DN2540 CCS questions

[RollE2k]

I have a Raleigh Parafeed SET linestage, and have been thinking about modifying this to balanced operation, and i have one question about this.
I don't have any big chassis so i will have a bit trouble getting room for the DN2540's if they need sinks for each phase, so i came up with an idea. Would it be possible to make a special cascode with three DN2540's with the two "lower" ones in parallel - feeding each half of the tube and then tied together at the bias resistor. This would make me only need one big sink for the upper DN2540. I have included a picture here so you understand what i mean.. Would this work? have anybody tried it?

[Zen Mod]

nope

sources of these lower two are short circuited , so they're in fact just paralleled

[RollE2k]

Is it possible to get the same result (3-fet cascode) with any other wiring?

[Zen Mod]

nope , if you want functionality

use 2 independent cascode cells

in any case - size of heatsink , dictated by thermal demands , is adequate for 2 DN's

edit : just started to edit my post when I saw that SY already wrote what I was intending

use lower ones as naked 220's

[SY]

Remember that the "upper" device is the only one that needs heatsinking- the lower ones can be in a TO92 package or a "naked" TO220 if space is at a premium. Also, because the drains are connected to B+, the two upper devices could share a common heatsink.

[zigzagflux]

Try this out:
Cascode CCS idea

Note the changes per the latest posts were done. Performance was excellent. I am mounting both devices to the same heatsink.

[RollE2k]

I was hoping to get rid of one of the DN2540's, but i'll have a look if there is room for both on the heatsink i have. Otherwise i'll probably scrap this idea and build a completely new amp later.