Slight hum problem

[DeathRex]

I have a Kenwood 1100U, but that don't matter. I kept getting a hum noise on the anode of the 6AN8, more on one channel than the other. I ripped out the 6AN8, put in Shannon's ST35 front end with a 12AX7(6N2P) in one 6AN8 socket, and a 6CG7 on the other socket. Now I'm getting just slight noise in one channel, and a nasty signal on the other. I removed the 6CG7 and it got a little bigger. I grounded the grid with a short wire soldered to it, and it didn't change. I removed the 6N2P and both cleaned up, but still had the noise from B+, a little less than the first picture.
Both pictures are at 10mv.

Pin 1 - 117
Pin 2 - 0
Pin 3 - 1.113
Pin 6 - 121
Pin 7 - 0
Pin 8 - 1.08
]
[URL=http://s19.photobucket.com/user/DeathRex/media/Img_2165_zps809dccb0.jpg.html][/URL

[DeathRex]

I got it down to coming from the feedback, and eventually from the 7591s. I have 2 extra and switching several around, I now have very little hum. Why would 7591s give me hum and still be good?

[Geek]

Why would 7591s give me hum and still be good?

Leaky heater-cathode.

[DeathRex]

I did more tests, 2 of the 7591s when combined in the right order, gives me 17mv ripple on the feedback line. Reverse their order and I get 12mv. On the other channel I'm getting 5mv. I can move both tubes from right to left and the ripple follows. I tried the other 2 7591s and can only get them down to 7mv of ripple. I did test them on a TV-7 and all tested great 50-54 and on the shorts test, they all tested the same.

I wonder when the 7591s are gonna be on the dollar list?

[azazello]

For 1 $ You can buy hummm, of course.... Try to rectifier heaters AC U.

[Shannon Parks]

As Gregg is saying, you have varying levels of noise from the heaters to cathodes. The trick is to match up pairs that can cancel each other out most effectively, but some will just be too far gone. It takes some luck to match up four old stock used tubes. A hum balance pot might help, too.

In my limited experience I think 7591s are very special tubes with plenty of that magical gm stuff, but they wear quickly. Little bulbs in a hot enclosed chassis with aggressive bias currents spelled their doom. Last I checked, modern production look out of spec with the transconductance but otherwise bias up and work.

Shannon