MXR Dynacomp, FET input mod?

[nooneknows]

Hi,
I've bought a used MXR Dynacomp. I didn't know it wasn't a true bypass and I didn't notice in the shop the little high treble loss... anyway I serched for the schematics and I saw a BJT as a first component.
What about changing this one to a FET, creating a high impedance buffer? I think this could reduce the hi band loss a bit, leaving the bypass switch as it is.
It could be done  shorting the input cap and the 10K resistor, then connecting the 1M Resistor from the base of the tranny to ground and not to V+, I believe a 2n5457 or a bf245 (keeping an eye on the pinout) could be ok.
Does it seem correct?
thank you,
M

[Mark Hammer]

The treble loss is a consequence of what compression does to the relative high/low balance.  Improving the input impedance will not fix that, sadly.  What you CAN do is to introduce some treble emphasis in the output to offset that loss introduced.  Stick a cap in parallel with the 10k resistor before the output level control and that will improve matters.  Tastes will vary, but .01uf is a decent place to start.

[markm]

Not for nothing but,
I always felt the Dyna was a bit brighter than other similar comps out there?

[davebungo]

I think the question refers to the unit when in bypass mode.  i.e. the input appears to slug the signal a little.  A FET input definitely would help in this.  Personally, if I were doing this mod, I'd just insert a little board as an extra and drive the existing circuit or possibly easier still, mod it for true bypass.

[nooneknows]

I think the question refers to the unit when in bypass mode.

Exactly, maybe I didn't explain well my problem. I hear a treble loss even when the pedal is bypassed and I bet this comes from the pedal input buffer.
I think it's not that easy to connect a newer board, it is also necessary to cut some trace on the pcb, I thought a simple change of transistor could be enough. The only doubt I have is if this mod, considering a source follower has a slightly higher output impedance than an emitter follower (and a input cap should be removed), could influence the compressor function.
Thanks,
M

[Mark Hammer]

 

No, you explained yourself well enough.  I just didn't read closely enough.  Why not just install a DPDT or 3PDT switch?  if you decide to sell it as a "vintage" piece, just reinstall the original switch.

Note that in those effects which normally lack any sort of volume adjustment capability (e.g., Small Stone phaser, wah), replacing SPDT with DPDT switches can produce volume-level discrepancies due to loading.  So your interest in a FET-buffer to offset the loading is in the right direction.  But if you have a volume adjustment which can offset any level discrepancies between effect and bypass, installing a true bypass with DPDT/3PDT switch is the easiest and probably least invasive approach to take.

[lovric]

hi,

dont cut any traces. the transistor is an emiter follower (no gain - you know that). just stick a jFET in its place. source -> emiter, gate -> base & drain -> collector.

the jFET can also be in source follower configuration with resistor divider bias network as here is the case (existing one gives you around 3V). you could even put a MOSFET because of that.

but, the ordinary bipolar transistor in the emiter follower configuration also has a big input impedance.

when you implant the jFET you might want to do the first suggestion by hammer. nail the right chromozome and fix the genome.

[bwanasonic]

Why not just install a DPDT or 3PDT switch?

If this is one of the Dunlop RI pedals, installing a 3PDT is possible, but a major PITA. My standard advice is to build a bypass loop. Very handy to have around. When I am testing a breadboard circuit, I put it in the loop, and it provides a convenient way to test the bypassed signal vs. the effected signal. A Hammond BB houses two bypass loops nicely.



Kerry M

PS I seem to remember the cap mod M Hammer mentions (which I performed on my DC) also effects the bypassed signal?

[Mark Hammer]

PS I seem to remember the cap mod M Hammer mentions (which I performed on my DC) also effects the bypassed signal?

That makes no sense to me, given that the cap precedes the output pot.  Possible this is having to do with someone else's mod to the input?

Sounds as if nooneknows has an older one, rather than recent Dunlop issue, given the SPDT switch.

[nooneknows]

It's a 2nd hand piece, I don't know if it's a new one or an old one. Anyway there is a spdt switch in and I'm short of dpdt so the fet buffer seemed to me the quickest way to achieve a good bypass. I'll work around a bit following yoour advices and I'll let you know.
Thank you very much for your help,
Marcello