help in debuggind dynacomp clone

[xfc]

hello,
I'm new in this forum and this my first build
a dynacomp clone layout and schematics from general guitar gadgets
with an addition of an attack knob by replacing R18 with a 10k in series with a 250k pot.
i get no sound out of it...

I've checked and found that i will hear sound out of leg 6 of the ic (it's signal out), but after q2 i'm getting just loud buzz
the same thing goes with leg 5 of the ic (voltage control input).
and at the base of the output buffer there no sound at all.

the voltage reads on those transistors:
Q2
C - 4.64
B - 2.44
E - 2.39

Q3
C - 8.96
B - 0
E - 0

Q4
C - 8.97
B - 0
E - 0

Q5
C - 9.33
B - 8.98
E - 8.81

all measurements made with a new battery (voltage about 9.4).
the transistors numbering is according to the layout (i believe theres an inconsistency between the schematics ans layout).

if anybody could help and give me some direction it would be great
thanks

[petemoore]

  Q3 and Q4 are NPN Bipolar, the base needs to be above emitter by one Si diode threshold voltage.
  Since both may be related....
  Look for a strand or solder trace from Gnd. to the base trace, perhaps both transistors likely share the miswire.
  q5 has the emitter at near V+, it should be nearer ground, test the E leg connects to ground.
  Go through the instructions on the debugging page.

[R.G.]

the voltage reads on those transistors:

Hi, and welcome. If you could read "Debugging thread: What to do when it doesn't work" and give us that info, it would be a huge help in debugging. It's possible that the other IC pin voltages don't matter - but it's also possible they do. Also, the actual links to the schematic and layout make it easy for those of us who want to help to find the actual plans you used, instead of what we remember, perhaps inaccurately.

[xfc]

sorry... i read the thread now, so here a repost

the effect is a dynacomp clone, here's the link:
http://www.generalguitargadgets.com/diagrams/geo_d&r_comp.pdf

with a mod for attack knob by replacing r18 by a 10K resistor and 250K pot in series
built on a prefboard but basically the same layout as the suggested layout in page 5

as i said before, i'm getting no soud at all at the output, but i do get sound along the signal path up to the emitter of q2 where i get a loud buzz
i also get that buzz at leg 5 (the control input) of the IC.
at the emitter of Q5 i get no sound at all.

the voltages:
Q1
C-9.35
B-1.4
E-2

Q2
C - 4.64
B - 2.44
E - 2.39

Q3
C - 8.96
B - 0
E - 0

Q4
C - 8.97
B - 0
E - 0

Q5
C - 9.33
B - 8.98
E - 8.81

D1
A-0
C-9.34

D2
A-0
C-0

D3
A-0
C-0

IC legs
1- N.C.
2-3.85
3-3.84
4-0
5-0.56
6-2.45
7-9.35
8-N.C.

a fresh battery measured 9.4v
the transistor are numbered differently in the schematics and layout, my numbering is according to the layout.

i believe i covered it all...

any one have a suggestion?

oh... i also checked for a grounding of the bases of Q4 and Q3 as petemoore suggested but failed to find any
i also checked it with continuity check of the Meter.

[R.G.]

the voltages:

Q2
C - 4.64
B - 2.44
E - 2.39

I think most of your problem is here. I believe that the emitter is not making contact with its emitter resistor, or the emitter resistor is not connecting to ground. Measure the DC voltage to ground on both ends of the emitter resistor to tell which it is.

[xfc]

i'm getting 4.6 voltage drop cross that resistor
the othe leg is in ground.thats weird is'nt it?
i will check the connection to the emitter tonight ans post...

[petemoore]

http://www.generalguitargadgets.com/diagrams/geo_d&r_comp.pdf
  Seems like I always get at least a little glimpse of something more when I read at GEO, thanks RG.
  Good luck xfc, I had perfboard experiences with the Dynacomp.

[xfc]

woo, i should never write a post before my morning coffe 

anyway, as it turn out i accidentally wired the emitter resistor and cap to the collector...
i rewired them and now i've got sound. but it acts very strange, when i'm turning the sustain knob i get a lound noise and a sudden volume raise when the knob is about middle way. and then it's drop back as i continue to raise the sustain.

already replaced to sustain pot, wich didn't help

[R.G.]

OK then! We have removed one layer from the onion.

Time to go back and see what the voltages say now that you've changed the circuit.

[xfc]

here are the new readings:

Q1
C-9.35
B-1.4
E-2

Q2
C - 7
B - 2.4
E - 2.21

Q3
C - 9.2
B - 0
E - 0

Q4
C - 9.2
B - 0
E - 0

Q5
C - 9.33
B - 9.2
E - 8.8

D1
A-0
C-9.34

D2
A-0
C-0

D3
A-0
C-0

IC legs
1- N.C.
2-3.83
3-3.83
4-0
5-0.59
6-2.4
7-9.3
8-N.C.


also when rotating the trim i will get to a point (about 60% the way up) where i'll get no sound out.

[R.G.]

here are the new readings:

Q1
C-9.35
B-1.4
E-2

This is a problem if you're reading it correctly. For an NPN transistor to work correctly, the emitter must be about o.4 to 0.6V lower than the base. If you don't have the base and emitter mixed up, it's a problem. Here's one other issue. The base is a very high impedance point here, biased as it is by a 470K or 1M. If your meter has an input resistance of only 1M or so, it's loading down the base, giving you a false reading and the circuit is OK, maybe. Hard to tell what's going on. However, the bias voltage on this board (junction of R13 and R15) is nominally 2.92V with a 9.4V battery, so maybe the emitter's OK.

Q2
C - 7
B - 2.4
E - 2.21

Still does not look right, but subject to the same problem as above.

Q3
C - 9.2
B - 0
E - 0

Q4
C - 9.2
B - 0
E - 0

OK for no-signal condition.

Q5
C - 9.33
B - 9.2
E - 8.8

Probably OK.

IC legs
1- N.C.
2-3.83
3-3.83
4-0
5-0.59
6-2.4
7-9.3
8-N.C.

And probably OK.

also when rotating the trim i will get to a point (about 60% the way up) where i'll get no sound out.

The trim is the source of more confusion than any other part of this board. Set it to absolutely middle and leave it there. Or replace it with two 1K resistors.

With everything potentially being OK, what's left? Ugliness. It's likely to be oscillating from a wiring problem. I suspect that because you said:

i rewired them and now i've got sound. but it acts very strange, when i'm turning the sustain knob i get a lound noise and a sudden volume raise when the knob is about middle way. and then it's drop back as i continue to raise the sustain.

That's pretty descriptive of an RF oscillation problem.

Is it still doing that?

[xfc]

ok

i rewired the output and route it away, it made the helicopter sound i heard go away.
i also tide up the wires that go to the pots and battery.

i still get some loud static noise.

i played a bit with that trim, and manage to tame that bump in the volume and push it to the end of the pot range (let say the last 20% of the rotation of the sustain knob)
it's lacks some clarity, and i don't think at compress enough at it's highest sustain setting.

what should i do next? short the pot's wire to the shortest length i can?
also i noticed that the effect is much clearer when i'm inside the "bump" of volume, but maybe that just because it's louder...
in maximum sustain setting i need the level to be about 60% to match the bypassed volume

[R.G.]

Just for fun, connect a 0.1uF ceramic capacitor directly between pins 7 and 4 of your CA3080, and a 100pF ceramic capacitor between pins 2 and 3. What happens now?

[xfc]

tried the 0.1 cap between leg 4 and 7 it didn't help (it's suppose to shunt high freq noise from the power supply to ground doesn't it?)

but, i found a failing cap - C11, i must've some how internally broke it so it was open.
replaced it and now the noise is some how quieter...

i think i can say for sure that the effect sounds way better at the end of the sustain knob, and still not as "squashy" as you would expect.
again when turning down the knob i'm getting a volume drop and almost no effect.

so i have the same situation only now i'm convinced that that area where the volume is raised is where thing working properly...

oh, and i didn't tried the 100pF mod yet, i would try it , i just don't have the value and it's hard doing it with two caps

[R.G.]

The exact value of 100pF is not critical; try 47pF or 220pF, or even 1000pF, whichever you have. In this case, bigger may be better, as you're trying to find out if using a cap like this will make it quit oscillating. The description you're giving me is still consistent with my guess about oscillating, so let's go with that until the guess is proven wrong.

[xfc]

well, i tried 47pF and two 47pF in parallel solderd directly to the legs of the OTL, it didn't work

i also took the jumper that connected to leg 2 of the trimmer and replaced it with a bit longer and from insulated wire and tried moving it around didn't help it.

i hope my English is not a barrier causing a misunderstanding so i will try to explain my self again

i get a diffrence at the volume between settings of the sustain toward the end of it cycle, and the rest of it cycle
the effect is much louder at the extreme. the effect sound better, but not as drastic as it should
when i sweep the sustain knob i reach to a point where i get that noise which sounds  like someone rubbing a plastic beg and them the volume goes up.

i hope that i don't confuse to much, i'm just trying to be detailed as possible

[R.G.]

No, you're not being confusing, I can understand you OK.

However, I'm at the end of what I can do remotely by typing. This is one problem with perfboard and veroboard builds - I can't see where your conductors run. I believe you have either a connection/wiring mistake that's causing the oscillation or a layout issue that's letting it happen. But not being able to see the board and test it with an oscilloscope, I can't go any further.

All I can do from here is suggest that you recheck it with an ohmmeter testing for connections from every point to every point, marking a printed copy of the schematic as you go. It is possible that one of your parts is broken/burned/etc., but that can't be told remotely either. Super-resolution closeup photos might help.

Sorry - this is one reason I went to PCBs exclusively a couple of decades ago and did not seed the effects making community with information about perfboard and strip board. These last can be very, very difficult to debug.

[xfc]

i might take it to a lab where i can put a signal generator in the input and measure it with a scope at different point, and i'll try to look it up.

i guess that in the end i will just rebuild it on a PCB, the only reason i didn't is that i can't buy the chemicals online
and the only shop i know to sell them is 2 hours trip in each direction from where i live.

i will repost success/failure, probably in two weeks or so.

thank you very much for your help, i hope that my next post here will be success...   

[R.G.]

I wish I could have given you more help. The vast majority of problems can be solved with DC-only measurements. But there are some where it requires more advanced equipment to find the flaw.

If I may suggest, is it possible for you to order PCBs by mail? I do know that the dynacomp/ross compressor from General Guitar Gadgets works, and it's not much more expensive than buying chemicals and board stock. It also eliminates the mechanical processes of cutting, resisting, etching, trimming and drilling. It's not possible for some people to buy PCBs, but if you can it will eliminate several sources of problems for you.

[xfc]

hello,
finally, i had some time and make it work

i etched a board and soldered the parts, problem wasn't solved.
debugged it a bit and found two cups inverted, corrected that. after that i found two traces on board that was broken and fixed them

now it works great.

thank you very much for your assistance.
David