Reply
 
Thread Tools Display Modes
  #1   Report Post  
Posted to rec.audio.tubes
Ian Iveson Ian Iveson is offline
external usenet poster
 
Posts: 960
Default mu feedback

Alex wrote:

Restoring an old 1953 radio I came across a funny way of
applying negative feedback in an audio 2 stage amplifier
(EBF80 + EL80). First stage was using a EBF80 pentode.
Input signal was applied to the control grid, while the
feedback from the speaker went to the screen grid via a
0.1uF capacitor of course.

Thus the gain of the amplifier was determined by the
internal mu (mu g1-g2) of the EBF80, which is in this case
is about 18.

I am wondering what order of THD one can expect from such
mu based feedback "divider"? Also the only way to reduce
gain is to find a pentode with a lower internal mu. I am
wondering if a pantode exists with the internal mu of say
10?

I tried to expand on the mu feedback concept. What about
arranging a cascode (series connection) of two triodes and
apply input signal to grid of the bottom one, and the
feedback -- to the grid of the top one? Then the mu of the
bottom one will solely determine the gain of the amp.
Advantage of this topology -- very high impedance on the
feedback input (grid of the top triode). It is convenient
to put a high impedance tone control circuit in the
feedback path.

I tried to find any low mu low power dual triodes. The
lowest mu i found was from 12AU7 (17), closely followed by
6CG7 (20). Again what order of THD can be expected from
such a trick?



Consider putting a triode in the feedback path. That would
tend to cancel out the loose input triode (or g1,g2, k combo
of the pentode).

This idea probably has a name, and a treatise to match,
which may in turn answer your questions.

Ian


  #2   Report Post  
Posted to rec.audio.tubes
Alex Alex is offline
external usenet poster
 
Posts: 65
Default mu feedback

Restoring an old 1953 radio I came across a funny way of applying negative
feedback in an audio 2 stage amplifier (EBF80 + EL80). First stage was using
a EBF80 pentode. Input signal was applied to the control grid, while the
feedback from the speaker went to the screen grid via a 0.1uF capacitor of
course.

Thus the gain of the amplifier was determined by the internal mu (mu g1-g2)
of the EBF80, which is in this case is about 18.

I am wondering what order of THD one can expect from such mu based feedback
"divider"? Also the only way to reduce gain is to find a pentode with a
lower internal mu. I am wondering if a pantode exists with the internal mu
of say 10?

I tried to expand on the mu feedback concept. What about arranging a cascode
(series connection) of two triodes and apply input signal to grid of the
bottom one, and the feedback -- to the grid of the top one? Then the mu of
the bottom one will solely determine the gain of the amp. Advantage of this
topology -- very high impedance on the feedback input (grid of the top
triode). It is convenient to put a high impedance tone control circuit in
the feedback path.

I tried to find any low mu low power dual triodes. The lowest mu i found was
from 12AU7 (17), closely followed by 6CG7 (20). Again what order of THD can
be expected from such a trick?

Regards,
Alex


  #3   Report Post  
Posted to rec.audio.tubes
John Byrns John Byrns is offline
external usenet poster
 
Posts: 1,441
Default mu feedback

In article , "Alex"
wrote:

"flipper" wrote in message
...
Funny you should mention it because I'm playing around with some dry
battery tubes and considered doing just that: screen feedback. I ended
up not including the pentode preamp in the loop, though.

Basically a toy at this stage but kinda cute, I think anyway.

http://flipperhome.dyndns.org/Batman.htm


Nice crcuit. I would remove (short out) redundant R11 (and reduce R10 to
maintain the overall gain). This would reduce AF voltage on the first
pentode 1N5GT plate.


Eliminating R11 appeals to my urge to minimize the parts count, but
beyond that what is the advantage reducing the AF voltage on the first
pentode 1N5GT plate? Is that the best thing to do, perhaps it would be
better to reduce the AF current on the first pentode 1N5GT plate?

--
Regards,

John Byrns

Surf my web pages at, http://fmamradios.com/
  #4   Report Post  
Posted to rec.audio.tubes
Alex Alex is offline
external usenet poster
 
Posts: 65
Default mu feedback


"flipper" wrote in message
...
Funny you should mention it because I'm playing around with some dry
battery tubes and considered doing just that: screen feedback. I ended
up not including the pentode preamp in the loop, though.

Basically a toy at this stage but kinda cute, I think anyway.

http://flipperhome.dyndns.org/Batman.htm


Nice crcuit. I would remove (short out) redundant R11 (and reduce R10 to
maintain the overall gain). This would reduce AF voltage on the first
pentode 1N5GT plate.

Also if you are concerned about a phase shift (mismatch) in the fhase
splitter at high frequencies, why not throw a small (trimmer) cap, probably
1...3pF, across R1 to cancel Miller effect in V2AA?


Thus the gain of the amplifier was determined by the internal mu (mu
g1-g2)
of the EBF80, which is in this case is about 18.

I am wondering what order of THD one can expect from such mu based
feedback
"divider"?


I can't help you there as I didn't breadboard it and, so, made no
measurements but I find it hard to believe it would be as good as grid
or cathode resistive feedback because you have, essentially, an
'active element' in the feedback path coloring the feedback signal.

I'm just guessing, though, and suppose serendipity could make it
coincidentally 'just right'.

Also the only way to reduce gain is to find a pentode with a
lower internal mu. I am wondering if a pantode exists with the internal mu
of say 10?

I tried to expand on the mu feedback concept. What about arranging a
cascode
(series connection) of two triodes and apply input signal to grid of the
bottom one, and the feedback -- to the grid of the top one? Then the mu of
the bottom one will solely determine the gain of the amp. Advantage of
this
topology -- very high impedance on the feedback input (grid of the top
triode). It is convenient to put a high impedance tone control circuit in
the feedback path.

I tried to find any low mu low power dual triodes. The lowest mu i found
was
from 12AU7 (17), closely followed by 6CG7 (20). Again what order of THD
can
be expected from such a trick?


You might try looking for some real old types. I seem to remember some
down around 9 or 10 but they weren't duals.

Out of curiosity, why are you looking for practically no gain?


Just to be able to have lower minimum gain. One can always increase the gain
by placing a divider in the feedback path, but it is impossible to get the
gain below the mu value.

Alex


  #5   Report Post  
Posted to rec.audio.tubes
Alex Alex is offline
external usenet poster
 
Posts: 65
Default mu feedback


"John Byrns" wrote in message
...
In article , "Alex"
wrote:

"flipper" wrote in message
...
Funny you should mention it because I'm playing around with some dry
battery tubes and considered doing just that: screen feedback. I ended
up not including the pentode preamp in the loop, though.

Basically a toy at this stage but kinda cute, I think anyway.

http://flipperhome.dyndns.org/Batman.htm


Nice crcuit. I would remove (short out) redundant R11 (and reduce R10 to
maintain the overall gain). This would reduce AF voltage on the first
pentode 1N5GT plate.


Eliminating R11 appeals to my urge to minimize the parts count, but
beyond that what is the advantage reducing the AF voltage on the first
pentode 1N5GT plate? Is that the best thing to do, perhaps it would be
better to reduce the AF current on the first pentode 1N5GT plate?


Reducing AF voltage on the pentode plate will reduce AF current (and THD) as
well, because of reducing of AF current flowing through R8...

However, R11 decouples C9 + 1N5GT plate capacitance + wiring capacitance
from the grid of V1AA, thus helping stability. Thus, you are right, R11 may
be beneficial, and I am retracting my advice. Let R11 be there.

Another thing which is not "right" are C7 and C8, compensating OPT leakage
inductance. Instead of these caps, series RC circuits should be used, while
R and C are empirically adsjusted (as taught by our GURU and MASTER Mr
Partick Turner) for the best reproduction of a square wave. (For a start,
use R abou half of the optimum load Rpp). Zobel-like RCs will help
stability. I have been putting such Zobels in amps since the MASTER brought
it to my attention. With very good results. Stability (phase) margin never
hurts.

Regards,
Alex





  #6   Report Post  
Posted to rec.audio.tubes
Alex Alex is offline
external usenet poster
 
Posts: 65
Default mu feedback


"flipper" wrote in message
...

However, R11 decouples C9 + 1N5GT plate capacitance + wiring capacitance
from the grid of V1AA, thus helping stability. Thus, you are right, R11
may
be beneficial, and I am retracting my advice. Let R11 be there.


Hmm. I'm not sure I follow. If it weren't for FB R11 wouldn't normally
be there anyway so you'd have the C9 + 1N5GT plate capacitance +
wiring capacitance coupling. So why is FB to that point any worse than
if, say, you could take it to the pentode cathode (meaning the grid
and capacitances are in the loop)?


In short, if you worked with op-amps, you know that loading "the summing
junction" (inverting input) with capacitance is generally a bad thing for
stability. In your case the grid of the triode where the feedback R10
resistor is connected -- is the sort of the summing junction (inverting
input). It is better avoid capacitive loading of it.

However, in your case since you have a small phase advancing cap in parallel
to R10, loading the summing junction is not that bad. In any case, if you
remove R11, it is good also to remove this 47pF (or reduce it).


  #7   Report Post  
Posted to rec.audio.tubes
Alex Alex is offline
external usenet poster
 
Posts: 65
Default mu feedback


"flipper" wrote in message
...
On Fri, 19 Mar 2010 22:45:00 -0700, "Alex"

This has been quite useful because it got me to rethinking about what
I'd done. One of the initial problems was motor boating and I
attributed that to C5, needed to block bias voltage, but now I'm
wondering about that A- B- bias string (shown in the battery
eliminator). I mean, they're all under the same filter cap and will
shift together. Sounds to me like a feedback path.

I suppose I could put multiple caps in to break up the synchrony or,
maybe better, use series diodes for the first few stages to make that
bias 'fixed'.

Whatcha think?


Motorboating is low frequency positive feedback. Typically through power
supply.
I think in your case C5 in the amp circuit has nothing to do with it -- time
constant is soooo large. But if you have concerns -- incerase it further to
2.2...3.3uF. IMO this increase will not anything significant.

Most likely the motorboating is caused by C6 in the power supply unit.
Increase it to 470...1000uF.

Regards,
Alex


  #8   Report Post  
Posted to rec.audio.tubes
Alex Alex is offline
external usenet poster
 
Posts: 65
Default mu feedback


"flipper" wrote in message
Most likely the motorboating is caused by C6 in the power supply unit.
Increase it to 470...1000uF.


Zowie, a thousand? That pole's bigger than the 'huge' C6 one

I do think the problem is related to the bias string but, as I
mentioned in a previous post, I tried adding a filter to the 'middle'.
That struck me as similar to a traditional B+ preamp filter since
things look sort of 'upside down' with the virtual ground but it made
things worse.

I did some more looking and I'm beginning to think I need to decouple
HT to A ground because B+ will shift just like bias does.

I think my situation is a little different than the typical old style
'battery radio' because they were usually Class A and I'm running
Class AB1.


Firstly increasing C6 from 47uF to 470uF is not about the pole, but about
the impedance. Say motorboating frequency is 10Hz. 47uF cap would have 400
ohm at this frequency and will not do anything. LF plate current variation
will directly be converted to bias modulation, which in turn is applied to
the grid of the first stage. Increasing C6 reduces this effect. Ideally a
regulator or a Zener shall be used for the bias. The Zener should be set to
generate maqqximum bias for the output tubes. others can be fed through
resistive dividers. In this case the resistors in these resistor dividers
should be relatively large (tens of Ks), so they do not draw current away
from the Zener. Even Zener is good to bypass by a large electrolytic.

A separate rectifier for the bias would be perfect. Better not to create
problems in the first place, rather than struggle to overcome them later.

Secondly, you are right, the motorboating can be caused by the feedback to
the anode of 1N5GT via supply rail. In this case I am not sure whether R11
does any good. It only increases the impedance of the supply rail at very
low frequencies. Pure emitter follower Q1 has less output impedance than
R11.


Reply
Thread Tools
Display Modes

Posting Rules

Smilies are On
[IMG] code is On
HTML code is Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
Feedback Lost'n Found Vacuum Tubes 16 October 12th 06 04:34 PM
ADK- GT-2- any feedback? [email protected] Pro Audio 2 December 6th 05 05:13 PM
16 ohm Feedback Tap west Vacuum Tubes 11 March 19th 05 03:52 PM
Need Advise on Feedback / Feedback Eliminators Marshall Pro Audio 23 June 26th 04 06:05 PM
Need Some Feedback Nick Miller Pro Audio 0 October 31st 03 03:27 PM


All times are GMT +1. The time now is 09:53 AM.

Powered by: vBulletin
Copyright ©2000 - 2024, Jelsoft Enterprises Ltd.
Copyright ©2004-2024 AudioBanter.com.
The comments are property of their posters.
 

About Us

"It's about Audio and hi-fi"