Ping SY

[JoshK]

Hey SY,

Did you ever draw up a schematic of your 150w pp sweep tube amp, the one that was screen driven operating in mostly class B? If you did, I'd love to take it a look at it if you are willing to share. I'd love to study it for learning purposes.

P.S. I found myself in the possession of a small quantity of some 6FW5's.

[SY]

No. I'll try to remember to scan the drawing out of my lab notebook tomorrow. Remind me if I forget (which I tend to do a lot).

[JoshK]

Just a gentle reminder. I have a tendency to forget these things too.

[Tony]

i am likewise interested....

[SY]

Agggh, I haven't forgotton, it's just become more complicated. Thank you Microsoft, my scanner will not work with my laptop or my "normal" computer because of Vista incompatibility. So I have to scan on my XP machine, which is not hooked up to a network, then move it physically with a memory stick. If I can escape child-care tonight, I'll do it.

Basically, it's a SYclotron input stage using a 6SL7 and 2N5114 pFETs direct coupled to a long-tail pair of 6SN7. Cap coupled to source followers (IRF820) which are direct coupled to the screens of a pair of 6LF6. Control grids are shorted to the cathodes. IIRC, B+ was about 700V. Plate to plate load is 1k2, idle current is 3mA (not a typo). I would NOT try 150W sustained, but it has no problem hitting that in tone bursts without clipping.

I'd do things differently these days for the input and drivers, but the basic output stage configuration is a keeper.

[ErikdeBest]

Hi Stuart

I'd do things differently these days for the input and drivers, but the basic output stage configuration is a keeper.

What would you do different? I remember the thread where you presented the SYclotron driver, including some measurements on the excellent performance of this circuit. Is there something 'better' around to produce large, balanced and undistorted sine waves?

Greetings, Erik

[SY]

I'd probably do a long tail CCS for the first stage with something like a 6AQ4 or triode-connected D3a, then 6S4 for the second stage. Or a pair of dissimilar triodes like 6GF7 with the same topology.

What the SYclotron really brings to the table is a reduced input capacitance, the Achilles' Heel of high mu input tubes.

[JoshK]

This seems like a very interesting learning/experimenting project. Thanks for the details, SY.

I have a pair of 1.25K p/p OPTs from Onetics. Bud said they were 150w cores. They were meant for another project but I can use them to test this idea.

I will search for details on your SYclotron input stage. It sounds like what your suggesting now is something akin to the glass palace input/driver, but with different tubes and mosfet source followers instead of cathode followers. Does that sound about right?

It seems on paper the 6FW5 should be able to operate similarly to the 6LF6's. 6LF6s don't seem to be cheap anymore. I got 38 6FW5's for $4/ea. I've read that electrically the 6FW5 is the same as the EL36 and I've seen schematics for the EL36 with screen drive.

[SY]

I haven't tried them, but paralleled 6JN6 seem to work; that was the approach Berning took when updating the EA2100.

The tail in the LTP could indeed be cascoded MOSFET (especially DN2540 or similar depletion mode). It could just as easily be a cascoded bipolar. The swing only needs to be one-half the input signal, making the design a bit easier.

[Boris_The_Blade]

SY, have you experimented with pentodes in your LT pairs? I've gotten good, large signals out of a pair of 6CL6s...