Aquaboy's Buffer

[delus]

Hi!
Has anyone noticed the following?
The Aquaboy (V3205 version) when on cuts some of the high frequencies of the dry signal. This is easily audible when using a fuzz face before the delay and cranking the gain, but it's also audible on a clean signal too, only a bit more difficult to hear.
I noticed it yesterday, so I took out my Carbon Copy to see if the same thing happened, but it didn't. It made the dry signal sound just a tiny bit louder. So I figured it must be a buffer thing.
Next I put the RC Booster before the Aquaboy, set to unity gain so it would act like a buffer instead of a booster, and it worked. When the Aquaboy was turned on it no longer cut any high frequencies from the dry signal.

So does anyone have an idea how we could improve the input buffer on the Aquaboy? Could we substitute the MPSA18 for another transistor? Or anything that would help keep the dry signal as "clean" as possible?

Regards,
Adam.

[gtr2]

Hmm...  I've built two aquaboys and haven't heard this problem in my own builds.  Did you make any substitutions on the BOM?

Josh

[delus]

None, only R19, which I used what I had handy, I don't remember excactly somewhere between 1.2K to 1.8K, in order to get my voltage around 7-8V. I don't suppose expirementing with the trimmer pots would make any difference on the dry signal, would it?
I made a brief search and found some transistors with the same pinout as the MPSA18. What if I substitute Q1 with 2N3904, or 2N5088, or even BC108?

[hoyager]

Hey, C35 seems to have the most affect on the highs cutting from dry signal.

I have around 1n there which only just cuts very highs, I think, standard is 100pf.

Lowering C1 (470pf) might help too?

The trimmers do adjust the bbd voltage a little bit (.20v or so)

I have a 1k resistor to the striped end of a 7.5v 1w zener diode (other end goes to the ground pad there) then a 1n4148 to the remaining pad, where R19 would usually be. Striped end of the 1n4148 goes to that pad which bleeds of some more voltage so the chips get 7v.


          (R19)
           
*---1k-- --1n4148 | --*
            |
           --
      7.5v zener   
            |
            |
        ground

Andy

[delus]

I'll try and experiment with the value of C35 and C1. But why do you suggest those modifications, hoyager? Did you actually notice a similar issue like me? Also does fine tuning the bbd voltage have any impact on the frequency response of the dry signal? I thought it would affect only the repeats (wet signal).

I have also noticed that on certain complex effects, even the wiring is critical and has an impact on the overall noise/hum. Perhaps there's something about my wiring that cuts some of the high freqs. Maybe a gut pic would help?

Thank you very much, people!

Regards,

Adam

[hoyager]

Sorry, no I haven't noticed any problem with the stock values, even raising C35 to 1n only just makes a difference for me. Is it possible you might have the wrong value in 1 of those 2 spots?

Can't tell otherwise unfortunately.

Andy

[jkokura]

I don't see anything wrong with your build. In fact, it looks pretty dang good. Because you used board mounted pots that makes it unlikely there's any problem with your wiring.

Jacob

[aflynt]

I've noticed that the high end is attenuated also when engaged, but I kind of like that smoothing effect. Perhaps it's just an input impedance issue if the buffer fixes it. I'm running all true bypass effects and usually use it after a Serendipity or Klone so maybe that's why I notice it too.

-Aaron

[delus]

Andy, I'm pretty sure I got the right values. I triple check at least, before soldering a component, and that's one reason why every build takes too much time for me to finish.

Jacob, thanks a lot, man! I just thought maybe the wires should be a shorter, or maybe the output should have been grounded with a wire instead of directly to the enclosure.

Aaron, it is kind of cool the way it smoothes the fuzz face, but the cleans I want them nice and bright! Well, with more than three pedals in the chain a buffer is always a must, so it won't be much of a problem anyway.

Thanks again,

Adam