Aquaboy Deluxe Calibration and General Questions

[peAk]

So I recently went back to my (2013) Aquaboy Deluxe (MN3005 & Road Rage) to see if I could maybe make some calibration adjustments and possible others modifications to make me like it more. I first tried calibrating by ear and I am currently using a scope.

One of the things right off the bat that I never liked was that I start to get self oscillation just a little passed the halfway mark on the feedback pot. That's even with T1 full CCW. Is there something I can experiment with adjusting so that I can only get self osc with the feedback at full CW?

I also tried running the clock at 6.8Khz and 6.5khz and I just still feel the repeats are just too degraded. Is it possible to run the clock even lower for cleaner repeats? I know that I will sacrifice delay time and I am okay with that if thats what it would take. Whats the lowest you recommend?

Reading through all three versions build docs, I see that in the latest revision Brian mentions this in the nerd notes:

Even though the circuit is running at 15v, I decided to leave the output voltage of the compressor
portion at the standard 3v as used on the DM-2. If you built older versions of the ABDX, you may
remember I suggested that if you wanted to run it at higher than 9v to change the voltage divider
resistors (here R15 and R16) to a larger value to increase the voltage output (pin7) of the compressor
(the idea being that a larger voltage there would increase headroom and also the bias voltage of the
pre-emphasis filter that follows).
After doing some more listening tests on the 2019 version, I ended up preferring the lower headroom
you get from the stock 10k resistors. To my ears, the higher voltage you get from an output voltage
of 5v or more ends up making the repeats too percussive and they have a tendency to sound more
"dithered" after 6 or 7 repeats. I did compensate for this in the expander portion of the compander
by including the 33k resistor on pin12 (R38). This bumps the output of the compander up about one
extra volt. This seemed to have no negative effect on the repeats.
If you want to tinker, then I suggest socketing R15, R16 and maybe even R38. Higher values for R15
and 16 will increase the output at pin7 of IC2. Lower values of R38 increase the voltage output at pins
10 and 11 (at low values it may stop operating). 

Is this something I can mod with the 2013 version? If so, how would I go about doing it? I feel, just like Brian mentioned above, the repeats right now are just too percussive. Don't get me wrong, I prefer analog delays and the darker, filtered repeats but I just feel right now this pedal is a little harsh, especially compared to the Total Recall.

Any advice on this would be appreciated.

[danfrank]

Yeah, see what those 2 resistor values are in the 2019 version (R15 & R16) and look for them on the schematic of the 2019 version. Then go to the schematic in your 2013 version and locate those 2 resistors in the circuit and change them out to the values used on the 2019 version.
Then change out the resistor going to pin 12 of the compander on yours with a 33k resistor

[peAk]

Yeah, see what those 2 resistor values are in the 2019 version (R15 & R16) and look for them on the schematic of the 2019 version. Then go to the schematic in your 2013 version and locate those 2 resistors in the circuit and change them out to the values used on the 2019 version.
Then change out the resistor going to pin 12 of the compander on yours with a 33k resistor

yeah, good point. I mapped them out on both schematics and will fiddle with trying some different values.

What about the feedback issue? you, or anyone, have an idea how to mess with this? It it normal to get self oscillation right after "noon" setting?

[danfrank]

If you're using 15 volts on yours, I'd try what the AQB DLX 2019 uses as a starting point.
Same thing goes for the feedback circuit. There needs to be more series resistance (bigger value T1) in your circuit.

[Scruffie]

The MN3005 gain isn't exactly 1 and multiplies with each BBD, reducing the value of R23/24, 29/30 to 47k (Edit: those are the part numbers on the 2019 version) will probably solve the feedback issue.

Higher clock will mean cleaner repeats, run it at 10kHz for 400mS and good fidelity.

[peAk]

The MN3005 gain isn't exactly 1 and multiplies with each BBD, reducing the value of R23/24, 29/30 to 47k (Edit: those are the part numbers on the 2019 version) will probably solve the feedback issue.

Higher clock will mean cleaner repeats, run it at 10kHz for 400mS and good fidelity.

Duh. Higher. Brain Fart.

I will give these suggestions a go and report back.

Thanks for the tips.

[peAk]

Just a quick update....

I lowered the R21/R22 & R28/R29 (2013 version) from 100k to 47K and that completely fixed my feedback issue. It actually allows T1 to actually have some flexibility of where I want the self oscillation to start depending on where my feedback is set. Which I wanted to to happen almost full CW. So that's great!

I changed the clock to 10kHz and changed the voltage divider resistors R13/R14to 10K and r37 to 33k and that seemed to clean things up a bit. At least to the point I am happy with it now.

Is it fair to say that this is just a much more lofi/dark/percussive delay compared to the DMM? 

[danfrank]

Yes, the DMM is about as good as it gets for an analog delay (the DC Echo is even better). It's mainly because of the component values used for the filters, etc., but yeah, the AQB/DM2 is kind of percussive sounding.
The AD900/Man O War is kind of on the other end... It's hard to get it to oscillate

[peAk]

Yes, the DMM is about as good as it gets for an analog delay (the DC Echo is even better). It's mainly because of the component values used for the filters, etc., but yeah, the AQB/DM2 is kind of percussive sounding.
The AD900/Man O War is kind of on the other end... It's hard to get it to oscillate

thanks for the info, thats good to know. I've never owned originals of either pedal so I have no clue. I have an Ibanez UE-405 rack unit with an Analog Delay. I assume that the same as the AD-80? I've never opened it up to look at the circuit. It's a pretty clean/tame analog delay, though.