DEBUGGING HELP: Ampeg Scrambler Clone

[Kinetic]

Hi folks, I have built an Ampeg Scrambler clone based on the General Guitar Gadgets schematic (which can be found here).
The pedal seems to work well when the texture control is set to minimum.  However, the higher up the texture pot is turned, the more the signal is squashed.  It's difficult to describe, so I have uploaded an MP3.  The same musical sequence is played three times: once with Texture at zero, once with Texture at 12 o'clock, and finally with Texture at full.  The blend control is maxed in all three examples.  MP3 clip is here.  All three are recorded one after the other into a PodXT on the clean preset.  The difference in volume is from the Scrambler, as is the kind of "fade in" effect.

MODIFICATIONS TO SCHEMATIC:
1) I have used a 2N5306 for Q1 and a BC169B for Q2 as directed, but I have used MPSA14s for Q3 and Q4.  I have checked my pinouts, Q1 and Q2 have the collector as the middle pin whereas Q3 and Q4 have the base as the middle pin (in what I consider to be the standard CBE configuration).
2) I have used a 3PDT switch instead of DPDT, with the extra pole for an indicator LED.

No ground conversion.

Q1
C = 9V
B = 3V
E = 3.35V

Q2
C= 2.75
B=0.9
E=0.3

Q3
C= 8.9
B= 4
E= 3.4

Q4
C= 8.5
B= 4
E= 3.4

D1
A (anode, the non-band end) = 0.3
K (cathode, the banded end) = 0.9

D2
A = 4
K = 3.7

D3
A = 3.7
K = 4

D4
A = 4
K = 4

D5
A = 4
K = 4

Any thoughts?

[Jarno]

Hi folks, I have built an Ampeg Scrambler clone based on the General Guitar Gadgets schematic (which can be found here).
The pedal seems to work well when the texture control is set to minimum.  However, the higher up the texture pot is turned, the more the signal is squashed.  It's difficult to describe, so I have uploaded an MP3.  The same musical sequence is played three times: once with Texture at zero, once with Texture at 12 o'clock, and finally with Texture at full.  The blend control is maxed in all three examples.  MP3 clip is here.  All three are recorded one after the other into a PodXT on the clean preset.  The difference in volume is from the Scrambler, as is the kind of "fade in" effect.

MODIFICATIONS TO SCHEMATIC:
1) I have used a 2N5306 for Q1 and a BC169B for Q2 as directed, but I have used MPSA14s for Q3 and Q4.  I have checked my pinouts, Q1 and Q2 have the collector as the middle pin whereas Q3 and Q4 have the base as the middle pin (in what I consider to be the standard CBE configuration).
2) I have used a 3PDT switch instead of DPDT, with the extra pole for an indicator LED.

No ground conversion.

Q1
C = 9V
B = 3V
E = 3.35V

Q2
C= 2.75
B=0.9
E=0.3

Q3
C= 8.9
B= 4
E= 3.4

Q4
C= 8.5
B= 4
E= 3.4

D1
A (anode, the non-band end) = 0.3
K (cathode, the banded end) = 0.9

D2
A = 4
K = 3.7

D3
A = 3.7
K = 4

D4
A = 4
K = 4

D5
A = 4
K = 4

Any thoughts?

I am no expert, but my Scrambler worked off the bat, looking at Q1 and Q2 I think those aren't biased correctly, maybe because of variation in parts. I don't recall what I did to bias all transistors correctly, I used totally different ones, I think BC550C's and two piggybacked for Q3 and Q4. I may have stuck in a trimpot to dial them in. Maybe you could do the same, to get the base voltage to about halfway between collector and emitter?

[Kinetic]

I think I took the alternative transistor suggestions from the tonepad layout, where it suggests the ones I've used.

If my voltages at Q3 and Q4 are okay then does that suggest that my problem part(s) is in the first half of the circuit?
Also, if the voltages themselves at Q1 and Q2 aren't right, does that mean that it's likely to be the other components (not the transistors) which are the issue?

[Kinetic]

*bump*
Anyone else got any ideas?  This one's really frustrating me.  I'm not really a massive effects user, but I like an interesting effect and I'd really like to get this one working.  Like I say, seems to work nicely... until I turn the texture knob.

[Kinetic]

I'm going to bump this up again - I put this one to the side and have now come back to it, but I'm still having the same problem.

I've analysed the issue a bit further - it seems to be related to the voltage at the base of Q4.  The circuit seems to work generally, but works best when the texture control (i.e. the 10k linear pot) is set to zero.

I'm looking at the schematic and specifically, at the voltage divider between the texture pot (R10) and R11.  When the texture pot is at zero, the voltage at the base of Q4 is just over 4V.  When the texture knob is maxed, the voltage at Q4 base drops down to around 0.15V.

Sonically, the output sound is pretty good with texture at zero.  Up to approx half-way, the output drops slightly in volume, but is still fairly good.  Once it gets above half way and the effective resistance of the pot is more than about 5k, the output sound has that "fading in" effect as you hear on the MP3 I posted in the original post.

I'm not really sure what to do with this now, I can't seem to find any logical reason for this problem and I haven't encountered anyone else having similar problems through searching.  Since the problems really kick in when the pot's effective resistance is over 5k, I wonder if replacing the 10k pot with a 5k would give an adequate result?

[Kinetic]

Well, I've had some success with this now, so I'm going to post here just in case anybody else ever has the same trouble I've had.

In my previous posts, the texture control was working fine when set to zero (i.e. effectively setting the variable resistor at 0 ohms) but was not working when turned up.
With reference to the previously linked circuit schematic, I replaced the texture pot (R10) with a jumper, effectively fixing its "resistance" at 0 ohms.
I then socketed R11 to do some experimenting.  I found that with the 8k2 resistor socketed, my output signal was equivalent to the sound of the Ampeg Scrambler with texture set to zero.  As I increased values of R11 up, the output starts to fuzz up a lot more and sounds like the effect you get when you turn up the texture knob.
I also noticed that if I leave the sockets empty (i.e. remove R11 and leave an open circuit here) then I get that really wacked out synth-esque Scrambler tone.  I suppose that this is equivalent to using an infinitive value for R11.

So, taking all of that into account, I have:
- removed the original texture pot R10 and replaced it with a jumper.
- replaced R11 with a 1M pot wired as a variable resistor (this is my texture control; I've used 1M because it was the highest value pot I could pick up).
- added in a SPST toggle switch which, when engaged, removes the R11 pot from the circuit - for those times when you want to turn the texture knob past maximum!

I'm not 100% sure why I've had to make these modifications given that so many people seem to get a good result from this pedal right from the word go.  But ultimately, I've ended up a pretty cool-sounding effect, and I feel quite smug for having successfully troubleshooted and modified it, so I'm not too worried.