6EM7 Headphone Amp

[Cressy Snr]

As I promised Paul I would do, I've just lashed up a 6EM7 headphone amp, using the power valve section of the two dissimilar triodes as a single gain stage. I've used my other headphone amp as the test bed for the job:

IMG_0638.jpeg (122.84 KiB) Viewed 48846 times

IMG_0637.jpeg (138.86 KiB) Viewed 48846 times

In the bottom picture, a Meanwell, 12V switch mode power supply can be seen next to the amp. This meant the valves had to be connected in series.
I'll have to see if I can get hold of a couple of 13EM7s so that the heaters can be paralleled up.

I can report that it sounds great: natural and un-fatiguing.

[Cressy Snr]

Audio circuit:

6EM7_HPA.png (26.19 KiB) Viewed 48841 times

[Paul Barker]

Thank you?

[Cressy Snr]

OK, I’ve finally started on a proper 6EM7 headphone amp rather than a lash-up. Theres a long way to go but it should be good (I think.)
Here’s the box made up with General Electric, coin base 6EM7 our Ant found during his house move. Rectifier is EZ81.

1000000045.jpeg (105.66 KiB) Viewed 33644 times

[Cressy Snr]

And we have heaters:

IMG_20250712_192753021.jpeg (90.9 KiB) Viewed 33059 times

[Mike H]

[Cressy Snr]

OK, so I’ve finished it.
Using both sections, of the 6EM7, there was far too much gain as you’d expect, but with the power tube section on its own there was not enough to play vinyl decently.
So using both sections and a ton of global feedback wrapped around it, I got the overall volume about right.

It’s a similar idea to the old WAD HD83 which used an ECL83 triode/pentode, with the pentode section triode strapped and feedback applied from a special winding on the OPT.
I don’t have that winding so applied my feedback conventionally.

IMG_20250714_181713302_HDR~2.jpeg (126.89 KiB) Viewed 30526 times

TBH, it sounds great; hard-hitting, fast and dynamic like an ‘80s Naim amp, but with an air of refinement and liquidity that the ‘80s Naims lacked. Nice big soundstage too.
I’ve a small amount of hum to chase down, but it’s all extremely promising.

[Paul Barker]

Yes when it is on speakers the soundstage excels compared to dht’s which are hard to coax out of their veil.

I also found the Rosenblit original 6as7 OTL design filled the room from corner to corner.

I’m glad you could sense it on headphones.

[Paul Barker]

Maybe the se 6em7 should amplify to a phase splitter to a push pull driver to a push pull output stage. All multigrid indirectly heated triode connected valves. The EL38 fits the bill for the driver and the output stage. Whether the 6em7 soundstage continues through all that to the speakers is unknown.

Or go back in time to James’s OB’s. Though I failed to copy them well enough.

To be fair my fostex plaster board 6ft by 3ft baffles sounded better to me. But you need a lot of room. I may trim them down to Briggs dimensions and go back to 6em7 at home. The German alnico woofers 12” I think, matched well in power with the fostex. They were a big part of the sound, and did it better than the woofers James used when I tried it, hence why I sold them to Mark, and is it Simon has them now? I failed with them. I never went perspex though.

But I got my synergy with those baffles I brought to Owston. They are still at the lockup.

On reflection James OB’s were marmite. I recall first hearing they sounded poor on 801a but Jono’s simple 45 amp brought out the nice things they sometimes hid. They sounded fantastic or not at all good. In the end the foam goes in the full range portion. Time isnt kind to expensive drivers suspended with foam. The foam went on both my AE’s time passes. That was the day I brought 212’s and Jono said “it’s not worth it” !

Now Guy Martin has blown up all that remained of our best ever venue. Time passes.

[Nick]

Time isnt kind to expensive drivers suspended with foam. The foam went on both my AE’s time passes.

That (and the 4k difference in price) is why I used PM6A instead of the AER's.

[Cressy Snr]

Picture of the inside:

IMG_20250715_152518658_HDR~2.jpeg (129.76 KiB) Viewed 28890 times

Diagrams will follow shortly.
I’ve managed to sort the hum. Still a bit of residual noise but that’s valves for you.

I’d need regulated HT and regulated DC heaters for silence. If you can’t be arsed with that malarkey (which I can’t) then don’t expect silence. There’s only so far you can go regarding noise, with unregulated HT supplies and AC heaters, especially when you have 113dB/w speakers right next to your earholes.

Of course you could always use transistors.

[Paul Barker]

Have you referenced the heater to ground?

[Cressy Snr]

Yes, but that’s the least of my worries!
Now a cathode resistor on the power section of one of the valves has burned out. The valve obviously ran away. Maybe a leaky coupling cap. But either way it’s out of action. The grid resistor of 500k was well within the 2meg limit.

Too many old components. Need to build with new stuff.

[Cressy Snr]

Well eventually I finished up reverting to the 1626, driven by an ECC82 as these can, of course, be configured for 12V heaters. It's about the best I could do given that I couldn't find another 6EM7 that would remain stable, even when the grid resistor was reduced to 220K. I wasn't going to risk Simon's 6EM7s being destroyed either.

I don't seem to be having a lot of luck with valve projects lately. However the 1626 driven by an ECC82 worked first time, which boosted the old morale no end.

There's a tiny amount of hum on the right channel, but it is easily livable with. The left channel is silent. I think either the toroidal mains transformer needs spinning round or choke number 2 in the power supply needs turning through 90 degrees. I'll take a look at that tomorrow.

1626HPA.jpeg (114.64 KiB) Viewed 20922 times

1626HPA_Int.jpeg (119.39 KiB) Viewed 20922 times

I need some better coupling caps than the orange drops, but apart from that, the amp sounds nice and tubey, which will do me nicely
There's none of that Darling stuff involving shared cathode resistors.

[simon]

I wasn't going to risk Simon's 6EM7s being destroyed either.

Easy come easy go