|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
203.129.49.2
In Reply to: RE: Tube pin straighteners ROCK and so do 50 year old French 12AU7's posted by mbnx01 on July 06, 2015 at 20:55:14
So, it's not a very good audio valve, at all.They do make good followers (ie with 100% local feedback) with a gain of about .95, for line-drive duty. I would never use one for gain duty.
If there is enough filament / heater reserve in the circuit you have, any reasonably high Gm <30 gain twin triode will better your french AU7s and any other 12AU7.
eg? 6CG7, 12BH7, 6BQ7....... You would need to change the circuit values and the pin connections, for the 6..7 types.
Warmest
Tim Bailey
Skeptical Measurer & Audio Scrounger
Edits: 07/07/15Follow Ups:
"I would never use one for gain duty."
AGREED! The non-linear 12AU7 makes a poor voltage amplifier. The type is satisfactory as either a cathode follower or "concertina" phase splitter. Degeneration, AKA local current NFB, saves the day in those situations.
BTW, if a unit uses 12AU7 sections as voltage amplifiers and extra filament current is not available {говно :> ((}, give plenty of thought to removing the Noval socket, mounting a Loctal socket, and installing the linear 7AF7. The 7AF7 data sheet happens to mention the 7N7, which is electrically equivalent to the 6SN7. ;> )
Eli D.
"The non-linear 12AU7 makes a poor voltage amplifier."
Just curious... do you feel the same about 12AU7s used in Williamson-type designs where the amp as a whole is "linearized" with NFB? I wonder how important individual tube linearity really is in such a case. Commercial manufacturers have used many tube types through the years that I wouldn't personally consider desirable in an audio amp, but apparently with little or no degradation to specs or sound. Again, I'm only referring to amps with NFB around those tubes.
--------------------------
Buy Chinese. Bury freedom.
In something like the Heath W5M, 'U7 non-linearity may well be of little consequence. Still, less open loop distortion has to be better. FWIW, focus on the HF error correction signal, which (if substantial) can induce slew limiting.
BTW, I calculate that the W5M has 800 mA. of spare heater current. Pull those damned 12AU7s and install pin compatible 12BH7s. Of course, minor Noval socket wiring changes allow use of the 6CG7/6FQ7 and the 6GU7.
Eli D.
" less open loop distortion has to be better. "
Yes, that certainly seems a logical conclusion. However, I was also thinking about this issue from the perspective of the non-linearities of pentode output stages that have to be assuaged by the loop. Relatively speaking, the 12AU7s may contribute so little to that mix as to be really insignificant. Of course, this might not be true if the outputs are triodes. In that case, the degradation created by the 'AU7s might be plainly audible. I admit that I haven't seen or built a Williamson design like that (triode outputs) in quite some time. The temptation of creating more power for free always sucks me in. It would be interesting to know whether the 12AU7's deficiencies are audible in a UL design like the one you posted. The reduction in GNFB that often accompanies UL might allow better tubes to shine through.
--------------------------
Buy Chinese. Bury freedom.
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: