The pot value and type is the only really contentious area I would say. Since it affects both the max bandwidth and the accuracy of channels matching across its rotation. But you specifically don't use any beside your DAC's internal.
There aren't any other parameter sensitive parts. It was one of my design concerns to avoid as many as I could. Only resistors are directly in the signal path and in the AC feedback path.
If you may be curious to try Z foils etc. for those, or to roll few preferably JFET-input type op-amps in the DC servo, that is up to you to explore.
But controversial for effect already as I asked Tham in the beta test stage who said he could pick at least the op-amps a little but I also asked Ammel rather recently given the opportunity he was exploring some for best offset reasons who said he could not pick his op-amps for sonic changes so there you go. Still your system is different and your mileage may vary.
There aren't any other parameter sensitive parts. It was one of my design concerns to avoid as many as I could. Only resistors are directly in the signal path and in the AC feedback path.
If you may be curious to try Z foils etc. for those, or to roll few preferably JFET-input type op-amps in the DC servo, that is up to you to explore.
But controversial for effect already as I asked Tham in the beta test stage who said he could pick at least the op-amps a little but I also asked Ammel rather recently given the opportunity he was exploring some for best offset reasons who said he could not pick his op-amps for sonic changes so there you go. Still your system is different and your mileage may vary.
I noticed a change when going from 10k pot (which has too much bandwidth, maybe dangerous) to 20k pot. Some internal wiring was changed as well which could influence things.
The 20k pot is more solid, more full and maybe, maybe less airy. Could be the source interacting with the pot though. It's not a big change but I feel it's a little different.
The 20k pot is more solid, more full and maybe, maybe less airy. Could be the source interacting with the pot though. It's not a big change but I feel it's a little different.
Thanks Salas for the info,
for pot I have a relay attenuator with constant output impedance of 9K but I'll try LDRs in the future.
Never liked Z foils, for that kind of money I could go for Rhopoint but for now I'm thinking of low ppm and higher precision for R1, R2. R4, R6, R7, R15. I'll try a small bypass for decoupling caps (MKP1837) for each rail or maybe change to Muse
Opamp rolling is something I leave for the future.
All in all, not very much to play with (I think I like it)
Nobody's perfectI' m not Savvas!
Last edited:
With 20K pot it has the scope check behavior of post #20 at Ohmic midpoint rotation value. That's enough worst case point speed (~1MHZ) for this circuit before risking squarewave edginess IMO.
How's your F5 respond by being driven at 5x now? (given your medium sensitivity speakers)
What does "Ohmic midpoint rotation value" mean? Does this mean at step 12 of a 24 step series attenuator of 20K value?
Thanks. nash
That would be at the 6db attenuation level, so the DCG3 will see 10K. Am I correct? My serious listening is between 22db and 30 db attenuation with a preamp gain of 3 which would mean that with a 20K pot the DCG3 would see about 1589R to 632R? Is this OK?
Thanks. nash
It's terrific. I used a 2K resistor so 5X-ish gain? Max volume before clipping is about 80% of volume dial, leaving some room for weak sources or low-volume albums. Really, 4X probably would have been enough. For headphones, 5X is a little too much gain for Grados but works well with AKG's. As a preamp with the F5, the added gain was a good choice though.
Compared to Pass BA-3 the DCG3 still sounds more refined, more precise and somewhat flatter and quicker. Bass is really quick, no overhang. Also, it feels very grainless. On some albums it can sound closed on top, other albums very open. Perhaps it is very true to the source. With my past preamps this was not as apparent. The transparency is really good, one needs to listen to a lot of different albums and music to understand.
The only negative is the quickness and speed leaves slower, simple music with a less romantic presentation than BA-3. But I believe DCG3 is more accurate and definitely more precise. Bass has a lot of overhang w/ BA-3.
Perhaps this is due to no feedback (BA-3)...
But I am enjoying the DCG3, for both headphones and preamp duties. Mine is single mono...I'd like to go dual mono in the near future.
I am talking about just the BA-3 front end as a preamp. I have never used the output stage (which is just a buffer, same as F4).
What I have is similar to this, but I have a jung super reg for regulation:
http://www.diyaudio.com/forums/pass-labs/258022-ba-3-preamp-build-guide.html
My F5 has 2 pairs of toshiba mosfets for outputs and a bit more bias (2.5A) and capacitance (200,000uf). So it's a bit more lush than normal. I have never tried any of the Aleph amps but i'd like to. Also, The TSSA amps and the First One interest me. Really, I'd like to have 50-75 watts vs the 25 watts the F5 is giving me now. But I can't seem to give up on having something Class A...hard to do 50-100 watts all class A without big mono blocks. My chassis right now is 5U.
I have played with class D too (mostly Tripath)...not bad but not at the same level. Hard bass, no air.