!
 
I wonder about the inductor noise. I've been poking at mine to see if I can induce some noise. Nothing. Just quiet.

What comes to mind is: Are the inductor wires solder joints OK. If they are not properly tinned or cold soldered then in certain mechanical orientations they could make a lot of noise or act as a diode.

I used a small syringe full of silicone rubber (RTV) to squirt some under the inductors and fix them in place.

If you have a mechanical problem with the inductors causing noist, sooner or later they will find that position. (And Murphy will decide when to do it)
 
Thanks for checking your inductors. Sounds like I better unsolder them. Make sure the lacquer is stripped off well and put them back in place. I know I'll be glad for doing it later. I wanted to put the 26 guage wire in them anyway and this gives me an excuse to do that too.

Both channels work well with a dynamic mic but my phantom power is all crackly and noisy.

Got to replace my local filter caps first. They are 35 and 50V and I'm putting 51.5V there. Hopefully that's it. Next, I would get 1% 6.8K resistors as I'm using 5% ones from Rat Shack. Finally, I hope my switches are OK. I'm using two that I robbed from a Behringer B2 Pro. They say 0.4VA max. Pretty light duty. I would hate to have to re-wire them.
 
Had a chance to take the inductors out of these. Threaded them with 26 guage wire and stripped the ends down to all copper. Didnt' realize the lacquer needed to be stripped off. Make sure you do this and do it well. Was definitely barely making a connection before and now its stable.

Now to order parts to fix that phantom power...
 
[quote author="sbranco"]Had a chance to take the inductors out of these. Threaded them with 26 guage wire and stripped the ends down to all copper. Didnt' realize the lacquer needed to be stripped off. Make sure you do this and do it well. Was definitely barely making a connection before and now its stable.

Now to order parts to fix that phantom power...[/quote]

Thats great! Sounds like you found the solution. Thinking about your 6.81k phantom feed resistors. I remember that the absolute value of the resistors is not important but matching the resistors to each other is.

You could build a wheatstone bridge to match them if you want to be ultra precise or just use a 4 and 1/2 digit ohmmeter ( should have 10 ohm or better readout) to try to select close pairs.

The tolerance value (5%) is to the labeled value. Thus a 5% that is 5% low could be matched with a 5% that is 5% high. Giving you a 10% mismatch. Using 1% could give you a 2% mismatch.

Regards



http://en.wikipedia.org/wiki/Wheatstone_bridge
 
[quote author="evm1024"]I was able to get the output transistors swapped the other night. 7 or the 8 I built were working. The non working 990 had the trace R6-q5-q6 near ground. Hooking up an ohm meter it registered 1.9 ohms.

I'm looking all around the traces on the bottem and top of the board and do not see anysoder bridges. So next step is to remove the resistor and transistors one at a time to see if that will show me the short to ground.

I did scrape away the silk screen just to eliminate wicking. All 3 parts out and still a short to ground. Looking closely through a magnifing glass shows the problem...

Take a close look at the emitter pad of Q5 in this photo (pad to the right):

q5-short.jpg


A little bit of photoresist in the wrong place and we have a 1.9 ohm resistor to ground.

Now they all work.

Here are 2 links to the test jig that I built (1 MB each photo):

http://www.arafel.org/audio/990/dcp_6065.jpg
http://www.arafel.org/audio/990/dcp_6066.jpg

You should note that I'm using a breadboarded JLM powersupply at +-12 volts and that I have current limiting resistors installed for testing. Just a simple non-inverting configuration. But it does have socket pins.

Regards[/quote]

Can you go into a little more detail about you jig, I might want to do this.

Are those 47ohm resistors coming off the secondaries of your transformer to limit the output current to about 255mA? How many Watts? 2?

It's kinda hard to see what else is going on, do share!
 
Its a simple jig that has socket pins so that a new opamp can be inserted easily. gain of 10 non inverting . Input off to a signal generator and input and output to a scope.

I can swep the freq and amplitude manually and look for anything wrong. It is a 100 MHz scope so any oscillations should show up.

The resistors do form a current limit into the regulators. No magic in the value. Just what I had around in that wattage. There are there to give you time to shut down before anything smokes. 200 mA is better than an amp or 2. (Too bad I bypassed them on the one I blew up)

By using this jig I was able to test all 8 opamps in a simple circuit. It would have been harder to test in the just build (an untested) JLM twin servo board I assembled.

Also it is available for the melcor and Sga opamps I have yet to build.

Regards, Ethan
 
Got my two 990 kits today. Stuffed and soldered them right away and now they sound awesome on my 312 boards. :green: :thumb: :thumb:
 

Similar threads

zzzzz
2520-style DOA to single DIP8 opamp adaptor PCB (open-source / gerbers included)
Replies
0
Views
190
zzzzz
zzzzz
zzzzz
Chandler
Replies
1
Views
500
zzzzz
zzzzz
F
MXL 990 mod question from Norway
2 3
Replies
44
Views
3K
Frdfrode
F
Y
DIY U67 Power supply
Replies
6
Views
457
RuudNL
RuudNL
zzzzz
thread for open source projects / gerbers (for members only?)
Replies
6
Views
217
pucho812
pucho812

Latest posts

Back
Top