6AQ5/6005/EL90, Parallel Push-Pull Project.

[Cressy Snr]

OK,
here is the concept schematic.


Like the KT120 which was developed over a long period, using a tried and proven
power supply, this project too, has had a long gestation period.
Starting with a transformer coupled, push-pull EL84 amplifier, a set of twin power supplies, partly based on the writings of
the late Harvey Rosenberg, and a hacked about red chassis, the design evolved gradually into a fully balanced, lowish power 6005 based output stage,
driven by 13D1 Brimar, twin triode, differential drivers.

What I ended up with was a superb sounding amplifier, easily as good and better in some ways than my KT120 SE setup, but it looked like hell.

The idea now is to increase the power output by using parallel output valves and build it into a chassis that doesn't look like an explosion in a junk shop.
More will follow as I finalise the schematic and get hold of a decent chassis to house it.
This is one build I am really looking forward to getting on with, I'm not dealing with the unknown
and I know it will provide high quality sound.

[Paul Barker]

OK, but how about hook it up first to make sure before punching the chassis?

[pre65]

Aye laddie, that is what breadboarding is for.

[Cressy Snr]

Bit of a revision.
Ideally I would like the design to use 7-pin miniatures throughout, so the drivers have been changed to a pair of triode strapped 6AU6.
The 13D1s were perfectly adequate to drive EL84s and sounded excellent with 6AQ5s but the fact remains that they do not have enough gain on their own, to drive them to full output and current is marginal for driving parallel pairs of triode strapped 6AQ5 output valves.

Of the 9-pin miniature twin triodes I could have used, 12AT7s are great as differential drivers, but their gain, at 60 is far too high for triode operation of 6AQ5 outputs, sans feedback.

I needed something in the 35 gain area; the triode strapped 6AU6 is perfect for this application and has adequate current to drive the parallel pair of outputs per side.
I don't like 6N1Ps or 6N6Ps and have already rejected 7044 as too microphonic, ECC88 as too sterile and grey, 6463 as bloody horrid, E182CC as strident and ECC82 as very nice but ultimately a bit boring.



Here is the revised schematic.
There is of course the additional possibility of pentode drive available from 6AU6s, if desired.

[JamesD]

Steve,

Some 7-pin alternatives

Twin triodes with common cathodes...

5J6, 6J6, 19J6 same valve different heaters.
5964
6101

Single triodes about the right mu:
9002
8532

Have fun! I suspect the 6J6 would be the best bet. €1:50 at Halfin www.electols.com

James

[Cressy Snr]

Cheers James.

6J6 looks good.
Those U shaped wrap-around anodes look spookily like sub-miniature versions of the early British AC/HL shortpath construction.

[JamesD]

Its very well aligned with use as a differential pair. the anode curves show it generates mostly (and quite a lot of ) 2nd harmonic distortion and that will cancel in the differential pair.

One useful tweak with the diff pair is to decouple the B+ to the cathode as you do with the output pairs. This provides a path for the non-cancelled part of the signal that removes them from the output current loop Its easy to try and see if it improves the sound - it has done when I have used it...

J

[Cressy Snr]

Thanks for the tip re the input stage HT decoupling setup. I wouldn't have thought of that little dodge in a million years.
The gain of the 6J6 at 38 is just about bang on for what I want, when driving 6AQ5s.

[Cressy Snr]

Ok so here is the schematic with a 6J6 as input,driver.
Looking at the transfer characteristics of this little valve, there is only a limited area of operation, where you can avoid generating higher order odd harmonics. The best points are below -2V grid bias where at around 100V and 8 and a bit mA, you keep away from the curved area of the line on the negative swing.
This conflicts with modern requirements for a 2V input.
I reckon I get away with it at -2V and 8mA with 145Va, just about, as I am at 1.16W dissipation (max dissipation is 1.5W)
It's a suck it and see situation, but these tubes are cheap so if I get hardness because of them, it is not that big a disappointment. I can always use something else to drive the outputs.

[JamesD]

Must admit I didn't try plotting op points when I listed them...

It is amazing how many small valves have to run near there dissipation limits to get them to work!

J

[Cressy Snr]

No worries James.

I'm sure you'll agree that thrashing the nuts off a cheap 6J6 is in no way the same as thrashing an ACP or MH4 to within an inch of its life. The need to thrash, in this case does not bother me at all.

[JamesD]

Indeed!

Changing one per week isn't difficult to contemplate

[Paul Barker]

No worries James.

I'm sure you'll agree that thrashing the nuts off a cheap 6J6 is in no way the same as thrashing an ACP or MH4 to within an inch of its life. The need to thrash, in this case does not bother me at all.

It is also great fun thrashing GM70's

[kuroguy]

My favorite little amp is a SE 6AU6/6AQ5 amp i built from scrap box parts. I think you're gonna like this one.[/img]

[Cressy Snr]

My favorite little amp is a SE 6AU6/6AQ5 amp i built from scrap box parts. I think you're gonna like this one.[/img]

Aye... I still have to get off my backside and build this one.
Our Anthony has some 1.5mm steel plate that could be very nicely bent
to make a chassis for one of these. 6AQ5s would have looked frankly stupid on the big steel sheet I used for the KT120 amplifier.