Buzzaround/Elka Dizzy Tone gating issues

[amptramp]

The bias on the output transistor stage is generated by leakage through the coupling capacitor and voltage generated by rectification of the AC signal coming in from the previous stage.  There have been a number of transistor radio amplifiers that were biased by leakage, but these radios did not work at low temperature due to the reduction in leakage.

If you rectify enough signal to create a strong bias tending to turn the output transistor on, you will have a time constant where the coupling capacitor has to charge to this voltage, so you may get blocking distortion (momentary loss of output, also known as gulp distortion).

The bias will change with the setting of the sustain control, so not all of the control sweep will be useful.  After all, at the zero setting, there will be no bias on the output transistor at all.

[KingSG55]

Thank you for the valuable input/response. So now we know what the problem is.

The question is, can it be fixed somehow without changing the sound too much?
Thanks guys.

[Rob Strand]

Thank you for the valuable input/response. So now we know what the problem is.

The question is, can it be fixed somehow without changing the sound too much?
Thanks guys.

My apologies for the late reply.  I've been out of action for a few days and probably won't be back to full working order for another week.

There's plenty of things you can try:

Based on the assumption you need more leakage, one or all of these might work:
1) Large value resistor between the base and collector of the last transistor.
2) Large value resistor between the base of the last transistor and the supply rail ("-9V").
3) Large value resistor between the top node of the sustain pot and the supply rail ("-9V").


In each case a different resistor value will be required.  (2) and (3) might need a 10M if you need more than 10M you have to fiddle about with series resistors. Solution (1) might allow more favorable and smaller resistor values.  Find the largest value resistor that works. Then perhaps use a one or two standard values less than that as a safety margin - but only if it stays working!  You might also find solution (1) behaves a little different to solutions (2) and (3).  Solution (1) might be more stable in the long run.

There's a chance the above recommendation is completely wrong and you need to shift the bias "the other way".  If so, you need to add a resistor across the base and emitter of the last transistor.  Much less than 10M though.  I have doubts you will need to go down this route as the 100k across the pot you tried earlier didn't work.

Got nothing to lose.  Once you get on the right track towards a fix you only need to spend some time tweaking the resistor value.

[amptramp]

You may need a series resistor from base to pot slider so the base side can be pulled into conduction even when the pot is at zero.

[mac]

1) Large value resistor between the base and collector of the last transistor.

+1

You can try a low gain silicon like 2n2369, mpsa42, bd139, etc, 70<hfe<150 and hear what happens. And 100pf to 220pf across the resistor to tame highs.

You may need a series resistor from base to pot slider so the base side can be pulled into conduction even when the pot is at zero.

+1

I posted a Silicon version using these two mods,
https://www.diystompboxes.com/smfforum/index.php?topic=111831.msg1267499#msg1267499
It might give you some ideas to fix your build.

mac

[mzy12]

The bias on the output transistor stage is generated by leakage through the coupling capacitor and voltage generated by rectification of the AC signal coming in from the previous stage.  There have been a number of transistor radio amplifiers that were biased by leakage, but these radios did not work at low temperature due to the reduction in leakage.

If you rectify enough signal to create a strong bias tending to turn the output transistor on, you will have a time constant where the coupling capacitor has to charge to this voltage, so you may get blocking distortion (momentary loss of output, also known as gulp distortion).

The bias will change with the setting of the sustain control, so not all of the control sweep will be useful.  After all, at the zero setting, there will be no bias on the output transistor at all.

Huh I know this is kind of off topic, but I think I'm getting this on my Tonebender MkIII build. (aion fx kit from musikding) Should I try some of these latest suggestions myself?

[mac]

Huh I know this is kind of off topic, but I think I'm getting this on my Tonebender MkIII build. (aion fx kit from musikding) Should I try some of these latest suggestions myself?

What transistors are you using?
Did you measure Q3 collector voltage?

mac

[mzy12]

What transistors are you using?
Did you measure Q3 collector voltage?

Unfortunately I don't have a way to test leakage/HFE. I got the kit from Musikding and they provide the appropriate transistors, marked according to their position in the circuit and tested beforehand.

Q3 Collector voltage is -3V. There's a trimmer on the Phobos kit that you can use to set the collector voltage appropriately.

[KingSG55]

Hello everyone. I have some bad news. Again, I have tried everything that you suggested and I'm still getting the problem.
Somehow I think it's really transistors ands that's it, I need a matched set.
Can some of you guys sell me a matched set for it? We can talk price and everything, please.
I just can't believe I can't make it work, it's driving me nuts, also made 6 tone bender MKIIs and 4 MKIIIs in the meantime. All of them work perfectly. But Elka is the best one I have ever played, I really want it.

So, anyone have a matched set? I ordered few from musikding but they only have AC125s (which have this issue).

[KingSG55]

What transistors are you using?
Did you measure Q3 collector voltage?

Unfortunately I don't have a way to test leakage/HFE. I got the kit from Musikding and they provide the appropriate transistors, marked according to their position in the circuit and tested beforehand.

Q3 Collector voltage is -3V. There's a trimmer on the Phobos kit that you can use to set the collector voltage appropriately.

Wait a second, you have the exact same issue? :O
Transistors from Musikding?

[mzy12]

Yep I do! Transistors also from musikding... not sure what's going on tbh. Considering making my own thread with voltage measurements out the wazzoo... tried using a 10M resistor from collector to base on Q3 but I think that only helped marginally. Gonna try two 4M7s in series next.

I also recently got in a couple batches of Russian Ge transistors, unsorted. Don't have any way of measuring their hfe for now. But I can try swapping some in to see what happens.

[duck_arse]

I also recently got in a couple batches of Russian Ge transistors, unsorted. Don't have any way of measuring their hfe for now. But I can try swapping some in to see what happens.

do you have a dmm? if so, build the RG Keen tester, even just with "near" resistor values. you can then say leaky or not, higher or lower gain.

[mzy12]

do you have a dmm? if so, build the RG Keen tester, even just with "near" resistor values. you can then say leaky or not, higher or lower gain.

Yep, I do. I really should just build one. Need to order some stripboard though as I've used all mine up already.

[KingSG55]

Hello there pedal building friends! The issues is still not fixed after all this time. Re-build the circuit a few times and can't figure it out. Spent 100 EUR on matched set of Soviet transistors (have enough for 4-5 pedals) and not of them worked - so the problem must be somewhere else. I still don't get how the problem goes away immediately if I just touch the capacitor leg that goes directly to Sustain pot. Should I add a resistor between the cap and the sustain pot?

Thanks.

[KingSG55]

One more thing to add - voltages.

Q1 - 3.5V
Q2 - 3.5V
Q3 - 1V

With everything on max. If I drop the sustain pot a bit the Q3 will end up on 0.5-0.6V.
I guess still wil proper leakage and proper transistors I cannot get it to bias properly. Ah.

[mac]

Sometimes the problem is "outside" the pedal.
This week I went crazy with a mini amp I have on the breadboard. The problem was the DC jack not making good contact.
So check for ground issues, filter caps, jacks, guitar cables, negative and positive ground living together, the weather, planetary aligment, etc.

Spent 100 EUR

I've just paid usd 100 on property taxes and expenses.  .. Yours is money well spended!! 

mac

[duck_arse]

post photos. why not?

[KingSG55]

Sometimes the problem is "outside" the pedal.
This week I went crazy with a mini amp I have on the breadboard. The problem was the DC jack not making good contact.
So check for ground issues, filter caps, jacks, guitar cables, negative and positive ground living together, the weather, planetary aligment, etc.

Spent 100 EUR

I've just paid usd 100 on property taxes and expenses.  .. Yours is money well spended!! 

mac

Terrible to hear that 

I've checked everything else.

[KingSG55]

post photos. why not?

Absolutely, making some photos tomorrow!

[KingSG55]

Can you maybe spot a problem in this vero layout?