Hi,
Can I use this to provide a selectable gain stage for the DCB1? The PS voltage requirement here is quite different from the DCB1 which takes 15VAC, Can I tab into the DCB1 regulated PS to drive this circuit?
The beauty of that particular schematic is that the 78L12 and 79L12 are the positive and negative DC regulators. You only need a raw bipolar supply that produces rails between 14.5-27 volts. (towards the lower end is probably better) Get a 12v+12v transformer, a pair of diode bridges, and two caps of at least 16v rating and 1000uf or more. Easy!
That's a nice gain stage, just take care of couple of things:
- for optimal results choose 2sk170 with Idss in 7.5-9mA range
- PS must be very clean (especially -3.3V)
Juma, referring to your nice design from post #53, there's any advantage increasing the p.s. to, let's say, +/-15V or even +/-18V?
If positive, should we modify any component value?
Thank you!
If positive, should we modify any component value?
Thank you!
Only if you need more than 10V_peak at the output, which is not the case with F5.
I used salas shunt regulators for the one I built. Jumas nice little circuit fit on a 2" square of perfect board and sounds great.
This past weekend, i listened to the F5 with the DCB1 extensively. While the F5 was not my favorite choice with Juma's schade circuit or my 6v6 preamp, the DCB1 and the F5 have a special synergy IMO. Based on this i think it would be wise to try his bf862 version of the B1. On paper, it has the potential to outshine k170 in same circuit due to better capacitance numbers. Don't know if you will be able to hear it or not, but should work well with very wideband F5. Never got around to building it, but will do so soon.
Buzz, I recently investigated behavior of the BF862 in the low Vds region with stiff cascode and this circuit succeeded in sounding better than simple SE buffer (B1-like). It's not a lot of work to test it but I find it worth the effort. J310 can be used as cascoding JFETs instead og BF245C.
Attachments
No. It's 20 Ohms 😉
One more thing: Q4 should have the same or a bit (few mA) higher Idss than Q1.
Last edited:
I suspect it is better due to gross power supply? What was the power supply? What if Salas' reg is used?
I tested with simple lab PS (uA723 based). Better PS usually brings better results...
Of course. The number 20 written near the resistor means that too.
I think i may try this with SMD parts. See hoe small i can make it. I have heard that smaller is better. May be backwards on this one.🙄 Juma, You think it wise to add small gain stage in front using nother BF862. I know the DCB1 is cranked just about full on when paired with F5. I like it LOUD!
We already did it here.
Now we can take it one step further by avoiding the output cap (just C1 stays as a small coupling cap between stages, so it can be of high quality and still not expensive).
Four PSs seems a lot but it's two actually: we start with +24V(V2) and -12V(V5). Then we regulate them down: +24V to +12V(V4) and -12V to -3.3V(V1).
Gain is 10dB (since you like it LOUD).
For clarity, schematic doesn't show bypass/decoupling caps - 10uF Low ESR on every rail will do the job, but depending on PS type,you can try whatever you like (RC combos, small caps in parallel, etc...).
The pot/volume control at input is not shown but 10 to 20K log device is recommended.
