Ideas For EL34 Pentode Push-Pull Amplifier

[Cressy Snr]

This is a good read about the history of the EF50.

http://www.dos4ever.com/EF50/EF50.html

Very interesting.

[Cressy Snr]

Ive got the EF50s warming their heaters for half an hour.

They're a complete pain in the arse to be honest.
The mere act of inserting and removing them from the 9 pin sockets caused the glass around some of the pins to show signs of cracking, on three out of the four valves.
It's not as if the sockets were even a particularly tight fit and the valves were able to be inserted and removed without rocking them in any way.

Because of the cracking around the pins, it was then necessary to cut the cans off all of them to inspect the getters. These appear to be intact, for now.
I cut the ally cans at the right place, to allow the locating key and base ring to remain fixed to the valves, and luckily they are nice looking without the cans.

But now I'm in two minds whether to use them at all, which is a pity as the curves look so promising. I don't want them going gassy after a couple of weeks, if the pin seals are showing signs of cracks.

I have four 12GN7 and four EF184 and am wondering whether these would be a better bet, but which ones out of these two types, will perform best, given the plate to grid feedback scheme?

I'm thinking the EF184 will be the better bet as its gm is less than the 12GN7, which should make getting decent match a damn site easier.

The cracking incidents on the EF50s have taken the shine off this project for the moment; they seem OK, but there'll be always that doubt. Six pins out of nine are showing cracking around the seals on one of the valves, after the heater warming session. Not good.

A bit disappointing, but maybe the EF184s will do the driver job as they are direct descendants of EF50.

[IslandPink]

That's very disappointing Steve.
I didn't realise there was a 9-pin loctal till you showed those, actually .
All of the 8-pin loctal valves I've used have been fine , which doesn't help you one bit !

[Paul Barker]

That's very disappointing Steve.
I didn't realise there was a 9-pin loctal till you showed those, actually .
All of the 8-pin loctal valves I've used have been fine , which doesn't help you one bit !

My 4p1l's are a pain in the but, I ended up cutting off the aluminium and just using them as glass. Yours must be different manufacture to mine. Get stuck and won't come out, would rather tear the socket from the chassis than come out of it.

~Isn't the whole point about loctals that you can use them in airoplanes?

[IslandPink]

May be socket-dependant then ? - I use those teflon sockets with the silver-plated contacts for Loctals. I think I got them from Crowthorne a few years ago. They are OK, grip enough but not too much - their only problem is that sometimes the metal sleeve comes part-way out of the middle of the socket when you take a valve out, and you have to push it back down .

[Cressy Snr]

~Isn't the whole point about loctals that you can use them in airoplanes?

Yes.
early on, in the development of the straight pin version, they found that the EF50s would vibrate right out of their sockets.
To cure this, developed a nine pin socket with a threaded collar. You put the valve loosely in the socket, then a screw-down ring fitted over the can, engaged with the flange at the bottom and pushed the valve down rigidly in the socket, once the collar was screwed down.

The early valves had hooked pins, which were a twist to lock arrangement, but the act of twisting the valve to lock it cracked the glass around the pins and the valve had a short life. The straight pin version was supposed to stop that problem, but on my 1940s vintage ones, the problem seems to still be there even with the straight pins.

The only way I can see to avoid this pin cracking problem, is to find some old screw flange sockets and collars. I can't be bothered.

[steve s]

I may have some locking sockets, but finding an exact pair may be a challenge i also have a bucket load of ef50's, i had a sort through the other night checking for some high gm valves of similar age and type, found none but there must 20 ef50's, they look the part with the ali cut off.

[steve s]

Ef 37a would be a good choice too, they say the quad 1sounded much better than the quad 2... Ef37a vs ef86....

[Cressy Snr]

Hi Steve,
I would like to try some more EF50s if possible.
the frustrating thing is that they look, on paper, absolutely ideal for use in an amp with partial feedback Maybe I have a bad batch. I can't believe in all honesty that the bases could be so weak.

[Deco-Doctor]

Steve

I found four VR91's in my valve stock, I think these are the RAF equivalent to EF50.

I just ran them up on the valve tester at EF50 settings and three of them test good, the other has a very small cathode to heater leak (~10M ohms), but is otherwise good.

My AVO CT160 has a ceramic nine pin loctal socket which requires some force to get the valve in, which has resulted in exactly what you had, a tiny chip in the glass around one of the pins on one of the valves.
I found you don't get that problem at all with the old paxolin wafer sockets as the small holes in the top wafer guide the valve pins exactly in line with the spring contacts below, not much force at all is required to get the valve in the socket.

If you need the VR91's and/or the paxolin sockets I will put them in the post on Monday for you.

[Deco-Doctor]

Steve

I also have a large stock of VR65

http://www.r-type.org/exhib/aaa0090.htm

Its an octal based HF pentode similar to the EF50 with a slightly higher GM also used for airborne radar.

I can send you some of those as well if you want try them.

[Cressy Snr]

Thanks Martin,

That would be brilliant.
Could you send four VR65s rather than the VR91s
I think, in the long term, these would be more reliable.

Not just me suffering glass cracking around pins.

[Deco-Doctor]

Could you send four VR65s rather than the VR91s

Certainly can, I will get some on the tester later today and select four good ones.

[Cressy Snr]

Cheers Martin, your generosity of time and help is much appreciated.

I had a look at the SP61 on the National Valve Museum, which as you know, is the domestic designator for the VR65.

Below is part of the blurb from the NVM.
The bit in (my) italics is interesting and perhaps explains why these EF50s are such pains.

The demand for efficient video pentodes for TV came a year or two earlier in Britain than elsewhere and the SP41 was Mazda's solution. Shorter leads to shorter pins, and compactness in general, are essentials where high gain-bandwidth product is a requirement and it is easy to see why Mazda's SP41 was streets ahead of (say) Mullard's SP4.
Of course, the later EF50 overleapt the SP41 but the latter and its 6.3V lookalike filled vital needs in military radar and high-frequency communications during the late 1930s and early 1940s, until UK (and later US) production of EF50s was sufficient to meet demands. Moreover, one reason for the huge demand for EF50s was that they had short lives and needed frequent replacement. By contrast, the SP41/61 were virtually indestructible.

Short lives?
Glass cracks around pins?
Nuff said....

[Paul Barker]

well researched.