The PL81 TV line output valve.

[Philips210]

That's a strange looking valve John.
The base seems to be odd as well since the Philips/Mullard numbering system would suggest it should have a B9A base when there's an 8 in the first part of the valve number.

Regards
Symon.

[Synchrodyne]

Originally Posted by FERNSEH

The really interesting line output valve is the PL500. This valve was I'm certain was developed by Philips. This valve employs what was described by Mullard as the cavitrap" construction. A Philips name for the beam forming technique perhaps.

From the radiomuseum: http://radiomuseum.org/tubes/tube_pl500.html

There is a description of the cavitrap PL500 but it is auf Deutch.

As best I can work out from the rmorg site and links, the “cavitrap” construction used the cavity or chamber anode form that reduced secondary emission. One may suppose that the “cavitrap” name referred to the fact that the secondary electrons were trapped in the anode cavities. Also, the irregular anode surfaces helped to reduce Barkhausen oscillation. The use of a small positive bias on the beam confining plates to suppress Barkhausen oscillation seems to have been a later development; for example in American practice it shows up on the 6JB6 but not on the 6DQ6.

Originally Posted by kalee20

I have seen two specimens of this amplifier, too.

If I recall correctly, the PL81 and the EL81 are identical apart from the heater.

It is a curious valve line-up as I'm sure that beefy output valves were in the Mullard/Philips line-up at that time - EL34 etc. so why use multiple EL81's when single quantities of these would do?

The EL81/PL81 predated both the EL84 and EL34, although EL37 was earlier. I suppose though that the efficiency of push-pull Class B might have been attractive for such applications as PA amplifiers. And the low anode voltage would have favoured the PL81 for DC-AC equipment, although the contemporary PL82 would have fitted there as well. Or maybe it was simply the case that Philips did it simply because it could.

Cheers,

[boxdoctor]

Originally Posted by Mike Phelan

I remember them, John, but cannot remember what set they were in.

Was it not the Ferguson 454 portable? I seem to remember they had some kind of special squat version of the PL81 fitted. Tony

[FERNSEH]

A Mazda PL36 which I guess is of DDR manufacture. "Foreign Made"
The beam forming plates can clearly be seen. As soon as I can find a faulty Mullard PL36 I'll open it up to determine if it is a true pentode, that is one with a suppressor grid or it has a similar construction as the one shown in the attachments.

DFWB.

[Synchrodyne]

Yes, that looks like a classic beam tetrode. It also has the anti-Barkhausen anode surface projections discussed in the RCA literature, but not the cavity anode of the PL500.

Something that comes to mind from looking at those pictures is that the beam tetrode, with its relatively large screen-to-anode spacing, may have had a structural advantage over the pentode (apart from other advantages) in terms of its ability to handle high peak anode voltages. From the PL81 through the PL519, the stated maximum peak anode voltage was 7 kV. For the UL44 it was 3 kV, and for the PL38 it was 1.2 kV. These numbers suggests that possibly the “break” point, the transition from pentode to beam tetrode, was between the UL44 and the PL81.

It is difficult to be sure from the pictures, but it looks as if the PL36 might have had the same base arrangement, that is a glass button base slotted into an octal base, as the EL34.

Cheers,

[FERNSEH]

Originally Posted by Synchrodyne

It is difficult to be sure from the pictures, but it looks as if the PL36 might have had the same base arrangement, that is a glass button base slotted into an octal base, as the EL34.

I think believe the PL36 was from the outset assembled on the glass button base.
There are few examples of the valve about in the shop including examples made by other manufactures.

DFWB.

[Heatercathodeshort]

Time to confess.. The 'odd' PL81 is a very early [1929] Mazda tetrode AC/SG. The words MOV and HAMMERSMITH were added with a simple photo programme given away with my digi camera. History will never be the same again. End of story.

The PL81 must have been the most widely employed line output valve. The PL36 did not see use in British receivers until around 1957. It was probably Mazda that sowed the seeds with the introduction of the 200m/a heater 20P4 in 1953. This was a very good valve, one of the bright spots from Mazda.
The PL36 must have been in service the longest with the ITT VC series mono chassis, around 10 years at least.
Thorn were early users of the PL500/504 introduced in the 900 chassis of 1964. It's all glass base offered easier assembly in printed circuit receivers and it was a very good valve. Many were changed in the hope of clearing width stabilization faults due to high value resistors but I dobt if many were actually faulty.
All these valves were developed from the original tiny PL81 and it must have been quite a shock to early television repair guys when they first encountered it! The PL38 was around four times the size! John.

[FERNSEH]

Three Tungsram EL38s. The one on the left has a different electrode structure and resembles the American 6BG6G in some respects.
It was made in England but certainly does not have a Philips/Mullard look about it. It is designated as EL38 and not the usual American number 6CN6 as it and many other Tungsram valves in the 50s and 60s were.
The other two are of Mullard Blackburn origin.
The EL38 had a predecessor, the EL50 which was intoduced in 1939. The EL50was used as the line output valve in the Pye 915 TV chassis. Model 12C.
Earlier Pye TVs used the Mazda AC6/PEN. Models 815 and 838.

DFWB.

[Synchrodyne]

Regarding the PL500 and its cavitrap anode, there is a brief description at this web page: http://www.thevalvepage.com/valvetek...2/wheotip2.htm. The paragraph is towards the bottom of the page. It confirms that the primary purpose of the cavitrap anode was to limit secondary emission.

Cheers,

[Station X]

Thread reopened.

[Robert Gribnau]

Originally Posted by FERNSEH

...
Very early PL81s were true pentodes but later on the valve acquired beam plates in lieu of the suppressor grid.
...

I found this earlier thread you started in 2013 and asked forum moderation to unlock it.

As far as I know, the PL81 always was a beam tetrode. I never saw evidence that the first PL81's had suppressor grids instead of beam forming plates.

In my view, Philips/Mullard calling the PL81 a pentode has no meaning in this respect. Philips used the designation "output pentode" also for valves that were clearly beam tetrodes, like for instance the PL36, and the power section in the ECL82. Philips sometimes also used the designation "tetrode". The QQV03/10, which does have beam forming plates, is an example of this. See: https://www.vintage-radio.net/forum/...43&postcount=7.

Only later on Philips occasionally used the more correct designation, like "beam power tube" (see for instance the PL500, the successor of the PL36). But the later PL509 and PL519 were again designated as "line output pentode", while they surely are beam tetrodes.

My guess is that within the Philips group there was some kind of unwritten rule not use the term "beam power tube" or "beam tetrode" because the beam tetrode (a British invention) was patent-wise the answer to the pentode (invented by Philips).

[kalee20]

They probably decided that the competitor's technology was more suitable in some applications, paid the royalties but didn't let on to the outside world.

And I guess the holders of the beam tetrode patent forgot to insist that the valves were designated as such (maybe didn't care, but that seems unlikely!) so Philips, after quietly paying the royalties, flogged them as pentodes.

The DL92 'pentode' is definitely a BT, but interestingly the later DL96 is a true pentode.

Coming back to the PL81 / EL81, although rated less than the EL84 in terms of anode dissipation, has a much higher current capability (just compare heater powers necessary to heat the larger cathode!) so makes for a more powerful Class B audio amplifier - and of course is much better suited to line output duties.

[RogerWalker]

There is a description of the cavitrap PL500 but it is auf Deutch.

DFWB. Here's the translation, according to DeepL



Almost 3 decades earlier, on September 9, 1932, RCA tube designer Edward Herold applied for a patent for a novel output tetrode that used a specially shaped anode to strongly suppress secondary electrons even at low anode voltages, eliminating the need for a suppresor grid.

This tube has only a control grid and a screen grid above one or two low-profile cathodes, followed by the anode. It is therefore an output tetrode, which is superior to output pentodes, especially when operating at low anode voltage. This tube was introduced in December 1932 as Type 48.

Translated with DeepL

[Richardgr]

Originally Posted by Synchrodyne

And Mullard Valves, Tubes and Circuits, #8, describes an unusual DC-AC amplifier using an EF86, an ECL80, a pair of PL81 and a pair of PY82. (I guess that the PF86 had not been released back in 1954.)

Cheers,

Hi Synchrodyne, that would be an interesting circuit, please could you point me to a reference?

I have this link ...
Mullard-Circuits-for-Audio-Amplifiers
... and Chapter 8 is a DC/AC amp, but alas not a PL81.

[kalee20]

It's in here, Richard, and it is No 8, as encyclopaedic Steve Synchrodyne has correctly said!

Curious circuit, using the ECL80 triode circuit as an oscillator to develop the -30V (approx) grid bias for the PL81's without eating away at the total HT voltage as would be the case with cathode bias.

And the 'L section of the ECL80 is the phase inverter, operating at sub-milliamp levels... I'm left thinking why not use a PCF80?

[ms660]

Originally Posted by kalee20

Curious circuit, using the ECL80 triode circuit as an oscillator to develop the -30V (approx) grid bias for the PL81's without eating away at the total HT voltage as would be the case with cathode bias.

The oscillator/bias trick was used in some domestic radio receivers.

Lawrence.

[Richardgr]

Thanks for that document.

That was an interesting circuit! Apart from the bias arrangement, there is that feedback, the tone control, lots of food for thought!

[kalee20]

It is interesting! In fact cathode bias wouldn't be much good with Class B anyway - look at the variation of current with signal, 32mA quiescent, 82mA max signal. (Ideal Class B would of course be zero quiescent).

It does utilise the high current capability of the PL81.

[Ed_Dinning]

Hi Peter, thanks for the Mullard data, it was an interesting read when originally published in WW way back. I always wondered if anyone had the full set

Ed

[trobbins]

With respect to the Mullard data and circuit #8, was it common to use such a bias oscillator to generate a negative bias for output stage valves in transformerless equipment? I can see the practical expediency of using a local oscillator coil, and using grid leak biasing of both the triode and for the output stage after filtering (which at 2MHz needs only small caps).