MUFF WIGGLER Forum Index
 FAQ & Terms of UseFAQ & Terms Of Use   Wiggler RadioMW Radio   Muff Wiggler TwitterTwitter   Support the site @ PatreonPatreon 
 SearchSearch   RegisterSign up   Log inLog in 
WIGGLING 'LITE' IN GUEST MODE

I destroyed my Multimix
MUFF WIGGLER Forum Index -> Oakley Sound Systems  
Author I destroyed my Multimix
catatemycaps
I did some module re-organizing the other night and in retrospect I should not have done that.
I forgot that I had fitted the PSU header backwards on the SVF and of course managed to reverse the + and -. This became very evident as soon as I switched on the power...the whole system was silent.
I immediately turned the power off and checked all power connections and that's when I noticed what had happened.

The funny thing is that SVF was ok. The only module that stopped working was multimix. No smoke, no funny smell, no popping sounds.

Now both leds are constantly on and there is no audio from any of the outputs.

I replaced the Q1 and U6 as I thought those two might have fried up. I have misplaced my schematics someplace so I'm pretty much doing guess work here.

Well, the problem didn't go away by doing that.

I did some measuring and the results are somewhat peculiar for U7 and U8. -13.4 V on pin 1 for both. I think the voltage should be about +15V.

Other ICs are showing +14 - +14.5 V except for U6 which is still at +2.9V.

Any idea what might be going on?

Thanks!
catatemycaps
I see I made a mistake in measuring. I did not mean to measure from output pin (pin 1).

I measured the voltage again for TL072s.
All are at 0.66 V at pin 4 and 0V on pin 8 except for U7 and U8 which are at -15.13 V on pin 4.

What on earth could cause something liket this?
Synthbuilder
catatemycaps wrote:
All are at 0.66 V at pin 4 and 0V on pin 8 except for U7 and U8 which are at -15.13 V on pin 4.

What on earth could cause something like this?

Nothing good I'm afraid. Let's hope you made a mistake in measuring. But if these are true readings then the copper track on the PCB that carries the negative rail has burnt out. That takes a lot of current to do that because I use 50 thou tracks.

So with no power attached and all the ICs removed from their sockets you should measure between pin 4 on the power header to pin 4 on all the ICs. Any that don't read zero ohms (or so) will have lost their negative rail. Since 7 and 8 are nearest the PSU headers it is likely the break is somewhere on the underside.

One of those op-amps is likely to be toast. And quite possibly all of them will need replacing.

Tony
catatemycaps
Well, that kind of current should not be possible as I'm using PA-20 and I'm sure all my other modules would be toast too.

However, when I started assembling multimix I noticed two scratches on the back side of the pcb. I thought that those scratches were only in lacquer (or whatever that coating is called) and continuity measurement didn't show anything wrong either so I thought the pcb was ok to use.



Hopefully this picture helps. (mobile phone is not the best choice for photography)
I checked those two tracks again and they still seem ok. I hope they will not cause problems in future.

I'll check the values without ICs as you suggested.
catatemycaps
All ICs read 0V on pin 4 except for U6 that is at -15.13V.
catatemycaps
Well, I had no luck in finding the source of the problem last night. I checked both sides of the pcb to no avail.

I also measured all the power cables from dizzy to modules and no glitches there either.

I'll continue hunting down the problem later today after work.
Synthbuilder
catatemycaps wrote:
All ICs read 0V on pin 4 except for U6 that is at -15.13V.


Hang on, that's different to what was written earlier and it makes all the difference. Pin 4 of U6 should be 0V, and pin 4 on all the op-amps should be -15V. Which may mean that you have -15V and ground swapped on your power lead.

Tony
catatemycaps
Ok. There is some light at the end of the tunnel after all.

Sorry about all the confusion. My handheld meter seems to be ok when measuring resistance and AC voltage, but DC voltage seems to give very odd results over all.

This time around I used my desktop meter (and buried the piece of s...garbage for a meter six feet under).
I'm doing my project in our study that I share with my wife and it's getting quite cramped in there and I need to keep some of my equipment stashed away.
This includes my desktop meter. That's why I'm so eager to use the handheld one. smile

Here are the actual readings:
Pin 4 Pin 8
U1: -7.936 15.050
U2: -8.130 15.050
U3: -8.220 15.049
U4: -8.205 15.048
U5: -8.215 15.048
U6: 0 14.928
U7: -15.113 15.049
U8: -15.113 15.048

Not quite what they should be but looks a lot better now.
Synthbuilder
With the power off and power lead disconnected from the module. Measure the resistance from pin of 4 of U8 to pin 4 of U1. It should be a near short, ie. zero ohms. If isn't you have a break somewhere on that track. Maybe just below pin 4 on U8 on the topside.

But it could be that left hand track in your photo above. A possible track break where the solder mask (the green coating) has been hand finished may cause the problems you are seeing. It may have been OK before the power reversal but it could have burnt out with the current surge.

Tony
catatemycaps
Seems fine as I get 0.
Synthbuilder
I'm stumped. I'm not sure how it can be zero ohms when there's a voltage drop...

Maybe it's intermittent... do the voltages on the errant pin 4s change when you flex the board?

Tony
catatemycaps
Yes it does. Now the meter is showing 0L in automatic range mode which indicates that the resistance is greater than 50M.

I fitted a jumper wires over those two cracks that are evident on the coating, but that didn't help so the break must be someplace else.

One wire goes from C15 to C10 and the other from U8 pin 3 to the other end of C10.

Edit: Erm...I see you asked for voltage and I measured resistance, but nevertheless...
Synthbuilder
Apply a little more solder to the left hand bottom side pad of C15. The left hand one being the one directly underneath pin 4 of U8.

Also apply a little more solder to the topside of the board at the same pad. The negative supply to the other op-amps goes through that through plated hole from topside to bottom. It's unusual but the power reversal may have blown out the through plating inside the hole.

Tony
catatemycaps
I bow to you oh master synth builder! we're not worthy

That did the trick! nanners nanners nanners nanners

I must remember to throw in an extra fiver or so in my next order so I can buy you a pint! Guinness ftw!


Thanks Tony!
catatemycaps
I can't help it but a couple of things still bother me.

When I applied some pressure on the board when measuring resistance between pin 4 of U1 and pin 4 U8 the readings varied.
That was before I added more solder on joint of C15.

Could this bad joint/blown out through plating cause that kind of behavior?
Surge opened a bad joint and pressing the board made the joint act like a switch?

Anyhow, readings are consistent now but could that surge have done something else too that is not yet evident?

Do I worry too much? wink
Synthbuilder
Glad to hear it's working now.

catatemycaps wrote:

Could this bad joint/blown out through plating cause that kind of behavior?
Surge opened a bad joint and pressing the board made the joint act like a switch?

Yes, I think so. It's difficult to tell for sure though but I wouldn't worry about it if the circuit is now working. I'd be tempted to replace the op-amps at some point. They won't have liked having the power reversed through them and this can age them prematurely or make them go off-spec.

Measuring resistance with any sort of meter will give you an average read out. So often a rapidly opening and closing of a circuit will make the meter give all sorts of wild readings. That's why I like a scope. You get a pretty much instantaneous reading and you can see the abrupt changes in real time.

The through plating on any hole is pretty rugged but that particular hole would have taken a lot of current for the time the power was reversed. By filling it with solder you have bridged the gap broken by the current surge. Back in the old days a lot of PCBs didn't have through hole plating and you had to solder both sides of the boards where you wanted current to travel between the layers.

Tony
catatemycaps
Thanks for the tip.

I thought that true RMS meter would give the measurement almost instantly.
I'll use my scope more for troubleshooting in future.

I fitted a fresh set of op amps into sockets. I reckon most of the old ones ..if not all...are Dead Banana as you said.
MUFF WIGGLER Forum Index -> Oakley Sound Systems  
Page 1 of 1
Powered by phpBB © phpBB Group