More SEPs

[Cressy Snr]

This is where I'm at now:

Screenshot 2023-03-12 at 16.56.41.png (55.18 KiB) Viewed 1285 times

Have yet to audition it properly, I just gave it a quick blast, to check there was no obvious distortion, but the main thing is that it now works properly, has a decent amount of gain, the D3as are no longer microphonic and are where they should be regarding the designed in screen volts and op-point. I'll see how it sounds later on.

[Nick]

In the interest of ideas from the past, you could try taking the cathode of the OA2 to the cathode of the d3a. This will allow you to both reduce the cathode resistor valve because of the extra current and also the g2 voltage will be fixed with respect to the cathode as it should be and not wandering in response to signal current in the unbipassed cathode resistor.

[Cressy Snr]

In the interest of ideas from the past, you could try taking the cathode of the OA2 to the cathode of the d3a. This will allow you to both reduce the cathode resistor valve because of the extra current and also the g2 voltage will be fixed with respect to the cathode as it should be and not wandering in response to signal current in the unbipassed cathode resistor.

Aye,
With the conventional small signal pentode screen supply by resistor, the bypass cap invariably goes to the top of the cathode resistor, but I’m normally too damned idle to do it properly. I’ll sort that tomorrow.
It’s the sweating of these little details that add up to the best sound isn’t it. As I mentioned in another thread, I could kick myself for not carrying on with active regulation of power supplies after the blindingly obvious superiority of the 6B4G monoblocks over anything else I’d built up to then.

Rather than the KISS mantra, IPSS! (It’s the Power Supply Stupid!) should be the operative mnemonic!

[Cressy Snr]

Got a bit of volume control rustle where there was none before. There may be some residual oscillation going on. I had the same years ago with the 12GN7 front end on the 6B4G monos. Once I got the right values of grid stoppers, the rustle disappeared.

So two jobs tomorrow then : the first being the cathode connection for the VR tubes, then trying to sort some oscillation prevention measures.

[IslandPink]

In the interest of ideas from the past, you could try taking the cathode of the OA2 to the cathode of the d3a. This will allow you to both reduce the cathode resistor valve because of the extra current and also the g2 voltage will be fixed with respect to the cathode as it should be and not wandering in response to signal current in the unbipassed cathode resistor.

Seconded - this sounds better, I tried the options years ago, bass is tighter.

[Cressy Snr]

I've been working on this over the last couple of days and have ended up dispensing with the services of the D3a driver valves, in favour of (as Nick suggested) something that would be a little bit easier to work with. I found a stash of 1950s, Brimar EF80, so thought I'd use those as drivers instead (7 point something mA/V.) I've used them triode strapped before but not as pentodes.

Had to estimate the pentode curves for a 150V screen grid value, as the data sheet only goes down as far as 170V.

They work beautifully in this circuit. There had to be a few changes of course to accommodate them, but it was worth doing. Cathode resistors were changed, pin 8 and pin 9 socket connections had to be transposed and there were a couple of other changes to grid resistors, both after the volume pot and before the output stage.

Schematic:

6L6SEP_EF80Drv.png (55.89 KiB) Viewed 1165 times

Sounds good. The positive change in bass response, with the EF80 cathode connection from the VR tube rather than ground was, to overuse a cliche 'not subtle.' But then everyone else but me has been taking advantage of this for years.

[Cressy Snr]

For the sake of completeness, here is the PSU:

6L6SEP_PSU.png (91.31 KiB) Viewed 1109 times

[steve s]

I threw away a box of of ef80's 20 years ago steve, probably still got a few somewhere, there where that many about they where worthless !

It will be interesting to see what the e180f is like amplifying at driver signal voltages and i would guess easy to overhead the input.
I know Nick has used them, so im sure he has a good idea of their qualities, for me it will be interesting to see and hear, so looking forward to how it sounds.

Would be great to arrange a meet up too ?

[Cressy Snr]

I threw away a box of of ef80's 20 years ago steve, probably still got a few somewhere, there where that many about they where worthless !

It will be interesting to see what the e180f is like amplifying at driver signal voltages and i would guess easy to overhead the input.
I know Nick has used them, so im sure he has a good idea of their qualities, for me it will be interesting to see and hear, so looking forward to how it sounds.

Would be great to arrange a meet up too ?

Hi Steve, drop us a PM or a text with a date and I’ll be glad to meet up. I hardly use the car these days, it could do with a good motorway blow through.

[Cressy Snr]

You know, I think it might be safe to put the KT66s back in. The sacrificial new production 7581As have done their job admirably whilst the redevelopment has been in progress.Time some proper valves were brought back into the fold.

[Cressy Snr]

Happiness is…

Valves, an old record player and Boom Radio:

BBDA2D42-322C-4A3F-AD3C-8420ED3684E6.jpeg (87.43 KiB) Viewed 971 times

…“So catch me if you can... I’m going back!”

[Cressy Snr]

AAAAARRRRRGH!
You have to take the rough with the smooth with Grandad Radio.

1E7DFF96-4C51-4FAB-928E-BEA371C88ACE.jpeg (73.27 KiB) Viewed 960 times

[Cressy Snr]

I think I can call the SEP a success now that one or two people have heard it. I've enjoyed getting to the current state of play. Although the SEP works superbly with Steve's hi-efficiency speakers, it still does a remarkably good job on my comparatively inefficient Mission 773 WAF friendly slimline towers. The smallness of our listening space helps out enormously in that department. I get away with the setup in that room, no strain being evident.

The SEP gives a clean, clear and musical performance with the Missions, particularly now I've bi-wired them. Mind you, don't get too excited because, as I mentioned to Scott yesterday, the fact that the shorting strips were corroded to buggery when I got them out, might have had something to do with the slightly recessed top end. I remember Ray commenting that they sounded a bit dull when he tested them out prior to bringing them up North.

The top set of terminals are connected to the tweeter half of the crossover and I was plugged into the bottom set. I really should have polished up the strips when I got the speakers home from Owston last October, but my head wasn't in a good place at that point and I couldn't be arsed with anything really. I've only recently got out of the depression. The speakers'll probably sound just as good single-wired now.

Building the SEP and getting a result has done me a power of good. Now I need to get on with sorting out this shit tip of a house over the spring, summer and autumn whilst the momentum is with me.

[Cressy Snr]

Well today I was having a look through Vacuum Tube Valley, issue 16 and came across a 2001 Lynn Olson article entitled "Ultrapath, Parallel Feed and Western Electric."
Now I know of course that in 2023, all this is old hat, but I was intrigued with Olson's argument that using the ultrapath connection with a regulated power supply is a better idea than not using it.
I've done ultrapath before, but was ambivalent about the sonic qualities and couldn't really make up my mind whether it was good or a passing fad, so I abandoned it years ago. I know Pete Millett used the connection with his e-linear amp published in AudioXpress, but his supply was not regulated apart from the usual shunt reg screens on the D3a input stage.

Anyway I've tried out the scheme on the SEP with its regulated PSU and the results are rather impressive: big sound stage and room filling sound, even with cheap, new production 6L6 types. Interesting.

Screenshot 2023-03-28 at 17.05.56.png (57.53 KiB) Viewed 574 times

[simon]

Must be very good indeed then Steve as it was very good before