Cascading 12au7 gain stages

[mauiboy81]

Have any of you gentlemen done much with the 12au7? I was wanting to build something simple like the ef86 channel of a matchless or vox using what I have laying around. I have no ef86 but I do have 12au7s that I don't have much use for.... Or so I thought.

I was thinking that cascading one triode into the other would give me gain much closer to an ef86 than parallel or cascading 12ax7 triodes.

What component values have given you the best results?

Page 3 in this data sheet has a schematic for two sections in cascade. Would this be optimal for guitar?

http://www.drtube.com/datasheets/ecc82-philips1954.pdf

[Firestorm]

12AU7 is 20 mu, so setting them as mu amps gives a theoretical max of 400 cascaded. Probably can't all the way there, but you might find a nice sound. 12AU7s are higher current devices than 12AX7s. Sometimes I like that sound.

[mauiboy81]

I was looking at getting around 15 mu from each triode.

For the bias do I want to get around 4v on the cathodes?

What kind of sound does the higher current device have?

[Firestorm]

I'd put cermet pots on the cathodes and dial them in. Maybe split them. Even if you fully bypass each Rk, I'm not sure you can get 15mu with common cathode. Maybe more like 10, but I didn't run the numbers. Lazy.

IMO, higher current tubes sound "crunchier" cause the voltage swing is more pronounced.

[Gaz]

Merlin Blencowe actually has a good cascode circuit for a 12AU7 in his preamp book that will get you closer to a pentode, albeit with lower input sensitivity.

If you aren't interested in the book just check out some cascode circuits.

[mauiboy81]

Thanks guys.

I was initially looking in to cascode but saw something Merlin posted somewhere that with a normal size anode resistor gain would be more like 65 mu so I settled on cascading the triodes.

I took the schematic on the datasheet and built it with close values into my test amp. I used 220k plate on both, 2k2 cathode and 22uf bypass (shared), .022uf coupling between stages, 1M grid load, after the second stage I made the coupling cap switchable into a 1MA volume pot into the pi which has .022's,1M's,1k,39k.

I believe I had 97v on the plates and around 3.5v on the cathodes. It sounded good but not magical. It was pretty clean up to about 2/3 on the volume where it had a smoother distortion. I would like a bit richer swirly kind of clean. I will split the cathodes and maybe bias them a bit differently.

Any advice on plate voltages and bias that would add a little magic with this tube?

[tubeswell]

Two 12AU7 cascodes cascaded together?

[img:766:388]http://farm8.staticflickr.com/7133/8169 ... 1b46_b.jpg[/img]