For a pentode, the 6V6-ish 1619 is a good valve. I linked its sheet below. It has a few very useful properties for this project. One is its filament. It is a 2.5V/2A directly heated cathode. Very similar to the 2A3. Matter of fact, change the sockets from the 1619's Octal to 2A3 UX4 and you're off to the races.

Next, it can easily be supported by the Guinevere power iron on a mono-block basis. Since we're likely using a custom OPTx, it will be fairly easy to arrange a means to take full advantage of both types of final.

New production 2A3 is fairly easy to come by, and NOS 1619 is even less expensive. Both will yeild similar output power, at between 7 and 10 watts.

Any design specifics we need to include? It seems that this ought to be as flexible a project as is possible so that changes can be made as testing and tweaking require.
regards,
Douglas

regards,
Douglas

• http://www.mif.pg.gda.pl/homepages/frank/sheets/049/1/1619.pdf

There are of course a whole column of folks with less than honest supplies of snake oil to sell.
regards,
Douglas

I checkesd several web sites.

Antique Electronic supply has them for $9.60 ea.
Tube Depot for $13.95 ea.
Fair Radio Sales in Lima has them for $7.00 ea.

None of these listed any quantities on hand.

I like the idea of it being directly heated. Been interested since I saw Gary Pimm's 47 amp. If we used your E-Linear topology, what are we looking at as possable drivers?

Keep it com'in...

Joe

The front end should be a cascode diff amp. Two 5687's are a fine choice. They have low enough plate z to allow an Ec2 of ~70V, and gm for enough gain. Figure B+ of ~300, and plate loads of 25-30k.

Special voltage ref for Ec2 to the common cathodes instead of the traditional ground. Building a voltage reference is no biggie. Just assume it is as I say for the time being....
regards,
Douglas

We could also do a KT88 monoblock. Full U-L or get more complex with a seperate g2 supply and higher plate voltage, CFB winding v. E-Linear with a faux pentode diff amp to tame its output z issues. More power does have its attraction. Also, it does not matter to me that there are a few designs going on at the same time.

Everybody does not have to build the same amp. I can make adaptations/alowances for almost anything within reason. There is a limited amount of designs I have acutually built. If you( colectively ) are willing to step out *WITH* me to tackle a variant/improvement I'd be quite happy to do that.

Alright to hell with simple. We're building 813's. Filament requirements are 10V/5A per valve. We'll run ~400V for g2 and 900-1kV on the plates.

Just as soon as I can get a proper OPTx designed for 20-20 @ 100W or so, and we'll be in business. I know where I want to start, but time is creeping quickly. Cathode feedback like a Quad II, with plate taps for E-Linear drivers. I am most of the way done. I don't see anything jumping out of the Long Grass at this point. It is debatalbe wether or not further design work is warranted, or if I should just make one and see if it works as intended, and modify as required.

I think I need to plot some g2 curves for that valve and see what g2 voltage I want to run pentodes at for a 1kV plate supply. If only Maggies were more efficient, this mess would not be required....
regards,
Douglas