5B/254M and the 807 versus the KT66

[Valvepower]

Hello,

Linking from the post on the Quad II and 22 - restore or risk? Thread.

Well, this had me thinking… What experience has fellow forum members had with the audio use of the STC 5B/254M and the 807 valves in comparison to the KT66? I suppose I’m thinking more Tetrode or Ultra Linear use?

The KT66 is favoured because of the ‘Kink-less’ characteristic, whereas the 5B/254M and the 807 are good ‘ole Tetrodes.

Herbert Keroes used the 807 in triode mode in the Willamson to replace the KT66. The 807 also pops up in amplifier circuits in the Brimar data books, and I’ve also seen them in some Practical Wireless articles and a variant of the Linear/RSC amplifier, but generally appear far less in audio amplifiers, except for some PA amplifiers like the Vortexion etc.

Maybe back in the day when the KT66 was readily available, the need to use valves like the 5B/254M and the 807 just didn't arise?

I’m avoiding snake oil audiophile stuff, but sensible and proper ‘old-school’ subjective differences would be welcomed.

Regards
Terry.

[Richardgr]

Here is an amplifier I intend to build one day, but I really need to know the specifications for those Partridge transformers:
Brimar 25P1

If anyone has the specs for these OPTs I would be very grateful!
- Partridge type P5352
- Savage type 5B29

The 5B/255M (a 5B/254M without a top cap) is used with HT of 440V in UL configuration, with a 390R screen grid resistor. Hence I am quite comfortable using them in my Quad IIs.

They are Loctal tubes, but I have found an easy way to make adaptors by buying cheap Bakelite octal bases, and using a wafer style Loctal socket on top, which fits perfectly. Pictures below.

[G6Tanuki]

For me, interested in getting the most audio out of the valves, the big differentiator is the presence/absence of anode top-caps, meaning you can run at the highest permissible anode voltage without worrying about potential insulation-breakdown in the valve-socket when the anode is going up to well over 1Kv on peaks.

[Of course you need to take appropriate safety-precautions].

You wouldn't want to use 750V HT voltage on a KT66!

[Valvepower]

This link may shed some light on the spec.

https://dalmura.com.au/static/Partridge%20info.pdf

Off the top of my head, this site belongs to fellow forum member Tim aka trobbins

Originally Posted by Richardgr

If anyone has the specs for these OPTs I would be very grateful!
- Partridge type P5352
- Savage type 5B29

Terry

[Radio Wrangler]

First, there was the 807 beam tetrode.

The 807 begat the 6L6, a version lacking the top cap, lacking the ceramic high voltage insulators and transplanted into the newer international octal base. Don't run it to 807 voltages.

Then the 807 begat the 5B/254M Just in a tighter-fitting bottle, and the newer base.

Meanwhile, in Europe we wanted some of this beam tetrode stuff, but the Americans had patents and registered trademark names. So Philips brought out their 'Power Pentodes' no, not at all beam tetrodes by another name, no, completely different, M'lud. Not to be left out, MO valve co produced their kinkless tetrodes. Look away! no beam forming going on here! Completely different! just don't look inside at the grid alignment etc.

So the KT66 was an invented-here slightly different 6L6 just different enough to stop lawsuits, but using closely related techniques.

The differences in these things tell you more about the legal business than about thermionic technology. Valve families got chosen on allegiance lines.

David

[Cobaltblue]

I thought the Beam tetrode was developed by MOV to overcome the Mullard Pentode patents but it was taken up and developed by RCA.

http://www.r-type.org/addtext/add029.htm

Cheers

Mike T

[Ed_Dinning]

A very useful listing by Tim of the Partridge transformers and a useful reference.
I can/ have replicated quite a few if those and even some of the C core types, but the cores are very expensive.

Ed

[Radio Wrangler]

I'm not sure who was first, the whole area was a mess and clouded by all sorts of claims and outright disinformation. The 807, 6L6 5B/254M path is clear, but how the KTs and power pentodes fitted into the jigsaw has several versions. They were all trying to come up with essentially the same thing while not getting killed by lawsuits.

Compared to the earlier technology of power triodes, higher power, higher gain and easier driving were becoming possible. Big money was at stake so people weren't playing nicely.

Interestingly, the KT's got turned into transmitter valves in the TT21 etc.

David

[ms660]

Have a read of this...Electronics, Feb. 1935, mag. page 65:

https://worldradiohistory.com/Archiv...cs-1935-02.pdf

In 1933:

https://patents.google.com/patent/GB...n?oq=gb423932a

Lawrence.

[Richardgr]

The ultimate 807 tube data - from STC with 43 pages and circuits
STC 807

[dougietamson]

Originally Posted by Valvepower

Hello,

Linking from the post on the Quad II and 22 - restore or risk? Thread.

Well, this had me thinking… What experience has fellow forum members had with the audio use of the STC 5B/254M and the 807 valves in comparison to the KT66? I suppose I’m thinking more Tetrode or Ultra Linear use?

The KT66 is favoured because of the ‘Kink-less’ characteristic, whereas the 5B/254M and the 807 are good ‘ole Tetrodes.

Herbert Keroes used the 807 in triode mode in the Willamson to replace the KT66. The 807 also pops up in amplifier circuits in the Brimar data books, and I’ve also seen them in some Practical Wireless articles and a variant of the Linear/RSC amplifier, but generally appear far less in audio amplifiers, except for some PA amplifiers like the Vortexion etc.

Maybe back in the day when the KT66 was readily available, the need to use valves like the 5B/254M and the 807 just didn't arise?

I’m avoiding snake oil audiophile stuff, but sensible and proper ‘old-school’ subjective differences would be welcomed.

Regards
Terry.

I have used a pair of CV428 valves (Mil spec equiv STC 5B/254M) in an amplifier that used a pair of 6L6GC with no issues but does require some work to change sockets to a pair of loctal bases and top caps for the anodes.
The amplifier was not ultra linear but that's for the screen grid, the suppressor grid/beam forming plates don't connect to anything other than the cathode.
My reasoning was the 5B/254M are good quality mil spec NOS made in UK and at the time half the price of the lowest cost modern foreign made 6L6GC.

The datasheet for the KT66 shows its schematic symbol with beam forming plates ie virtual suppressor grid - not a physical electrode and no royalties due.

Doug

[Radio Wrangler]

Thanks,

That was when 'Electronics' was a serious magazine, I saw its latter years where all the articles seemed to just be product placement stuff from sponsors. A sad decline.

Interesting, r-type.org describe MOV kinkless tetrodes as being a way to get around Philips' patent on pentodes, and of early MOV tetrodes being passed to the USA to stimulate production there and that resulted in the 807 and friends.

I've obviously got some of the flow reversed in my head.

The Philips patent explains why MOV called them tetrodes and didn't mention the extra electrode as being a third grid. RCA followed the tetrode name too.

Interesting to look at from a distance. THe lectures on valve operation and circuit design I was on the receiving end were all about the electron optics and how to exploit device characteristics (and how to dodge the ones you don't want) History and family trees went without a mention.

David

[Gabe001]

I can't comment about these valves per se, but I've got a single ended 6l6gc in ultralinear with 335v on anode and screen running at 0.4% thd (mostly 2nd harmonic) at 80% anode dissipation and about 15db of global negative feedback. Distortion remains below 1% until about 4w. Maximum output just over 5w RMS per channel.

[dougietamson]

Before I attacked with a soldering iron I McGyvered a 6L6 to CV428/5B/254M adaptor to do some testing.

[joebog1]

Like Dougie I have only "tested " CV428's. I am 95% complete on the Laboratory Golden Ear amplifier, with transformers specifically wound by me for that amplifier. It's a new year, and I've had a break from it for quite a while now, so I will get back to it. I have modified a few Fender guitar amps using both 807's and CV428's. Both work very well, but remember these amplifiers are a far cry from Hi-Fi.

AS far as bases go, I use Chinese white ceramic bases that fit an octal hole to perfection. I pay around $2 each for them. Top caps, likewise, white ceramic, about $1.50 each. The bases are like all Chinese ceramics and extremely high quality, the metal sockets ( rolled round sockets ) are extremely tight to be honest, so they shouldn't cause problems down the tracks but are not held into the ceramic very well. ( Just a tiny extrusion punched from the sheet metal ) After they are soldered they are fixed very tightly.

My valve tester will be using them for the output valves. I have tested them on the bench and they work superbly well. Due to the tiny bottle they run VERY hot. My negative supply is only -150 volts so I can't cut them off completely so actual real low voltage use is unknown, but so far I can get 90 volts at 50 mA without any red plating. Thats with 400 volts pushing the HT. In any case, as far as my valve tester goes, its' "gunnado " and less than 90 volts is not required very often ( unless you play with battery valves ).

My take so far, but as always I am waiting patiently for others comments and experiences.

Joe

[Synchrodyne]

Originally Posted by ms660

Have a read of this...Electronics, Feb. 1935, mag. page 65:

https://worldradiohistory.com/Archiv...cs-1935-02.pdf

In 1933:

https://patents.google.com/patent/GB...n?oq=gb423932a

Lawrence.

The full .pdf of the EMI patent is available for downloading at Espacenet, see: https://worldwide.espacenet.com/publ...C&locale=en_EP.

This was evidently a development of the Harries valve (as manufactured by HIVAC), with improvements to overcome some practical shortcomings of the latter caused by the interferences caused by electrode mounting struts, etc.

I haven’t found a patent for the Harries valve, but it was described in WW 1935 August 02 p.105ff.

Harries Valve WW 19350802.pdf

As I understand it, Marconi-EMI had a patent exchange agreement with RCA, so that would have allowed the latter free access to the beam tetrode concept.

The RCA 6L6 was described in Radio Craft 1936 July p.12ff (and probably elsewhere at about the same time), at which time it was described as a “beam power tube”. One feature was the use of aligned grids to produce essentially flat beam sheets. That did not seem to be mentioned in connection with either the Harries valve or the EMI patent, although it might have been the case. The term “beam control plates “ was used in respect of the 6L6. Later the terms “beam forming” and “beam confining” seem to have been used. The latter is probably more accurate, since the beams were largely formed by the electrode geometry, with the plates then keeping them away from places it was not good for them to go.

RCA 6L6 Radio Craft 193607.pdf

The Osram KT66 was announced in WW 1937 July 16, p.61.

Osram KT66 etc. WW 19370716.pdf


Cheers,

[Synchrodyne]

A fuller description of the RCA 6L6 development was provided in RCA Review, 1936 July, p.101ff, in an article entitled “New Developments in Audio Power Tubes”, (https://worldradiohistory.com/ARCHIV...w-1936-Jul.pdf.) There was a footnote on p.106: “Some developmental work had been done in Europe, especially by the Electric and Musical Industries Limited in England, along this line.” It looks as if “much" was translated to “some” in its westward transatlantic journey.

The RCA 807 was announced in Electronics 1937 January, p.60.

RCA 807 Electronics 193701.pdf


Cheers,

[Diabolical Artificer]

I've never used KT66's in an amp but have built several amps using 807's with good results. The thing that disqualifies all the 807 type valves though for UL use is the screen grid max voltage of 300v (this has been discussed a few times here in the past) though the datasheet says Vg2 can be 400v in triode config. I've run 807's with the screen grid at 400v in test set ups but never for any length of time.

I like the shape of the 807, it has character, I'm not keen on the ceramic valve bases though, they're a bit tricky to mount and I've bought UX5 & UX7 ceramic valve bases in the past where the spring on the base pin holders have broken, maybe a bad batch? NOS UX5 bases with a similar profile to octal sockets are a lot easier to mount and have a more condensed footprint meaning the layout can be tidier.

Here's some pics of 807's doing their thing. Andy.

[Richardgr]

Hi Andy, If you get a chance take a look at that Brimar 25SP1 amp schematic from the Valve Museum, linked above. There the 5B/255M is used in UL with 440VDC to the plate, and a slightly beefier screen grid resistor (390R).

Maybe the quality of STC tubes was higher than the mass produced 807's?

Whatever, if you don't mind a 19V heater, 5B/258Ms can be had in quantity for £8 (Billingtons). That seems pretty cheap for an uprated NOS 6L6 equivalent.

[Richardgr]

Originally Posted by Valvepower

This link may shed some light on the spec.

https://dalmura.com.au/static/Partridge%20info.pdf

Terry

Thanks for that! I did some sleuthing a couple of years ago, and I think I managed to get that far as well. I was in contact with Majestic Transformers in Poole at the time, and I had a quote to have them made. Think it was about £500 for the pair, so I put the idea on hold, but maybe that was the potted price.