NetNews Usenet Archive 1992 #27/NN_1992_27.iso/spool/alt/guitar/10138

Internet Message Format | 1992-11-23 | 6.7 KB

Path: sparky!uunet!mcsun!uknet!glasgow!dam
From: [e-mail] (David Morning)
Newsgroups: alt.guitar
Subject: Re: Power Tubes (long) was Built your own amp?
Message-ID: <[e-mail]>
Date: 23 Nov 92 13:43:12 GMT
References: <[e-mail]>
Organization: Glasgow University Computing Science Dept.
Lines: 131



[very long reply..hit 'n' now if you're not into valve/tube construction :-)]

Neal writes:

>In article <[e-mail]> [e-mail] (David Morning) writes:

>>The purpose of 'beaming' is to improve the low level linearity of a multi-grid
>>valve. Most pentodes do it.
>>It's fairly common practise in the US to connect the extra screen grid
>>internally to the cathode whereas in the UK all the grids are brought out to
>>the pins and the cathode connection made externally. That's the only reason
>>an EL 34 appears to have a true screen grid, a 6550 has one too, it's just
>>connected to the cathode inside the valve to save you the trouble.
>>
>>Maybe your thinking about a tetrode like the KT88.
>>.
>>
>>>EL-34/6CA7's. The American made GE and Sylvania 6CA7/EL-34's were beam
>>>power tubes and not power pentodes.
>>
>>I don't know about the US firms but I'm sure Mullard would be extremely
>>surprised to discover that their EL34 wasn't a pentode.

>Screen grid connected to Cathode ????? HuH ???? Here's the terminology
>I'm using (this is from ARRL books, these are the US terms. Your UK terms
>must be different from what we're familiar with over here):

I beg your pardon, I'm getting my translation terminology in a twist.
You're right, that should read suppressor grid. :-)

>    All the 6550's I've ever seen (or taken apart) have no third grid
>    ( suppressor grid is US terminology). In it's place, they have a much more
>    solid piece of metal, like a plate, (the "beam" electrode) but with holes
>    cut in it as little "windows" thru which the electron beams pass. This
>    electrode is usually connected internally to the cathode like the 3rd grid
>    in a lot of pentodes. Having this kind of electrode (instead of a real
>    "grid") is why we refer to it as a "beam power tube" even though it is
>    technically a variety of pentode.

Ah! Hang on a minute - you're talking about a different technology approach to
deal with the same problem, namely secondary emmision. That's the main job of
the suppressor grid.
That little "window" also acts as the suppressor grid - or rather the metal
around it.

The purpose of the suppressor grid is to try and prevent secondary emitted
elecrons from reaching the screen grid. There are inumerable ways of doing this,
either with coiled wire, as in the EL34, or a solid sheet of metal as in the
6550. Of course if you adopt the 6550 approach, you have to leave a hole in
the sheet of metal to let those electrons you want to reach the anode get
through.

The 'beaming' aspect is usually done with a couple of 'wings' either side of
the cathode. The type of 'wing' fitted varied from manufacturer to manufacturer
as well as country to country. In its basic form the beaming plates looked a
bit like this - missing out all the screens:-

         /      \
        |  KKKK  |<---Beaming plate
         \      /

..where KKKK is the cathode.

By connecting the beaming plate to the cathode, it takes up the same potential
and effectively repels any electrons emitted by the cathode back into the
mainstream which will be heading up or down in this diagram.

However, there is a problem which relates to multi grid valves, namely that
of secondary emission from the anode striking the screen grid.
While the beaming plates will help protect the screen grid from *some* of
these much slower electrons, it's not ideal and still leaves a great deal of
non-linearity. Thus an extra grid, the suppressor grid was introduced, which
was at the same potential as the cathode and thus would repel the secondary
electrons back to the anode.

How you approach the construction of a suppressor grid varied widely.

Mullard and Phillips opted for a spiral of wire between the screen grid
and the anode in the case of the EL34. Since it was going to be connected to
the cathode to maintain the proper potential (it's proper potential is 'less
than the instantaneous potential reached by the anode at any time in its
potential excursions') but wanting to leave the options open for the valve
in situations where a connection *might* not be appropriate, they brought it
out to an external pin. As it happened, virtually everyone tied it to the]
cathode.

In the case of the 6550 a quite different approach was adopted.
Since the suppressor grid was almost always connected to the same potential
as the cathode it made sense to connect it internally, however, since it was
at the same potential as the beaming plates why not just extend the beaming
plates round the front with a small hole in it rather than wind a seperate
spiral and then connect it up? Cheaper to manufacure and the 'wing' extentions
act as the suppressor grid repelling secondary emission.
So technically it is a true pentode in its method of operation i.e. it does
have a third suppressor grid, it's just a little bizarre in its design.
Interestingly RCA, in their data manual, list the 6550 as a 'beam power tube'
...and show it's schematic diagram as a pentode!

I've never known anyone to connect up the suppressor grid to anything other
than the cathode, but then again, I'm sure someone will prove me wrong. :-)

>    for a little variety. I've thought a lot about just using the EL-34's but
>    I'm hitting the output tubes with nearly 725 volts DC and I was afraid
>    that'd be a bit rough on EL-34's. (I can vary the 1st grid bias from
>    anywhere between -25 to -75 volts). I guess I could cut the B+ down to
>    half and use the EL-34's on 350volts since I really don't need all the power
>    output.... I'd rather sacrifice some raw output power for better output

That shouldn't be any sweat *if* it's the real McCoy EL34. They could handle
an 800 volt anode potential. Mullard published a design for a 100watt Class
AB1 Ultra Linear amp using 2(!) EL34s and an 800volt HT.
They could handle 1000 volts cold :-)
The current crop of Chinese EL34s cave in at around 500 volts.
Currently EL34s are being made again in Britain by a company called Selmar.
They are claiming the old Mullard specs, i.e. 800 volt HT.

I've currently got a pair of old Mullards running 550v HT with no problems
giving out 50 watts or thereabouts.

>    Ob KT-88's> I've got access to four ancient KT-88's and a pair of KT-66's, I
>    wonder what they'd sound like in a guitar amp???

Rough...KT = 'kinkless tetrode" as someone mentioned to me in a not very
serious mood.
Gets kinda bumpy down at low HT voltages despite the best attentions of the
beaming plates.

Dave
--