DIY Dynacomp Debugging help (please!)

[jjg]

Hi there, I've completed an MXR Dynacomp circuit (from Perf & PCB effects, http://effectslayouts.blogspot.co.uk/2015/04/mxr-dynacomp.html) and I'm getting the following problem: the effect works well when my GUITAR volume is low, but as soon as I wind the guitar volume over about 3, there is distortion as well as compression.

I've followed R.G.'s excellent debugging steps as best I could, and haven't yet been able to identify any problems, although clearly I've done something wrong.

Here are my voltages:

Battery: 8.84V

2N3904's:
         C       B       E
Q1   8.65   1.24   1.44
Q2   6.24   2.57   2.35
Q3   7.52     0       0
Q4   7.53     0       0
Q5   8.65   8.65   8.06

The voltages between base and emitter on Q1 and Q2 are both 0.6V.

IC (CA3080E):
1: x
2: 3.56
3: 3.57
4: 0
5: 0.63
6: 2.57
7: 8.65
8: x

I've switched all the transistors and the IC, but the problem persists.

(I think these voltages look consistent with those from other Dynacomp posts.)

Could anyone please offer advice about what the source of the distortion might be, and why the circuit appears to be working just fine when the guitar input signal is low?

Any help very gratefully received!

Best regards,

Joe

[Kipper4]

Welcome Joe

See reply 10 here


http://www.diystompboxes.com/smfforum/index.php?topic=114668.msg1064703#msg1064703

[jjg]

Thanks - so if I happened to get the chips dirt cheap on a slow boat, you think that could be the problem?   

[antonis]

2N3904's:
         C       B       E
Q1   8.65   1.24   1.44
Q2   6.24   2.57   2.35

The voltages between base and emitter on Q1 and Q2 are both 0.6V.

So,  1.24 - 1.44 = 0.6
and  2.57 - 2.35 = 0.6

resulting in: 1.24 + 2.57 - 1.44 - 2.35 = 1.2
         or in: 1.24 - 2.57 - 1.44 + 2.35 = 0

[jjg]

So ironically I have a PhD in maths but clearly am amateur in electronics.

I measured the table of voltages with black probe on ground and red probe on the leads, whereas base to emitter was just that. Is that not correct?

[antonis]

I measured the table of voltages with black probe on ground and red probe on the leads, whereas base to emitter was just that. Is that not correct?

Can't follow you but you shouldn't mind for black/red probes placing..

For DC, we usually place Red probe on point with higher voltage..

If brobes are reversed, we read the same value with a minus sign (-) in front of it..

So, if you kept black probe to GND and measured 1.24V on Base & 1.44V on Emitter of Q1 you had a V_BE of -200mV.
For Q2, you had +220mV..

None of the above V_BEs is close to 600mV (according to your statement..)

edit: Now I see what you mean and is absolutely correct..!!
(as correct as can it be: a-c=x, b-c=y => a-b=x-y.. )

[jjg]

Thanks again for taking the time to reply!

Any idea why I've got distortion in the circuit as well as compression?

[antonis]

As soon as we have a circuit schematic we'll be able to help you...

Meanwhile, you may read about it: http://www.electrosmash.com/mxr-dyna-comp-analysis..

Especially about Q1 buffer stage and Q2 phase splitter voltages..

[duck_arse]

can you post some photos of what you have built please?

[jjg]

Thanks. I used the perf board layout from here: http://effectslayouts.blogspot.co.uk/2015/04/mxr-dynacomp.html?m=1

I followed this layout through with the schematic from the Electrosmash site linked above, and it checks out (although there seem to be a couple of additional electrolytic caps in the perf layout).

With the following photo, there are a couple of minor differences from the perf board traces (near the bottom left), where I only board mounted one of the pots.

Please note I am a novice at soldering so all advice gratefully received. This is my third project.

The circuit 'works' and gives good compression when my guitar volume is very low, but when louder than 2 or 3 there is significant distortion. It doesn't sound bad, but that's not the point! It should be clean.

[jjg]

Apologies, I don't think I'm able to post pictures (do I need to get permission?) - here are links to the build so far:

https://drive.google.com/open?id=0BwUWfwo09NyFa1FMbGNTOU5FZGc

https://drive.google.com/open?id=0BwUWfwo09NyFRm96Z01DZnRrMlU

[duck_arse]

well now, it seems you have fitted a 20k trimpot instead of 2k. I don't know what diff this might make to the circuit .....

also, if those yellow-ish caps are tantalums, please make absolutely sure 100% certain that they are correct polarity. tants are marked for (+) while ordinary electros are marked for (-), and a backwards tant can really put a crimp on your day.

as for image posting, you need a link to an image (hence, ending in .jpg or .png or .gif), NOT the page it resides on (hence no suffix, or .html, or .php, or something). then you poke the picture of mona, and paste in your image url. you would need to keep an eye on the size of the image posted in-page, tho, because people will tune out if they have to scroll more than 2 clicks.

[jjg]

Wow very good spot - not sure how I missed that! I'll order a 2K replacement and see if that helps.

Thanks so much for taking a look and for all the advice