Thank you for the reply salas. What is the uniqueness of using AD823 for the servo? I believe what we use is to correct the DC offset right? so any Jfet input opamp with low noise should do the job like opa134 as such? Since its not amplifiying or used as buffer I think using any Jfet input low noise opamp should do the job. Any reason why particularly AD823 to be used?Offset minimizing before putting the servo chip in the socket
There are posts about that. Its not the only option. Any good modern JFET input dual opamp of +/-18V rail capability can do it. The AD823 was chosen on a subjective level about subtleties. That is why there is a socket to roll any other preference. Some said they could not find a difference when rolling chips some said they could. An OPA2134 is a good example to try against the AD823 and see for yourself in the context of your own audio system. In the finishing design phase I have tried about ten different opamp models relying on different headphones including Sennheiser HD 600, AKG K612 Pro, 712 Pro, Fostex T50RP MK3, Grado SR 60, and other. So to avoid power amps and loudspeakers + room further influence/synergy.
Hi Salas
i have now tried first step on the preamp..
use now my old supply from amb.org σ22
give more life in sound- with better dynamik
so now i want to know ,,are the Salas DCSTB supply,, in teori ..better than the AMB
i will like to try something with the caps on the preamp
are there some caps who are important for the sound (the small wima)
will more bias give better sound,,,,only for line pre
need to wait with the 10k pot,,i have to order one
best Bjarne
i have now tried first step on the preamp..
use now my old supply from amb.org σ22
give more life in sound- with better dynamik
so now i want to know ,,are the Salas DCSTB supply,, in teori ..better than the AMB
i will like to try something with the caps on the preamp
are there some caps who are important for the sound (the small wima)
will more bias give better sound,,,,only for line pre
need to wait with the 10k pot,,i have to order one
best Bjarne
Hi Bjarne
This σ22 supply you replaced the one from NewClassD seems to take your audio system's sound synergy to the right direction so lets keep it for now. Its a competent discrete series supply with good characteristics that covers this application. So I would have not only to to model it in simulation but to also build it and listen to it with the preamp to answer your question vs DCSTB.
Regarding the Wimas, C5 has a minor influence but its already MKP. By the way you may try other types but don't change values in the main board capacitors including C6 & C7 electrolytics.
150mA bias in line pre only mode can be used with no negative effects if the sink is easily taking more than 100mA but it does not produce better measurements at line level loads. A couple of members said they could hear better sound in line mode also but I did not hear something right out noticeable when comparing bias for non demanding loads at the time of experimentation. I had many things to check and maybe I did not focus enough though.
What is the easiest experiment to keep you busy until having the 10K pot (better a cheap switching pot) is roll OPA2134 and TLE2072 in the servo socket. Those could progressively punch up your sound but maybe the last one could also roughen it up a little.
This σ22 supply you replaced the one from NewClassD seems to take your audio system's sound synergy to the right direction so lets keep it for now. Its a competent discrete series supply with good characteristics that covers this application. So I would have not only to to model it in simulation but to also build it and listen to it with the preamp to answer your question vs DCSTB.
Regarding the Wimas, C5 has a minor influence but its already MKP. By the way you may try other types but don't change values in the main board capacitors including C6 & C7 electrolytics.
150mA bias in line pre only mode can be used with no negative effects if the sink is easily taking more than 100mA but it does not produce better measurements at line level loads. A couple of members said they could hear better sound in line mode also but I did not hear something right out noticeable when comparing bias for non demanding loads at the time of experimentation. I had many things to check and maybe I did not focus enough though.
What is the easiest experiment to keep you busy until having the 10K pot (better a cheap switching pot) is roll OPA2134 and TLE2072 in the servo socket. Those could progressively punch up your sound but maybe the last one could also roughen it up a little.
Hi Salas,
which value of volume pot do you recommend?
As written earlier I wan't to use a ladder volume control and its possible to use
either constant input or output resistance.
My experience is with series pots carbon and switching with 20K log nominal as a good spot. Given there is no perceived tonal tilt in other parts of the audio system including the speakers in room interface when using an active preamp with gain. That means 20K constant input and variable output resistance of 5K max at half point. You should experiment from there.
or is a TKD 2CP601 a better choice?
Good tone pot. Discrete resistors are precise though. Subjective.
As written in the GB For Salas DCG-3 and DCSTB Power Supplies threat:
I would like to build the DCG-3 as the PreAmp/Headphone stage for the “Soekris dam 1021 R-2R DAC”.
Because the DAC integrated volume control destroys the pure DSD path I will use a volume control within the DCG-3. Any hints for that?
Because the DAC will be my only input I need no input selection.
The DAC RAW output will deliver 1,4V and my Hypex and Croft PowAmps will handle 2V. That's why I think the standard DCG-3 gain of 2 is sufficient.
I would like to build the DCG-3 as the PreAmp/Headphone stage for the “Soekris dam 1021 R-2R DAC”.
Because the DAC integrated volume control destroys the pure DSD path I will use a volume control within the DCG-3. Any hints for that?
Because the DAC will be my only input I need no input selection.
The DAC RAW output will deliver 1,4V and my Hypex and Croft PowAmps will handle 2V. That's why I think the standard DCG-3 gain of 2 is sufficient.
Last edited:
The cheapest precise 20K log pot you can get is one of those small jobs found on Ebay advertised as DACT-like which other than neither mechanically made for a lifetime's service nor sporting some top notch minimal ppm SMD resistors they still do keep a dead on channel centering and a short signal pathway area inside them.
P.S. I currently have Dim Dim's Soekris build on my rack as he kindly offered it to me so to get by until my Aune DAC returns from a factory repair. It pairs very nicely with the DCG3. The standard gain is 3 and with this DAC I can listen loud enough with -6dB peaking recordings at one o clock pot position with a gain twenty power amp at four to five meters distance but my speakers are 95dB SPL at 2.83V/1m. From this clue you may somewhat estimate your situation and what gain to use.
P.S. I currently have Dim Dim's Soekris build on my rack as he kindly offered it to me so to get by until my Aune DAC returns from a factory repair. It pairs very nicely with the DCG3. The standard gain is 3 and with this DAC I can listen loud enough with -6dB peaking recordings at one o clock pot position with a gain twenty power amp at four to five meters distance but my speakers are 95dB SPL at 2.83V/1m. From this clue you may somewhat estimate your situation and what gain to use.
Using soekris dac myself together with dcg-3. Works like a charm. The dac has a unreleased potential with regards to digital filters IMO.
With good filters the dac is extremely good.
However:
Dont understand that part.
With good filters the dac is extremely good.
However:
Because the DAC integrated volume control destroys the pure DSD path
Dont understand that part.
Using soekris dac myself together with dcg-3. Works like a charm. The dac has a unreleased potential with regards to digital filters IMO. With good filters the dac is extremely good.
True, its all in the filter codes. It's a chameleon.
volume controll is made in PCM. You have to leave pure DSD path. There for I would like to have the controll within the DCG-3.
Did that make sense?
The DAMs, being R-2R devices, need to convert DSD to PCM to make sound. Disabling the digital volume control will not help with that. There can be no "pure DSD path" with R-2R dacs.
OK, thanks for that. That was not my understanding. I will ask there again.The DAMs, being R-2R devices, need to convert DSD to PCM to make sound. Disabling the digital volume control will not help with that. There can be no "pure DSD path" with R-2R dacs.
end off toppic ;-)
Received my chassis from Diyaudio a bit earlier than expected. It is rather large and has plenty of room. The rear panel is cut for IEC plug, the part I have would work, screw hole spacing is fine, but out line of part doesnt quite cover the entire hole, leaving some small corners exposed. The part for the precut power amp size Chassis from Diyaudio store contemplates a different sized part, too large as it also holds the fusing.
Is there a plug everyones been using that fits this cutout and includes switching as well?
I am looking at Mouser and Digikey as I type this...Thanks in advance to those who have located this part.
Russellc
Is there a plug everyones been using that fits this cutout and includes switching as well?
I am looking at Mouser and Digikey as I type this...Thanks in advance to those who have located this part.
Russellc
Deluxe case? Store has pn of IEC under rear panel accessory kit listing.
BK
Never mind - now notice you differentiated the power amp case. Darn phone display..,
I know, after searching myself blind, I have come to the conclusion that there is not one that small that has a switch. I found some with a fuse, but none with switch. So, one more hole must be drilled. It would be nice if the store made it large enough to fit a switched one but thus is life. I really like the way they do the power amp size one with the built in switch.
Russellc
- Home
- Source & Line
- Analog Line Level
- Salas DCG3 preamp (line & headphone)