Dyna comp help! Driving me insane

[jonny]

Alright I've been trying to build a dyna comp from one of the factory boards sold through here.

Bypass works fine however engaged I hear nothing - no matter what I do with the knobs, maybe a crackle haha.
I've followed it through with an audio probe and it gets to pins 2 and 3 of 3080 but doesn't come out pin 6. I've quite thoroughly checked all the resistors and caps and transistor orientations but still can't find a problem!  

I mainly used this as my guide(checked most of it with the schem and seemed good) - but if there's a problem I didn't see with it then most likely will  be a problem with my board (I did notice one resistor was wrong the 27k at the top should be 15k) :





Schem:



Things i've done a little differently from normal(which I don't think should matter):
-the 2k2 trim i've replaced with two 1k1 resistors (I didn't have a 2k trim and I hear that for most occasions it works best in the middle).
-all the transistors are MPSA18's, I've also tried 2N3904 but the same problem (they are all socketed)
-tantulum 1uf caps so can't screw up the orientation on those.

Voltages:

3080
1) 0V
2) 3.7V
3) 3.7V
4) 0V  (V-)
5) 0.6V
6) 0.73V <<< (output from OTA) something wrong here this is where my I think I lose my signal, I get nothing on the audio probe
7) 9.15V (V+)
8 ) 0V

Q1
C) 9.15V
B) 1.3V (with an audio probe I hear signal on both collector and emitter so think it's okay)
E) 1.9V

Q2
C) 8.8V
B) .73V << connected to pin 6 of OTA, something wrong here..
E) .23V

Q3)
C) 9.15V
B) 8V
E)8.6V

Q4)
C) 8V
B) 0V
E) 0V

Q5)
C) 8V
B) 0V
E) 0V








Any help would be greatly appreciated! I don't have another 3080 to test it with which I wish I did!

[Johan]

I believe you should have close to 9volts at pin 5 of the 3080 when no signal is precent.  check for shorts around Q3
J

[jonny]

Reading through another thread R.G said this about that pin:

Following up on the note above, Pin 5 is where the current telling the chip how much to amplify comes in. This pin is **always** at 0.4V to 0.7V on a 3080 if it's working. 0V on this pin indicates it is being TOLD not to amplify. This is what would happen if the wiring through the 27K resistor and sustain pot to the emitter of Q5 is open.

So maybe it's okay, something is definitely wrong around there somewhere though  

[Johan]

I would trust R.G's word then..especially since I havnt built this myself. but on the other hand it would be easy to test. just desolder the wire going from the transistor to the sensitivity pot and  and let it go from +9volt to the pot instead...if you get signal through when adjusting the pot, you  know where to look next.  if it doesnt make a differance, hook the cable to ground and retest...still nothing, then I'd suspect the IC

[jonny]

Good idea, I'll try that tomorrow... I'm starting to believe I may have killed the IC, at first I accidentally put in 2k7 resistors instead of 27k resistors. This could have possibly killed the chip because apparently pin 5 is very sensitive and you can kill the IC if you feed it 1ma, 9V/(2k7+sensitivity knob) = 3.3mA(with sensitivity knob=0)   and I would have moved the sensitivity knob when I first tested it. Oh well, I have more 3080s coming soon. But maybe there is another error and the chip is actually fine, who knows
Thanks heaps

[petemoore]

Voltages:

3080
1) 0V        N/C
2) 3.7V   
3) 3.7V
4) 0V  (V-)     
5) 0.6V
6) 0.73V <<< (output from OTA) something wrong here this is where my I think I lose my signal, I get nothing on the audio probe
7) 9.15V (V+)
8 ) 0V
   Check every pin node, pin 1: ''measure every mark'' [measure with DMM, physical board, compared to marks on schematic] including resistors, wire-traces, everything]. Skip the caps or check they do DC block and pass a 'click' [metal contact-touch may put a click-signal through it] or buzz [connect your thumb to the input side] ie pass AC signal, but they always do [knock on wood].
Q1
C) 9.15V
B) 1.3V (with an audio probe I hear signal on both collector and emitter so think it's okay)
E) 1.9V
   ...looking for 1:1 signal at buffer output, seems 'some' signal would be at the collector [since the emitter moves] but that it'd sound like a 'rail', V+...as long as the emitter outputs, but the base should be a diode drop above emitter.
Q2
C) 8.8V
B) .73V << connected to pin 6 of OTA, something wrong here..
E) .23V
Q3)
C) 9.15V
B) 8V
E)8.6V  Complicated around here...the emitter goes through the .5M pot to the OTA?
Q4)
C) 8V
B) 0V
E) 0V  *B/E [or any two pins] at same voltage is always suspect,

Q5)
C) 8V
B) *0V
E) *0V
  The diamond shape of resistors at the left side of the board appear to be 100ohm, I couldn't find this value on the schematics.
  The board is different than layout ?

[twabelljr]

The diamond shape of resistors at the left side of the board appear to be 100ohm, I couldn't find this value on the schematics.
  The board is different than layout ?

 

On the schematic it is a 2.2k trimmer between pins 2 & 3 with the wiper going to the 470k then 9v. As it is connected now it is split down the middle with 1.1k to pin 2 and 1.1k to pin 3. Maybe not being able to adjust this is causing an issue?? Perhaps a pot could be substituted for troubleshooting.

[ayayay!]

Something is super wacky on Q3.  I'd desolder that whole socket to ensure nothing is touching on both sides of the pcb.  Good luck. 

[R.G.]

First, thank you for sending a complete set of information on this. It helps me debug - a lot.

Things i've done a little differently from normal(which I don't think should matter):
-the 2k2 trim i've replaced with two 1k1 resistors (I didn't have a 2k trim and I hear that for most occasions it works best in the middle).

This works perfectly well in almost all instances. It is very, very unusual that an actual trimpot is needed. It was needed on the old 3080s from the time the Dyna was designed. ICs are better now.

-all the transistors are MPSA18's, I've also tried 2N3904 but the same problem (they are all socketed)

Not a problem.

-tantulum 1uf caps so can't screw up the orientation on those.

Tantalum is polarized. Be sure to check for orientation. I *have* messed up tantalum orientation before.

1) 0V
2) 3.7V
3) 3.7V
4) 0V  (V-)
5) 0.6V
6) 0.73V <<< (output from OTA) something wrong here this is where my I think I lose my signal, I get nothing on the audio probe
7) 9.15V (V+)
8 ) 0V

All OK, except as you note, pin 6 is not right. The low voltage on this pin is why you're getting no audio. There are two possible issues with this. One is that your bias network may be incorrect or faulty. The other two are that the 3080 may have been damaged, or the transistor hooked to pin 6 may be damaged - or both.

Check the DC voltage on both ends of the 150K connected to pin 6. If the voltage on both ends is less than 2V, then you (at least) have a problem with the bias section, the 56K and 27K that should give about 2.9V to the output. If the voltage on the 56K/27K side is 2-3V and the pin 6 side is low, then the bias is OK, and there is a problem with the 3080 or the transistor. If it's the bias resistors, check to be sure that the 1uF cap is oriented correctly.

Tantalum *can* be oriented backwards. At one time IBM had special 3-lead tantalums made up with + on both outside ends and - in the middle so it was impossible to insert them backwards. They had power supplies big enough to not only kill the tantalums, but start fires when the tantalums failed.

Q1
C) 9.15V
B) 1.3V (with an audio probe I hear signal on both collector and emitter so think it's okay)
E) 1.9V

It's quite unusual to hear audio on the collector of this. The base really can't be at 1.3 and the emitter at 1.9 and have it work; I think you may have mixed up which lead is which.

Q2
C) 8.8V
B) .73V << connected to pin 6 of OTA, something wrong here..
E) .23V

I agree - something is wrong with the OTA or this transistor. Since the transistor is socketed, pull it out and then measure the pin 6 voltage. If the voltage on pin 6 rises to about 2.9V, then the transistor was the problem. If pin 6 stays low, it is a problem with the 3080. Divide and conquer.

Q3)
C) 9.15V
B) 8V
E)8.6V

Q4)
C) 8V
B) 0V
E) 0V

Q5)
C) 8V
B) 0V
E) 0V

S'OK.

Check the bias voltage and pin 6 without the transistor, and report back.

I'm starting to believe I may have killed the IC, at first I accidentally put in 2k7 resistors instead of 27k resistors. This could have possibly killed the chip because apparently pin 5 is very sensitive and you can kill the IC if you feed it 1ma, 9V/(2k7+sensitivity knob) = 3.3mA(with sensitivity knob=0)  and I would have moved the sensitivity knob when I first tested it.

You're right on here. More than 1ma into pin 5 will kill the 3080 dead. We'll find out more when you do the tests I recommended.

[jonny]

Thanks very much for all the suggestions. Just double checked caps are orientated correctly, I've measured the 150K and it = 2.7V and 0.65V. With 0.65 being connected to pin 6 3080. If I remove the transistor pin 6 3080 remains at 0.65V. I'm starting to think I killed my 3080? I have more on the way but they'll take 2 weeks to get here.
Thanks so much for all the replies!

[R.G.]

Thanks very much for all the suggestions. Just double checked caps are orientated correctly, I've measured the 150K and it = 2.7V and 0.65V. With 0.65 being connected to pin 6 3080. If I remove the transistor pin 6 3080 remains at 0.65V. I'm starting to think I killed my 3080? I have more on the way but they'll take 2 weeks to get here.

Sorry - the bias voltage seems OK at 2.7V, and with the transistor removed, the output pin only "sees" the 150K load. If it's at 0.65V, my first guess would be that you killed the 3080 with the resistor foulup when you were bringing it up. With 2.7K instead of 27K, the current could go to 9v/2.7K = 3.33ma. and that is enough to kill the 3080. They are very unforgiving on this.

Take that one out and put another in when you get them. Watch the orientation, especially if you put in a socket for the IC. One of my personal favorite tricks was to insert ICs backwards in sockets.   

If you did not use a socket for the IC, clip the IC pins off next to the body of the IC, and then hook the tip of your soldering iron under the curve where the pin went into the body on each pin. This should heat the pin enough to melt the solder in the pad, and lift the pin out of the hole. When you've done all eight pins, re-melt the solder in the pad and stick the pointed end of a wooden toothpick into the hole to clear it.

[jonny]

Okay thanks for all the help. Lets hope it works first time in 2 weeks when the new chips arrive.  After screwing up so much these days I just socket transistors and ics it's so much easier than desoldering.
Also when I said I hear audio on the collector of Q1, I said the wrong thing and meant emitter and base not..

Thanks everyone.

[R.G.]

Okay thanks for all the help. Lets hope it works first time in 2 weeks when the new chips arrive.  After screwing up so much these days I just socket transistors and ics it's so much easier than desoldering.

I used to do that. These days, I find that I think of sockets as more pain than they're worth. I guess I mostly use low-end chips (jellybean opamps and such) and the chips are often cheaper than the sockets they go in. But you're right, if you replace things a lot, sockets make sense.

[jonny]

Just a bump to say that it was the IC. Sounding great now, Thanks everyone 

[R.G.]

Hey, congratulations. Good work.