6c33c SET

[Nick]

Ok, I finally got round to drawing out the schematic of what I have at the moment.

EDIT: I have updated the circuit.

[simon]

Ooooh, that's tempting. Are the heaters AC, I can't remember? The hum was pretty impressive if they are.

[Cressy Snr]

Hi Simon,

If the heaters are AC there will still be negligible hum. Although my 6AS7s are not as powerful as Nick's valves, they are of a similar type. These types seem relatively insensitive to heater induced hum. My PP amp uses AC heating and is silent.

It is one less complication and it is a good job they don't need DC. Due to the heater current requirements (2.5A per valve on the 6AS7 and 7A per valve for the 6C33C) I would not like to have to put together a regulated DC supply for either valve, least of all the big Russian one.

Steve

[simon]

Hi Steve,

Good point about the current requirement of the 6C33C. I think Nick did say that they were AC but they were so quiet I guess I forgot thinking they were DC. There's a certain reassurance to a little hum - at least you know the amp's working

[Nick]

Nope all AC. As I said to Simon, there is a little hum on one chan due to a hum loop I haven't fixed yet (it goes away when the input is disconnected). But on the bench the hum/noise is down in the 200uv range, and most of that is between the output TX's and the mains torroid, its just about nulled out by rotating the torroide to find the low point.

I think the 6C33Cs are further helped by the two heaters being wired out of phase.

[Andrew]

That's an interesting Aikido-type driver, Nick. Does the 83 sound better with the higher gm and the partially bipassed cathode (didn't I read this on SB web page?)

I assume that's your thinking?

cheers,

-- Andrew

[Nick]

Yes, I think the 83 is better in parallel, seems to have more body for what of a better word. I haven't measured what the spectra looks like.

The partial bipass isn't that clever, is so I have to a place to apply loop feedback to if I feel the need. At the moment with the OB's there doesn't seem to be that much of a need, so I haven't.

Actually, I could remove the bipass alltogether, its got enough gain when used without feedback anyway.

Interestinly, if you look at he FR of the spice model I have for it all, you can clearly see the effect on the low freq end the two chokes have, but adding the aikido cap to the lower valve masks that, so I think the "noise reduction" action may have an additional benifit

My first try at the driver stage used the 6n6p also in parrallel as a cathode follower, but the additional heat dissapated under the top plate when added to the heat above just got silly. It doesn't need any gain in that stage. I considered a SRPP but was put off by the hassle of calculating the thing for minimum distortion into the load, which would change as the bias was altered.

[Andrew]

Interestinly, if you look at he FR of the spice model I have for it all, you can clearly see the effect on the low freq end the two chokes have, but adding the aikido cap to the lower valve masks that, so I think the "noise reduction" action may have an additional benifit

Yes, I'd be interested to see that, for educational purposes, of course...

cheers,

-- Andrew

[Nick]

I will send you it tonight.

Oh, and the 56R in the coupling caps, should be 560R

[simon]

Would the 560R be the same for the coupling cap in my phono, or is 56R right in that instance?

[Nick]

No, it should be 560, my flawed memory.

It works by cauing a xover at the frequency given by the small cap and resistor, so the xover point is the frequency where the Xr of the small cap is the same as the resistor.

So the xover frequency will be

560 = 1 / ( 2 * 3.14 * f * C )

or

f = 1 / ( 560 * 2 * 3.14 * C )

So for the case of the 0.1u its at 2800hZ, I just tried the values in the diag (I will change it tonight) as thats what I had, but with a 56R resistor, its at 238kHz So that explains why when I tried it last night, I didn't hear a change. I will try again tonight with 560.

In reality, it would be better (and thats how it was in the phono) to use (say) 0.68uf as the small cap and 560R, that would put the xover at 400hz

Or I still have it all wrong

[simon]

Thanks Nick, I think I understand that. So thinking of the 470nF Russian PIOs I have, to bypass them with a 10nF silver mica I should use 39k to give me approx 400Hz. 39k seems quite large - might this have an effect on the sound? I think I have some 4n7F silver mica too, maybe I'm better using these as a 100:1 bypass without the filter?

Or perhaps I should just try some different 470nF caps.

[Nick]

I don't think there is any point of doing the bipass on a 470nf cap, T did it in the phono as it neeed very big (6uf big) caps, so getting nice ones at that valve is very very pricy.

In your case, I think its only worth thinking about for the final cap you have (I think)

[simon]

Yeh, that makes sense. I was coming to that conclusion when I did the sums.

[Andrew]

Hi Nick,

Do you think you the amp would benefit from regulation of he driver stage? I'm contemplating this on the 300B's and have been amassing bits for a few experiments, currently a double choke LCLC affair.

Why do I ask? Well on higher current stages regulation seems a winner, yet I recall one of our starting points on this was my comment that even on a reasonably lowish current phono some form of regulation seemed to help the bass and we do know regulation was a step forward on your LCR.

I guess I was surpised you hadn't regulated the driver, in fact.

cheers,

-- Andrew