DCAudioDIY.com

Since I see that Paul ordered an SE triode KT-88 amp, I'd be careful before dropping in an 807, without taking some quick measurements on how hard it's running. Paul, what amp did you order?

The amps have three adjustments to A range of tubes from 6v6 to kt120

Oh, then you're fine. In fact this is a very fun alternative to the other tubes, and cheap as well. I think if you set the amp for 6L6 operation you'll be fine. I would avoid the Russian versions, affordable as they are. Not as sweet as the NOS types. You can get a nice matched pair for $50 or less on eBay.

An interesting adapter, and very good information from Grover: thanks!

_________________
- Guy

_________________
Walt

Are you saying that they can handle the higher plate voltages as kt88

_________________
Walt

My RCA transmitting tube manual gives a G2 rating of a maximum 300V for both CCS and ICAS for the 807. If ultralinear this limits the plate voltage to 300V as well. Sooooo, the 807 would not be expected to provide reliable service for your amp, Paul. I recently reworked an 807 amp that Jim G has and it turned out quite well according to Jim. I used a separate (less than) 300V fixed voltage for the screen G2 and over 600V on plate. Also autobias with around 10mA idle current per tube.

By the way, the screen voltage is a more important parameter for determining tube stress than plate voltage. The reason this is so is that the screen voltage plays a more important role in setting up the fields in a tube, and it is largely the screen voltage that determines the needed negative G1 (input grid) bias voltage. This is why I use very high plate voltages which gives lots of power with very low screen voltages to get good reliability. I can take full advantage of this with the ZOTL circuitry because it doesn't have the voltage breakdown limits imposed by an output transformer. But also to take full advantage I need tubes with plate caps or ones that have the plate connection on the base that is spaced far away from other connections. An example is the 6JN6 tube with pin 7 being the plate. Pins 5,6, and pins 8, and 9 are not used. My highest-voltage designs that give the highest power and efficiency use G2 as the input grid and connect the normal input grid G1 to the cathode. With plate voltages around 1300V the G2 bias is actually slightly negative for 5mA idle current.

over my head so this is something I will not try
I just saw this converter widget and thought I would ask the community

David, with all due respect, the 807 was used in thousands of Williamson amplifiers in the early '50s after Sarser and Sprinkle published their American version of the British original. It appears to have performed very reliably in that circuit both in triode mode and "ultralinear" as long as you don't exceed 400Vp/G2. Sarser and Sprinkle operated them at about 360Vp in triode mode and 50mA, but I haven't had any problems running them a bit higher. It's also true that it was soon replaced by the 5881, which allows for a G2 max of 400, and was also safer and more convenient for consumers, but in all my reading about the early Williamson circuits, I've never come across any reference to failures with the 807s. I just restored a vintage homebrew Williamson that used 807s and an original Partridge transformer that probably dates from around 1950, and by all accounts ran for years. There were no signs of tube failure that I could see. I myself have used them for many hours in my Williamsons without any problems, bu of course there's always the chance I'll be proven wrong.