That 6AG5 tube looks pretty exposed internally, with a very high output impedance. Any charge building up on the inside surface of the glass could affect the present operation. I'm guessing that the carbon coating prevents that. The sound stage improves when past signal is not disturbing the present signal. The coating is probably required for audio. Do you really want a tube with such high Zout? Lower Zout tubes likely have the plate wrapped around the internals.
I made a little test board to evaluate 7 and 9 pin pentodes in four different operating configurations, pure pentode, UNSET mode, Saturator mode, and UNSET + Saturator mode. I ran 15 different types of 7 pin tubes and a couple 9 pin tubes in pentode, and U+S mode. I have not got to the 6AG5's yet. So far, the absolute "best of show" is a little 7 pin pentode that looks just like the 6AG5 or 6CB6 in construction.
Oddly, I did all of this looking for some cool sounding distortion for a guitar amp build but wound up with a great driver for a HiFi amp. It's seen here driving a sweep tube to 20 watts of output in SE UNSET configuration.
More details are scattered throughout this thread starting in post #107. The whole thread is really about breadboarding and prototyping and possibly worth scanning through. Somewhere in that thread I posted my vacuum tube "master chart":
https://www.diyaudio.com/community/...he-way-i-always-wanted-to-do-it.410333/page-6
Oddly, I did all of this looking for some cool sounding distortion for a guitar amp build but wound up with a great driver for a HiFi amp. It's seen here driving a sweep tube to 20 watts of output in SE UNSET configuration.
More details are scattered throughout this thread starting in post #107. The whole thread is really about breadboarding and prototyping and possibly worth scanning through. Somewhere in that thread I posted my vacuum tube "master chart":
https://www.diyaudio.com/community/...he-way-i-always-wanted-to-do-it.410333/page-6
