6HF5 sweep tube single ended amp

[Tony]

A mu-stage would cover a lot of bases for you perhaps?

that crossed my mind, i have a pair of 12by7's and a 12bh7 with which to make an allan kimmel mu stage! the author claims 280 volts peak to peak output and low output impedance to boot! at a supply of 350 volts!

but i also wanted to make my amp as simple as possible!

well, i would like to find out how the circuit will response out!

thanks,

[kevinkr]

The zener clamps between gate and source of your mosfet are drawn in series not back to back as they should be.

Kevin

[EC8010]

Although a mu-stage produces admirably low distortion, it isn't good value for money in terms of output swing vs HT voltage. I'd stick with your existing circuit and test it to find out whether it has enough poke for your output stage. Don't forget to put the gate-stopper right on top of the MOSFET gate - you really don't want that MOSFET to decide to whistle at umpty-ump Megahertz.

[Tony]

Originally posted by kevinkr
The zener clamps between gate and source of your mosfet are drawn in series not back to back as they should be.

Kevin

Originally posted by EC8010
Although a mu-stage produces admirably low distortion, it isn't good value for money in terms of output swing vs HT voltage. I'd stick with your existing circuit and test it to find out whether it has enough poke for your output stage. Don't forget to put the gate-stopper right on top of the MOSFET gate - you really don't want that MOSFET to decide to whistle at umpty-ump Megahertz.

guys, thanks for the inputs, will consider them.

[tubelab.com]

I would consider using a CCS IC like the Ixys 10M45 instead of your transistor based CCS, if you can get these chips. They are rated for 450 volts across the chip, and I have used them successfully in driver circuits that operate at 500 volts using a 45 triode. That circuit provides over 400 volts peak to peak to the grid of an 845 operating A2. Frequency response goes out to 345KHz. I have used it on an 833A also where the grid can draw over 100mA on peaks. I like the 2SK2700 fet because of low gate capacitance, and it sounds good. The circuit is on my web site:

http://www.tubelab.com/845SE.htm
http://www.tubelab.com/powerdrive.htm

I use a 45 tube because it will deliver 400+ volts cleanly. The 6BL7 works good too, and a 6SN7 probably would, but I haven't tried it. I don't think that any of these tubes have enough gain to use in a two stage amp. The 12AT7 might have enough gain.

[Sch3mat1c]

A 6AU6, or given the SS load, something beefier like 12BY7 or 6CL6, ran in pentode mode, would make a better driver. You'll need a whole lot more gain to drive that thing with NFB.

Figure a 10x reduction in gain by NFB (20dBV). 1V input for max. power output needs a gain of what, 70, to drive the output? That's not even 12AT7 already. Then you need another 10x, for a total of 700x gain, which is even out of the range of RC coupled pentodes. Cascading the two 12AT7 triodes might suffice (gain circa 40 for each = 1600 total).

Tim

[Tony]

hey tim,
what i had in mind was large voltage swing from the input stage, to be able to swing the screens of the output tubes. actually the NFB was an afterthought and not really a design goal, that is why the switch!

@tubelab.com,

unfortunately, those 10m45 devices are not available here,

will try out the circuit incorporating suggestions of SY and kevinkr!
will report findings.

thanks to all.