DCAudioDIY.com

My Aikido preamp isn't in a metal case (it's built on a delrin plate with a wood base) and it's dead quiet--and it has a lot more gain than this. Then again, the Aikido is designed to cancel PS noise and has a regulated supply.

FINALLY. I didn't add enough capacitance to the DC filament supply. Duh. Anyway, I've got an Antek 275-0-275/6.4/6.4 shielded tranny. 6X5 rectifier, DC filaments on the 6V6's (yes, I went back to those). Two 30H chokes and the Maida regulator, which might stay or go. It's quiet now and I can play around with slimming down the PS. I'll draw something up when I'm done messing around.

Grover,

Glad you got the DC filament issue resolved.

What did you find more appealing with the 6V6 vs the 6AH4?

David

It feels smoother, lower distortion, plus there are about 1000 varieties you can try.

Now I've added 6dB of feedback, posted by a guy on the DIY forum. This really kicks it up to a competitive sound.

Grover,

Looks good. The feedback is a nice touch, gets the overall gain down to where it should be (at least for those of us who have plenty of gain in our amps. It should lower the output impedance a bit as well.

What are you using for the 470uF cathode bypass capacitor? I have thought about using an LED string to avoid the bypass cap, like 5 red leds (maybe 10 with two parallel strings since 22mA may be a little hot for a single string). I need to build this to try out sometime.

David

Yes, that bypass cap is going to "stand out," isn't it? I have some cheap Nichicons in there now, but I just found some old Black Gates I could use. I have some Elna Cerafines on order. LEDs would be interesting. Could also do a current source if I knew how.

Interestingly, I've been trying out different combinations of rectifiers and power trannies to get the right B+ and sound. One Antek was dead quiet, another gave me some 60Hz hum. Now I'm using one of those massive old Angela Instruments universal trannies with a 5AR4. It has 640 vct secondaries. It's overkill but it's also the quietest of them all. I think I'll build the PS on a separate chassis.

I really didn't have high expectations for this project but the speed and transparency are pretty captivating.

Oh, and I definitely need both chokes before the Maida board. It seems I could reduce the filtering quite a bit but it's not that way at the moment. I wonder if I have it set up wrong. David, when you have a moment, could you look at the brochure and see if I'm missing something? I'll email it to you.

If you use a current source in the cathode, you're still going to need a bypass cap...

Roscoe

_________________
I can explain it to you, but I can’t understand it for you.

Grover,

Please check the drop-out voltage on the regulator (needs to be more than 15V). Based on your B+, you need to setup the regulator for operating range IV (315-420V). Also, pay attention to the guidance on the maximum output capacitor after the regulator...notice the recommendation is to use 10uF decoupling capacitors at the load. I also would not expect you need as aggressive filtering, the regulator should take care of most of that. It would be curious to see what a good solid stage rectifier using 1200V SiC Schottky diodes would do. At such low currents thru tube rectifiers (for a series reg like the Maida, only 40 mA or so total in your circuit), may be a bit low. Just surprised you are hearing so much difference in your raw supply with the two series chokes.

Also, concur with Roscoe, you do not want to use a current source (sink) for biasing for a common cathode circuit without a bypass cap, the current source blocks AC. For a cathode follower it does not matter, since you want to block AC from the cathode to ground, or on the plate of a common cathode in place of the plate resistor, but you still need something to bias the tube. The string of LED s is a way to implement (fixed) diode bias via with a much lower dynamic impedance than a cathode resistor so that it does not need bypassing. Battery bias could work as well, in fact, a 9V battery would do the job nicely, either in the cathode OR at the grid (need an input cap). In the cathode, a NiMH or NiCad rechargeable that trickle charges, or a long life non-rechargeable at the grid that just sets a voltage (no current). These are well documented.

David