EF91 Line Stage questions

[JDeV]

I recently got hold of 5 new Mullard EF91 valves and want to know:

1) Is it a good valve to use for Hi-Fi quality projects?
2) Are there any good line stage circuits,using this valve, that I may copy?
3) Can anybody design a nice line stage or buffer, using this valve - if it is worth to do?

Thanx for all assistance

 

[kopite]

Probably not as this valve is a pentode and usually used in IF stages.
The EF94 (6AU6) is a similar valve and can be wired as a very good triode.
You may want to have a look at this.

 

[Morgan L]

EF91 has a very high mu (70) in triode connection, which isn't desirable unless you want to use feedback (unnecessary in a line stage). I haven't seen any triode-ct curves, but it seems likely it would make a fair (1st) voltage amplification stage, as in a power amp. The plate resistance will be fairly low trioded (ca 10k), which is nice. But for line stage VA, no. If you're going to include a cathode follower on the output of the line stage, however, the EF91 ought to be about as good as anything in this function. Output impedance ca 150 ohm, and the high mu won't hurt linearity either (due to the 100 % degenerative FB of a CF). So don't throw them away! They also make good constant-current sources (ref. M. Jones' Valve Amplifiers). Not sure which op point might suit line stage CF use, but guessing 120-150 Vak and 5-8 mA or so (trioded, that is).

Morgan

 

[JDeV]

I found this article but can not see the relevant schematics. Can somebody maybe re-draw them?

http://members.lycos.co.uk/griffonconsulting/line_stage.htm


Thanx to all.

 

[Morgan L]

The schematic seems to be missing from the page. But there is an e-mail address on the index page. Maybe you could write the author and inform about the problem (or just ask for the schematic)?

griffon 'at' ntlworld 'dot' com

Do you need just a cathode-follower buffer (unity gain)? I don't have access to a server to upload to, but if you write me, I could probably draw something that might work with an EF91.

morgamp1 'at' yahoo 'dot' se

Morgan

 

[hacknet]

i`ve seen tons of mullard ef91s at a local surplus store too... i had a look at the curves and they don`t look too linear on top of that i have doubts about the microphonics. am i wrong? i`m think about them in a phonostage. arn`t these 70mu valves built for stages like these?

 

[ray_moth]

The EF91 was intended for RF and was used in the thousands 50+ years ago, including in some early computers. It is not a partricularly good tube for audio. Used as a pentode, it will be noisy. It also suffers from microphonics. However, as already stated, it can probably do duty as a CCS or CF.

 

[nastorrent]

Hi, I know I'm replying to a very old post but I came here from the search of "EF91 Audio" and ... I have a good lot of EF91 (and equivalents) looking for a good project. On other search, I've found this: An EF91 Audio Amplifier

Please, make some comments, suggestions, critics...

Thanks

 

[piano3]

I think it is pretty linear triode wired and high mu but I wouldn't use it to drive the grids of output valves without a follower in between. Probably ok for first stage of phono amp or a single stage mic amp with a follower output. Trying to use it as an output valve would be a waste of a transformer, in my opinion. Look at Morgan Jones's book for CCS use.

As a line stage, you could use split supplies, put an EF91 cathode follower, triode wired, up top and use one in pentode mode as a current sink underneath. +/- 130 volts should probably do it.

 

[leadbelly]

I found that power amp schematic you posted puzzling. I thought letting g2 get positive of the plate is a complete no-no, leading to runaway IIRC, but in that schematic g2 would be slightly positive of the plate a fraction of the time since g2 gets full B+, but the plate gets B+ minus OPT losses.

 

[PRR]

> I thought letting g2 get positive of the plate is a complete no-no

Only on True Tetrodes. Nobody has used them since 1926.

Note that ALL pentode power output stages have Vp swung WAY low of Vg2 half the time. Is not a problem. This is why we have G3. (Or the equivalent side-rods in "beam" tubes.)

 

[DF96]

Practical Wireless in the 1960s and 70s was full of valve amplifier designs. In most cases sound quality was not a major factor; choosing the same inappropriate valve for every stage in an amplifier is a sign of this.