OD-1 build completely baffled (again)

[jhaneyzz]

I have built a board using this plan from tagboardeffects:



But my output is really low and accompanied by a sound like whooshing wind.

I am currently walking through the board with an audio tester (first time using this tool but it "seems" really useful" while consulting this schematic:



But I'm crap at reading schematics like this when they break the IC into two separate components. totally messes with my head and I've no idea what I'm supposed to be experiencing.

1. Works fine in bypass.
2. output and input are definitely correct
3. I have re-flowed the entire thing 3 times and checked the paths between strips with a magnifying glass and a DMM.
4. I have rechecked the values of every component
5. I have swapped in 3 different 4558s and two sets of 2n2222 and 3n3904s
6. Confirmed the double link is connected
7. Checked all the cuts and jumpers
8. Replaced the foot switch because why not?
9. Voltage and polarity (at least at the power jack) are correct as well.

Any ideas?

[Mark Hammer]

Well that's the wrong schematic.  The OD-1 has no Tone control, just a drive and a volume knob.
There are two versions of the OD-1.  The first uses an RC3403 quad op-amp.

The second issue uses a 4558 for the gain/clipping sections and filter, and transistors for the input and output buffers.

The electronic switching is not obligatory.
Does the quad op-amp version sound "better"?  I don't know.  Would a humble LM324 yield the same tone as an RC3403?  Again, I don't know.  What I do know is that I built one for myself with a 3403, and I like it very much.  The filtering provided by the 18nf cap in the "tone" stage is just right.  I have to note that I added a 3-position DPDT toggle to provide stock rolloffs, add more bottom, or add more bottom plus less top.

[jhaneyzz]

Ive kinda been figuring that out since my post.

I haven't found a schematic that matches the strip board layout.

I think there was at least one iteration of the OD-1 that was 4558 based before the tone knob version.

hmmmm...

[idy]

Post voltages for all pins of opamp and transistors.

the path is a little hard to follow for already known reasons.
The audio goes in Q1 base, out the E, sounding the same
In the opamp pin 3, out on 2 (and back in 1)
then it goes to the second opamp, in the  - input, pin 6,
It should come out pin 7 (and won't be audible on pin 5)
goes through level pot and into B of Q2, and out E.

Check these places, the sound dies somewhere, where?

[antonis]

Signal across items (resistors, capacitors, diodes) flows through them..
Ignore vertical crossing items..

[ElectricDruid]

That diagram is a good indication of why trying to debug stripboard layouts is a b!tch!! 

[duck_arse]

I have built a board using this plan from tagboardeffects:

........

9. Voltage and polarity (at least at the power jack) are correct as well.

Any ideas?

10. post photos of what you have built.

[jhaneyzz]

Thanks all for jumping in.


photo of tag board.




Voltages of 4558 pins.

[antonis]

What's weird enough is almost 20% Vref "offset" upwards whereas your 33k resistors are rated at 1% tolerance..
Also, Q1 Base bias resistor looks like 4k6 one (instead of 470k) which it should load Vref (lower than 4.44 V)..

Maybe it's too late here..

[jhaneyzz]

well, apparently the soruce is near that frankenstein "stack-o-caps" i built bridging pins 6 and 7 of the 4558.

I swear to god my DMM showed 18ish nf when i built it, but I unsoldered it and by multimeter with the little hook probes seems to suddenly be giving me wonky readings, even with caps of known values. So' I'm having a bit of a time figuring that out.

In the meantime I plugged in a 15nf cap which seems like a major step in the right direction but is still way too quiet.

FWIW, the audio probe tells me that everything is fine on the left side of the 4558, all the way though the diodes it all sounds perfect.

[antonis]

the audio probe tells me that everything is fine on the left side of the 4558, all the way though the diodes it all sounds perfect.

Your right side DC voltages are almost the same with those on the left side which means it should also sound perfect..

Delete 15nF cap (take it out of the board), remove any solder residuals in the area and try again..
(don't worry about cap removal - it's just there to form a LPF for the unity gain right side amp..)

[jhaneyzz]

BINGO....

As you mentioned, without that cap at all it sounds fine. I put in a temporary socket in its place and tried a few different caps and stuck with the 15nf.

Sounds amazing!!!

That signal path diagram was soooo helpful I can never seem to decipher what all of the "extra" components do.

The audio probe is such a great tool as well

The clouds are slowly beginning to lift.... I might actually be learning something.

I have a box full of 99.99% completed pedals that i abandoned after not knowing enough to troubleshoot them. Now I have a holiday project.

big thanks antonis.

p.s. For the time being, Im ignoring your comment "What's weird enough is almost 20% Vref "offset" upwards whereas your 33k resistors are rated at 1% tolerance..
Also, Q1 Base bias resistor looks like 4k6 one (instead of 470k) which it should load Vref (lower than 4.44 V).. for the time" since it is WAY over my head still.

[FiveseveN]

For the time being, Im ignoring your comment

You shouldn't though. Just make sure this resistor is 470K:

[jhaneyzz]

I checked it. it is a 470k. (466k to be exact)

Oh. and going back over my previous cap-sandwich mistake... I have apparently provided evidence that there actually IS a difference between 18uf and 18nf... so I never have to wonder about THAT anymore...

[PRR]

....I can never seem to decipher what all of the "extra" components do. ....

http://www.geofex.com/circuits/what_are_all_those_parts_for.htm

[jhaneyzz]

That is very helpful and enlightening thank you.

[antonis]

Oh. and going back over my previous cap-sandwich mistake... I have apparently provided evidence that there actually IS a difference between 18uf and 18nf... so I never have to wonder about THAT anymore...

You realized (the hard way) that 18μF cap in place of 18nF one lowers down X1000 the LPF corner frequency..
Just place 10k for Rf and 18μF (instead of 18nF) for Cf in R.G.'s formula posted above by PRR..

[duck_arse]

two points I'd make - DON'T franken. use a single part, R or C, somewhere near the correct value. when the circuit works, you can worry about the value. and photos, always, first step. well, top 3 steps.

[Steben]

The filtering provided by the 18nf cap in the "tone" stage is just right.

Just wondering .... a TS has treble filtering in the clipping stage. Here we see filtering after the clipping.... does make a difference

[antonis]

a TS has treble filtering in the clipping stage. Here we see filtering after the clipping.... does make a difference

Here we have LPF fixed (constant) corner frequency..
(in TS it's variable according to Drive/Gain pot setting..)