#27 mu stage

[zivhackman]

Hello,
I'm currently working on my first desing,
its a #27 mu stage and the uper tube is D3a.
there is nothing spacial about this desing..
I dont have any simulation programs or enough experience to be confident about my calculations,
i would realy appreciate if you can take a look at my schematic and suggest corrections.

the D3a should see approximatly 175V @ 9mA
and the #27 - 135 @ 4.5mA

Thank you very much!
Ziv Hackman.

[ruffrecords]

You don't say what the intended application is.

If you want minimum distortion you should bypass the cathode resistor of the lower tube.

Cheers

Ian

[kevinkr]

In addition the 1uF screen bootstrap needs to be larger if you want to maintain good pentode follower performance to 20Hz.

The 27/227 is extremely microphonic and also has issues with filament to cathode insulation breakdown, if you bias the filament supply above ground with DC to help quiet things down be sure not to exceed 12 - 15V. You could just use ground referenced DC to heat the filaments. I recommend you find a good sized stash of 27 before you build this line stage.

You can download and install LTSpiceIV for free and add tube libraries. In the case where there is no model available you can use paint_kip for pentodes and paint_kit for triodes - find these at sourceforge.net You have some reading and googling to do..
(These programs will run in windows and linux)

[ruffrecords]

To be honest, I don't much see the point of putting a pentode as the top tube in a mu follower. Once the effective ra is about 10 times the triode ra the stage gain is very close to mu and the distortion pretty much reaches the intrinsic level of the bottom triode. So long as the top tube has an effective mu of at least 20 there'll be no noticeable improvement..

Cheers

Ian

[pieter t]

It is still worth reading what Alan Kimmel had to say:

Mu Stage Philosophy © 1993 Alan Kimmel

Pentode CF's do have advantages over triode CF's.
Drawback of these circuits IMO is that there are already two coupling capacitors before the signal comes out...

[ruffrecords]

Originally Posted by pieter t

It is still worth reading what Alan Kimmel had to say:

Mu Stage Philosophy © 1993 Alan Kimmel

Pentode CF's do have advantages over triode CF's.

Not really. They just present an even higher load to the bottom triode. Once you get past 10 times ra you have all there really is to have.

If you pick a triode with poor inherent linearity then no matter how good a CCS you put on top of it you won't get any better than the inherent linearity.

Better to choose a triode with very good intrinsic linearity and stick a simple triode bootstrap on top - the results will be much cleaner. The 6SN7 is a good candidate. See Morgan Jones book for more details.

Cheers

Ian

[kevinkr]

To Ian's point the D3A makes an excellent follower in triode connection.

The 6SN7/6J5 would also be good candidates but the much higher mu might be problematic as the gain triode - ok as a follower however.

If you are interested in a headache no worse than the 27 is going to give you - you might also want to consider the following DHT/DHPs.. In the old American DHT camp the 201/01A, 112/12 and the 26/226 are all worth a look, none are more microphonic IME than the 27, linearity is quite good too.

The Russian 4P1L DHP is also reputedly very good in triode connection as a line stage device, but is extremely microphonic. I have a bunch, but have yet to build anything with them. My comments on the 4P1L should be construed as hearsay as I have no direct experience working with the tube to date.

[pieter t]

Originally Posted by kevinkr

To Ian's point the D3A makes an excellent follower in triode connection.

Excellent yes, but pentode CF's are even "more excellent", so why not use it that way?

[pieter t]

Looking at Ziv's circuit I would not use that extra feedback; any reasons for it?

[pieter t]

Taking another close look it seems to me that the 33k resistor is not correct; it should have a value to provide the D3A with the right bias. Some 360 ohm should be good.