Salas DCG3 preamp (line & headphone)

diyAudio Chief Moderator
Joined 2002
Paid Member
Hmm, different stuff. One is playing with the system's rise time/sharpness the other with the OLG figure and its width of shape due to different Cob, Rbb'. Also the two BJT types have different self noise. It can have an impact on the bass resolution too. The trick is to have a couple of alternatives to can unlock your installation's potential. Of course when it already performs in a system's context just let it sing. You know its not easy to give alternatives in a circuit like this one, fast things demand to be exact, but I tried to leave a couple of windows open because we are diyers and we love to tweak.
 
And then there was smoke

So on the DCSTB I had reversed the polarity of both DB3 and DB4 diodes. Didn’t realize this until I plugged it in and very quickly C1 let out it smoke.

I replaced C1 and both diodes but the LED doesn’t light. What else might I have smoked? Any suggestions on what to check next would be appreciated.
 
Member
Joined 2009
Paid Member
Having listened to my now Ultrbib powered DCG3 for several hours over the past week I have to agree with others comments that this p/s is a step up in sound quality. The DCG3 and DCSTB work really well together but somehow there seems to be more space and definition around and to instruments with the Ultrabib. There can't be that many wow moments left surely.
You've done it again Nick thank you:cheers:
 
Having listened to my now Ultrbib powered DCG3 for several hours over the past week I have to agree with others comments that this p/s is a step up in sound quality. The DCG3 and DCSTB work really well together but somehow there seems to be more space and definition around and to instruments with the Ultrabib. There can't be that many wow moments left surely.
You've done it again Nick thank you:cheers:


Thanks for your impressions and for describing the heatsinks used in an earlier post. Besides space and definition have any of the other sterling attributes of the DCG3/DCSTB combo been improved upon or diminished somewhat. Did you use the 327's in the mirror in this build and for comparison the previous one? I wasnt aware of any other builders who have tried the UltraBiib's/DCG3 as you are mentioning. Who?

I am using the Diotec 327's in my balanced build since around April of the year and the overall tonality and presentation is stunning. A step up from the 560c.

nash
 
diyAudio Chief Moderator
Joined 2002
Paid Member
The two pots output impedance will add up. That impedance is source to the preamp and augments the R element of its RC input filter. The more RC slows down the slew rate response and narrows the HF -3dB frequency response point. You would want the total to stay within the range of the 20K-50K recommendation for DCG3. The article's Fig.7 standard configuration that says below it "BAL = 2.5 × VOL. For example: VOL = 10k log, BAL = 25k linear" should be like a 35K standalone pot for output impedance.
 
diyAudio Chief Moderator
Joined 2002
Paid Member
Set up that pots system alone first and measure its resistance from input to output when rotating so to find the max. See if it falls in the recommended range or to change pot values. The equivalent max for series type 20k-50k nominal pots is 5k-12.5k when at half point value travel.
 
diyAudio Chief Moderator
Joined 2002
Paid Member
In any case, if its not about some low level hum to solve but its hiss, you can reduce the gain to x2.33 with R7=750R. That's when the standard x3 is inadequately loud for your system and your headphones 115dB/V sensitivity regarding useful volume pot range and/or background noise creeping up.

Maybe its an opportunity to test with the alternative BC327-40 Q1 Q2 also. You may put female header pins for those components to can swap them with the BC560C originals and compare. The DC offset setting up procedure with VR1 must be repeated when changing Q1 Q2.
 
0.25W Resistors

Just doing some final checks before powering up my DCG3 build, and found that I may have made a mistake - can someone please advise?
I have used Takman REY metal film, but through lack of attention have installed 0.25W resistors in about 50% of the positions, rather that 0.5W, which is probably closer in characteristics to the CMF55 recommended in the BOM.
R3 is a 1W resistor.
Question: is it okay to leave the 0.25W pieces in the circuit, or, where is it better to have 0.5W (like R6 perhaps)? :confused:
I expect to bias at about 110-120 mA, and gain of about 3.4x.

-Keith-