[Ian Thompson-Bell]

I have found a reference to the need for a screen grid stopper resistor
in pentodes wired as triodes in order to stop parasitic oscillations:

http://www.webace.com.au/~electron/tubes/triode.htm

Can anyone explain how this happens and does it apply to preamps as well
as power amps?

Cheers

Ian

P.S. The grand children have gone home and peace reigns once more!

[tubegarden]

On Dec 25, 5:15�pm, Ian Thompson-Bell wrote:
I have found a reference to the need for a screen grid stopper resistor
in pentodes wired as triodes in order to stop parasitic oscillations:

http://www.webace.com.au/~electron/tubes/triode.htm

Can anyone explain how this happens and does it apply to preamps as well
as power amps?

Cheers

Ian

P.S. The grand children have gone home and peace reigns once more!



Hi RATs!

No, I cannot explain how RF oscillations are excited ... but, I can
explain how to determine if a screen, or grid, stopper is required.
Build the circuit with stopper(s). Listen and measure to your hearts
content. Note color of plate after using successfully for at least an
hour.

Remove one stopper. Re-test. If plate does not glow more than with
stopper, relax, and remove other stopper and retest.

We really do not need to know what frequency is being produced, only
that it affects operation or temp of plate.

If it ain't broke, don't fix it.

This applies to preamps, too.

Happy Ears!
Al

[Ian Thompson-Bell]

tubegarden wrote:
On Dec 25, 5:15�pm, Ian Thompson-Bell wrote:
I have found a reference to the need for a screen grid stopper resistor
in pentodes wired as triodes in order to stop parasitic oscillations:

http://www.webace.com.au/~electron/tubes/triode.htm

Can anyone explain how this happens and does it apply to preamps as well
as power amps?

Cheers

Ian

P.S. The grand children have gone home and peace reigns once more!



Hi RATs!

No, I cannot explain how RF oscillations are excited ... but, I can
explain how to determine if a screen, or grid, stopper is required.
Build the circuit with stopper(s). Listen and measure to your hearts
content. Note color of plate after using successfully for at least an
hour.

Remove one stopper. Re-test. If plate does not glow more than with
stopper, relax, and remove other stopper and retest.

We really do not need to know what frequency is being produced, only
that it affects operation or temp of plate.

If it ain't broke, don't fix it.

This applies to preamps, too.

Happy Ears!
Al


So, if I don't have glowing plates in my preamp then I don't need
stoppers, correct?

Ian

[Tom Schlangen]

Hello Ian,

if both anode and screen grid are available on pins at the base,
a straight piece of wire between both pins will do.

If triode-strapping a top-capped tube like, say, 807, you very
likely will need a screen grid stopper, maybe even an anode
stopper, depending on your wiring.

Always keep screen stoppers as low in ohmic value as possible,
means, not more in value than is needed to reliably prevent
parasitics.

Any non-bypassed resistance in the screen feed will produce
distortion in any connection (pentode, UL, triode-strapping),
because the Ia/Ig2 ratio is not exactly constant, especially
when the anode swings widely "down", entering the current
takeover area, screen current rising in a non-linear manner.

This non-linear screen current behaviour will produce an
accordingly non-linear voltage drop across the screen
stopper resistor, inevitably rising distortion.

If you want to learn more about triode-strapping of
tetrodes, pentode and BPT, you may have a look at the
presentation foils of my ETF06 lecture on that
theme.

See http://www.mynetcologne.de/roehren/etf06_lecture_e.html

Regards,

Tom

--
A consultant is a man who knows 40 ways to make love,
but doesn't know any women.

[Ian Thompson-Bell]

Tom Schlangen wrote:
Hello Ian,

if both anode and screen grid are available on pins at the base,
a straight piece of wire between both pins will do.


Excellent. They are available in my preamp design. I am using a couple
of 6AU6 pentodes strapped as triodes. The second stage is a cathode
follower and it is noisier than I had expected. There is no obvious
oscillation visible on a scope but I thought is was at least worth
eliminating the possiblity.

If triode-strapping a top-capped tube like, say, 807, you very
likely will need a screen grid stopper, maybe even an anode
stopper, depending on your wiring.


Ah, so I would guess lead inductance is issue.

Always keep screen stoppers as low in ohmic value as possible,
means, not more in value than is needed to reliably prevent
parasitics.

Any non-bypassed resistance in the screen feed will produce
distortion in any connection (pentode, UL, triode-strapping),
because the Ia/Ig2 ratio is not exactly constant, especially
when the anode swings widely "down", entering the current
takeover area, screen current rising in a non-linear manner.


OK

This non-linear screen current behaviour will produce an
accordingly non-linear voltage drop across the screen
stopper resistor, inevitably rising distortion.

If you want to learn more about triode-strapping of
tetrodes, pentode and BPT, you may have a look at the
presentation foils of my ETF06 lecture on that
theme.

See http://www.mynetcologne.de/roehren/etf06_lecture_e.html

I just tried this URL and got the following message (amongst others)

Sie haben eine Seite aufgerufen, die auf unserem System nicht vorhanden
oder vorübergehend nicht erreichbar ist.

Cheers

ian

[Alex]

I came across 6BW6 tube (noval) which is similar to 6V6 in characteristics,
but surprisingly (!) has an independent connection to the beamformer plates.
Perhaps those can be connected to the plate and screen grid and give the
ultimate triode connection performance without any risk of damaging the
screen grid because of current distribution towards the screen grid at high
current and low voltage.

Regards,
Alex

"Ian Thompson-Bell" wrote in message
...
I have found a reference to the need for a screen grid stopper resistor
in pentodes wired as triodes in order to stop parasitic oscillations:

http://www.webace.com.au/~electron/tubes/triode.htm

Can anyone explain how this happens and does it apply to preamps as well
as power amps?

Cheers

Ian

P.S. The grand children have gone home and peace reigns once more!

[John Byrns]

In article ,
"Alex" wrote:

I came across 6BW6 tube (noval) which is similar to 6V6 in characteristics,
but surprisingly (!) has an independent connection to the beamformer plates.
Perhaps those can be connected to the plate and screen grid and give the
ultimate triode connection performance without any risk of damaging the
screen grid because of current distribution towards the screen grid at high
current and low voltage.

Why would this be expected to be a problem in triode mode even with the
beam former plates connected to the cathode? I would think that if the
tube can survive operation in "pentode" mode, triode mode should be a
breeze?


Regards,

John Byrns

--
Surf my web pages at, http://fmamradios.com/

[Alex]

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

I came across 6BW6 tube (noval) which is similar to 6V6 in
characteristics,
but surprisingly (!) has an independent connection to the beamformer
plates.
Perhaps those can be connected to the plate and screen grid and give the
ultimate triode connection performance without any risk of damaging the
screen grid because of current distribution towards the screen grid at
high
current and low voltage.

Why would this be expected to be a problem in triode mode even with the
beam former plates connected to the cathode? I would think that if the
tube can survive operation in "pentode" mode, triode mode should be a
breeze?

Originator of the thread, Ian Thompson-Bell referred to some questionable
article
http://www.webace.com.au/~electron/tubes/triode.htm

In that article it was recommended for the triode mode of operation to
connect the suppressor grid to anode, if the suppressor grid is
independently connected to a pin of a valve. Reasoning was that it helps
current distribution towards anode.

I agree with you, that if a tube can work in a pentode mode with Ua Ug2
then it will be even easier to work in the triode mode at Ua = Ug2.

However, still space charge near the suppressor grid slightly affects
cathode emission because of finite "mu" between grid1 and grid3. My guess
this mu is of the order of 100...400. Connecting grid3 to anode would
disperse this space charge and MARGINALLY increase cathode current for a
given plate voltage.

There are quite a few pentodes with separate connection to grid3. However,
the only beam tetrode I know with independently connected beamformer plates
is 6BW6, and I believe these should be connected to anode for triode
operation.

That is why I drew attention to this tube for those who want to experiment
with triode connection. I do not know whether power dissipation on these
plates will be low enough not to overheat them.

Regards,
Alex

[mick]

On Sat, 29 Dec 2007 05:46:58 +0000, Alex wrote:

snip

That is why I drew attention to this tube for those who want to
experiment with triode connection. I do not know whether power
dissipation on these plates will be low enough not to overheat them.



I doubt if that would be a problem. You still have g2 linked to anode as
well, I suppose, and that will have far less dissipation capability than
the beam plates.

However, connecting the beam plates to anode causes another problem. It
prevents them working as intended - as a virtual cathode. That removes
the "beam" control and actually encourages electrons to hit the grid
support structures. IMHO that would probably reduce efficiency (these
valves have aligned grid structures) seriously and, possibly, increase
distortion quite a bit. I would like to see this tested - I don't have
either a 6BW6 or the necessary test gear though!

--
Mick (Working in a M$-free zone!)
Web: http://www.nascom.info http://mixpix.batcave.net

[Tom Schlangen]

Hi Ian,

See http://www.mynetcologne.de/roehren/etf06_lecture_e.html
I just tried this URL and got the following message [...]

Sorry, should be: http://www.tubes.mynetcologne.de/roe...lecture_e.html

Tom

--
The opposite of a correct statement is a false statement. But the
opposite of a profound truth may well be another profound truth.
- Niels Bohr

[Ian Thompson-Bell]

Tom Schlangen wrote:
Hi Ian,

See http://www.mynetcologne.de/roehren/etf06_lecture_e.html
I just tried this URL and got the following message [...]

Sorry, should be: http://www.tubes.mynetcologne.de/roe...lecture_e.html

Tom


That's better. Strangely enough I had already found your article and
downloaded it before I asked my question here.

Cheers

ian

[Tom Schlangen]

Hi Alex,

Connecting grid3 to anode would disperse this space
charge and MARGINALLY increase cathode current for a
given plate voltage.

It does, indeed.

When triode-strapping pentodes or BPTs, g2 to anode,
the difference in Ik when changing g3 (or beam plates)
between cathode and anode level typically is 10% or less,
hence well within individual tube production tolerances
anyway.

The effect really looks like moving the std anode curves
diagram very slightly on the horizontal (X, Ea) axis.

BTW, there are some people swearing that, when triode-
strapping a pentode or BPT, strapping g3 (or beam plates)
to anode _always_ will sound superior. Whatever that means,
it is utter nonsense since the very same effect can be
reached by very slightly lowerig Eg1. Again, we are
talking about effects so small they are swamped by
individual tube tolerances, anyway.

There are quite a few pentodes with separate connection
to grid3. However, the only beam tetrode I know with
independently connected beamformer plates is 6BW6, and
I believe these should be connected to anode for triode
operation.

I don't see any argument why this believe should hold true
for any technical reason.

Also, there are many more BPTs out there that have the
beam forming plates available for separate connection at
the base, notably BPTs intended for horizontal or
vertical deflection duty in TVs. For example, EL509/6KG6,
PL509/40KG6 comes to mind, being produced in two-digit
numbers of millions. These tube had no pentode anchestors,
but were designed as BPTs from the start.

Also, many BPT copies of real pentodes feature BFP pinout
to maintain 100% operational compatibility to the pentode
versions. EL83/6CK6 or PL83/15A6 comes to mind, which
originally was a very high linearity and bandwidth driver
pentode for TV tubes, but I estimate about 3/4 of all
EL83/6CK6 or PL83/15A6 ever produced actually are BPTs.

However, the only beam tetrode I know with
independently connected beamformer plates is 6BW6, and
I believe these should be connected to anode for triode
operation.

Another - probably the most famous - example of a BPT
copy of a real pentode with beam forming plates pinout
at the base is 6CA7, originally a BPT copy of the EL34
pentode, to circumvent the Philips pentode patent.

Regards,

Tom

--
Life: All in all a bad game, but graphics are really good.