Reply
 
Thread Tools Display Modes
  #1   Report Post  
Jon Yaeger
 
Posts: n/a
Default A mystery to me

OK, I admit that my understanding of tube electronics is rudimentary at
best. But one thing really puzzles me.

On a Williamson UL PP amp, the screen is at about the same DC potential as
the anode (as measured with a DVM). I thought that the screen had to be at
a lower potential for normal PP operation.

What's up with that?

TIA,

Jon

  #2   Report Post  
Troglodite
 
Posts: n/a
Default

On a Williamson UL PP amp, the screen is at about the same DC potential as
the anode (as measured with a DVM). I thought that the screen had to be at
a lower potential for normal PP operation.

What's up with that?


A reasonable question. No doubt you'll get a multitude of answers from the
folks here, but here's a simple one:

If you tied the screen directly to the plate, you'd have - a triode. What the
UL tap on the transformer does is put some signal on the screen, but not the
full signal that appears on the plate, so the tube runs somewhere between a
triode and a pentode. It's actually a lot more complex than that, but that's
the basic idea.

Note that some tubes cannot be operated in this mode due to screen voltage
limitations. The 6146 is an example. If you wanted to run these in UL mode,
you'd need a separate winding on the OP trans.



  #3   Report Post  
Tim Williams
 
Posts: n/a
Default

"Troglodite" wrote in message
...
Note that some tubes cannot be operated in this mode due to screen voltage
limitations. The 6146 is an example. If you wanted to run these in UL

mode,
you'd need a separate winding on the OP trans.


Well, it's rated for triode mode, but it'd be much better to have a seperate
25% winding, or parafeed from the same tapping. (I presume you'd want a
similarly lower AC screen voltage when reducing DC voltage. Example, +600V
plate and +300V screen, with around (theoretically) 500V peak swing on the
plates, would give 250V peak swing to the screens, nearly cutting them off,
i.e. triode mode, if ran from a typical approx. 50% UL tap.)

Tim

--
"I've got more trophies than Wayne Gretsky and the Pope combined!"
- Homer Simpson
Website @ http://webpages.charter.net/dawill/tmoranwms


  #4   Report Post  
Tim Williams
 
Posts: n/a
Default

"Jon Yaeger" wrote in message
...
I thought that the screen had to be at
a lower potential for normal PP operation.


Nothing says you can't run +150V plate supply and +250V screen supply,
however you won't get much power output, distortion will be high and
efficiency (especially Ip/Is ratio) will be low.

Tim

--
"I've got more trophies than Wayne Gretsky and the Pope combined!"
- Homer Simpson
Website @ http://webpages.charter.net/dawill/tmoranwms


  #5   Report Post  
Fabio Berutti
 
Posts: n/a
Default

As some other RATs have already pointed out, this is the so-called
ultra-linear arrangement, which is used by the largest part of HiFi amps
using PP tetrodes, particularly these days because it is getting difficult
finding OPTs fitted with cathode windings suitable for "distributed load"
stages like McIntosh or similar. I suppose You can find the original
article written by Hafler back in 1950something describing his findings
about this arrangement.
The main points about UL are the following:
- G2 is operated at the same DC voltage of the anode (in fact 2-3V higher
due to the OPT winding resistance), therefore this should be checked when
swapping tubes
- for the very same reason, tubes allowing a high G2 voltage are best
suited to UL (ie. EL34, 6L6GT, or other types where the max V is not very
different between G2 and A)
- the grid is fed back some signal, and this (negative, local) feedback
contributes to reducing output impedance and distortion. In fact, during
operation, the applied load creates a very complicated situation which is
different from the one measured in static conditions. Look for the old M-O
data sheets for the KT88, they show the anode curves in UL mode: they're
definitely neither "pentode" nor "triode".
- the behavior in terms of distortion products is half way between triode
and pentode, depending on the primary tap position
- construction is easy, efficiency is very good and so is the sound (IMHO:
all my tetrode amps are UL, including some SE types, and I think they're
good, but... I'm not an unbiased listener)
- a damping resistor should always be connected between G2 and UL tap

As per Your understanding of electronics, with that Williamson You managed
to get one of the finest pieces of tubed gear ever designed: You don't
necessarily need to be a Nobel winner to enjoy it, as far as You have ears.

My compliments

Fabio


"Jon Yaeger" ha scritto nel messaggio
...
OK, I admit that my understanding of tube electronics is rudimentary at
best. But one thing really puzzles me.

On a Williamson UL PP amp, the screen is at about the same DC potential as
the anode (as measured with a DVM). I thought that the screen had to be

at
a lower potential for normal PP operation.

What's up with that?

TIA,

Jon





  #6   Report Post  
Patrick Turner
 
Posts: n/a
Default



Jon Yaeger wrote:

OK, I admit that my understanding of tube electronics is rudimentary at
best. But one thing really puzzles me.

On a Williamson UL PP amp, the screen is at about the same DC potential as
the anode (as measured with a DVM). I thought that the screen had to be at
a lower potential for normal PP operation.


The screens don't have to be at a lower than anode potential.

But with pentode/beam tetrode operation its best that Eg2 *is* no more than
the makers recomendations, so an 807 should never have more than Eg2 = 300v
regardless of the anode supply voltage.

But 807 were amoungst the first to be used for UL amps with 450v
applied as the anode supply.

With UL, the screen gets signal volatge applied from the anode circuit via the
OPT,
and the dissipation of heat in the screens is much reduced because of the
signal present.
With triode op of 807, the G2 diss is at its minimum.
With pentode op, when the anode voltage swings low, the G2 current peaks,
and when anode swings high, G2 current is reduced; the electrons seek the
most easy and attractive target.
But with UL or triode, the G2 voltage swings down with anode voltage,
so its current input is lower, and when anode swings high, the is still g2
current reduction.

Screen current input is about proportional to anode current input though.
So in class AB UL amps which are pushed hard with a load which is too low,
the screens are probably at some risk.

UL operation limits the B+ which can be used because of screen limitations.

A pair of 807 can be cajoled into making 80 watts in beam tet mode,
with 600 v anode supply but only 300v G2 supply.
EL34 can make 100 watts with 900v anode supply, but only 450 G2 supply.
Neither of these tubes can be used at the anode voltages for UL, unless
a special G2 winding is wound onto the OPT to give the wanted UL
signal voltage and supplied with an appropriate G2 supply.
This works OK with 6146 type tubes.

But if I was trying to
make a high voltage output stage I'd either stick to
beam T or pentode op, or use cathode feedback windings, which are more
effective
at reducing the thd and Ro of multigrid tubes, and you then have the
option of setti9ng g2 at the lowest efective value which will keep the tubes
happy,
and the screen current at its minimum, as well as reducing the wanted G1 bias
to a minimum.

EL34 and EL84 were designed to be able to run with equal anode and G2
voltage supplies, with 450v limit for EL34 and 350v limit for EL84,
but I have seen each used with 20% higher voltages for UL amps.

Screens are grids which draw current, and they run warm, and will run red hot
if their current is eccessive, and then start emitting electrons of deforming
in shape and alignment
and ruining the tube function.

Patrick Turner.



What's up with that?

TIA,

Jon


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
Can anyone help me with a pair of mystery Wharfedale boxes please? Tim Pro Audio 5 March 22nd 05 05:55 PM
New tape could solve JFK mystery JoVee Pro Audio 0 August 5th 04 02:46 PM
eBay: Mystery laser unit, looks like a special sensor Bob Hedberg Vacuum Tubes 3 May 31st 04 05:30 AM
Knight Tube Tester Mystery ???? Mista T Vacuum Tubes 9 February 20th 04 07:42 PM
Subwoofer miniplug mystery Arthur Kimes Tech 1 September 6th 03 08:20 AM


All times are GMT +1. The time now is 05:00 PM.

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"