Eye witness info from GEC

[Jeff West]

Thought some of y'all might dig this as much as I do. Bryan Lockey is a vibrant being, judging from his writing, who worked at GEC at Gateshead and Hammersmith 1955-61, right in the middle of all kinds of tube making adventures, starting as a teenager in an advanced electronics apprenticeship. He says he loved it. He's been gracious enough to correspond repeatedly and address lots of my pesky questions about valves and valve making there during that prime time, from an eye witness and participant perspective. I've copied some of our correspondence below; afraid it's in reverse chronological order. He specifically said he doesn't mind at all if I put it up on Marstran for your consideration and enjoyment!

JEFF




To: jawket@hotmail.com
Subject: MOV/GEC/Osram codes and more
Date: Tue, 2 Dec 2008 14:21:23 +0000

Thanks, Jeff, for your reply.


Now, where was I, part 3!

The Gateshead factory only produced "miniatures" on the B7G, B9A and their wire and clip versions, The wires were all military (CV)versions of the domestic types. I saw some in the RAF Hendon Museum in some serious ordnance. The clips were used by our "General Post Office"--at the time, (it was a national telecomms company, like ATT, only it had no competitors).

However, we did also make 3 sub-miniatures for proximity fuses for anti-aircraft shells. Tetrode amplifier, triode amplifier/oscillator, gas- filled triode fire-atron! I remember blasting them with compressed air down a steel tube at 100psi, mounted in a "steel bullet" into a lead disc, 150mm diameter; penetration had to be no less than 6mm, and the little beggars had to test within 10% of original parameters! (CV122 and CV123, and CV61). George Adamson, the production manager, (on these subminiatures, and the X78,and X79), a real nice guy, was always furiously one year behind becoming a member of the Institute of Production Engineers, as they raised the qualification levels every year.

Many M-O V types were RCA derived. Hence, they were not to be sold in North America.


All M-O V tubes, at Hammersmith, and Gateshead, were tested 100% on the production line.

Date Codes

These were etched into the glass immediately after the visual inspection test. Hydrofluoric acid and a white pigment were used. It was almost impossible to remove. Yep, several boxes of Z77s became L77s. These were rescued by sticking the GEC or Osram label over the error, and re-stamping them.

Bought in Tubes

This must have happened. Don't know of it happening while I was at Hammersmith. Certainly it happened later, as I saw Osram/GEC TV valves with that small coding Mullard used to date and factory identify. The M-O V sharp square date/type/factory marking was kosher, but there were loads of markings which looked dodgy; but were genuine. I handled a huge pile of returns at one point in my career, all were GEC/Osram manufacture, but the etching would have failed visual inspection at Gateshead, and looked like a kid with a white pen had drawn them!


Matching

This was done elsewhere when I was at Hammersmith. There was no time during production. It may have been done by Hirst at Wembley, where life testing was done, or by the Dev. Lab.

The production lines ran, on average, I guess, at 15seconds per tube. The Z77/6AM6/EF91/, for example, was running at one every 7seconds on each of three pumps, 24 hours a day at its height of production at Gateshead, as well as at Hammersmith. I ran U78, N 78, X78, X79, B309, B729, B739 and Z77 for almost a year before we were shut down. (not all at the same time!) Bigger tubes took a little longer to pump, up to 20 seconds a position.

Hammersmith ran about 10hours a day maximum around 1957-1960, for all production.

One of your links to the KT66 showed a KT66 with a KB/L label. Now, what can I say about that? If the date stamp shows it to be after 1951, KB/L was CSF France! CSF probably made KT66s for M-O V, too, as M-O V had a strong link with CSF, as well as RCA. In 1957 I was in the Drawing Office at Hammersmith, and was working on converting drawings for CSF Carcinotrons to the M-O V house standard. (CSF CV6124 BWO).

Colloidal Graphite/Aqueos carbon/ Dag80.

Oh dear! This caused many early grey hairs. After I migrated to London, I got a copy of the McGraw-Hill/MIT book on Tube manufacture. It was hindsight. In the Gateshead factory, we worked on reports from Hirst Research Centre for help if we had excessive failure rates. (above 3%). We used Dag80 at Gateshead on a number of trials on different types that needed rf screening or heat dissapation(disappation?). N78/108, HN/LN309 spring to mind, and also the wonderful Z77(6AM6). There was no doubt that for heat disappation, it was the best thing, if it didn't scrape off and then contaminate the cathode coating when the electrode assembly was pushed up into the bulb. and the Z319 secondary emission pentode used it, too.

That KT66 link also shows a bulb with the grooves/scrapes where the locating micas contacted the glass.

There's a price for everything. A quick splurge with a spray gun--you have an envelope carbon coated from here to eternity. Dead cheap to do. Then life test reports come in. So you now nask/scrape off the coating on the top end of the bulb . Better, says Hirst Labs. Then you scrape off at the bottom of the bulb. Not bad, says Hirst. Then they say, stuff the coating, try this carbon-coated nickel-iron or carbon-coated Ali.nickel stuff for the anodes. Yo! Eureka! Then, 6 months later, ---this nickel-iron stuff is s--t, needs more outgassing(time on the pumps) -- , maybe put the carbon back?

The carbon scraped off when the assembly is pushed in, is loose in the bulb, most falls out on the assembly line. During pumping some of it gets on to the cathode. The carbonates of strontium, barium do not like this. Two things can happen- the simple one is the loose blackstuff lowers the effective cathode temperature, and chemically, when combined with the rare-earth oxides a lowering of emissivity happens. Prevention? A more expensive base glue in the Aquadag to make it scrapeproof, but not degas/degrade at high temperature. So, costs versus life expectancy! Yuo pays your money---.

So, KT66s can have carbon coatings of various lengths in the bulb. Some have none, using better heat radiating anode material. Different bulb shapes were usually based on economics. It was cheaper to use one shape that was consistent with the types being made in the factory.


At the time I was at Hammersmith, the 66 was on line, as well as the 77 and 88. The A1834/6080 was also happening. sometimes in the KT66 or KT88 style bulbs, but the 6080 went back to straight tubular glass when it wouldn't fit into some Marconi TV Camera power supplies ('cos they were mounted too close together!).

The U19 rectifier used a top cap anode similar bulb to the KT88. 2 production technicians, into amateur radio, thought that the 1930's M-O V KT8C was a crap version of the 807, and that the fat-bastard KT88 with its low interelectrode capacitances would be the bees knees with a top cap anode as a TX tube, and so we got the girls on the U19 and KT88 production lines to sneak through 6 KT88s assemblies with top-cap anodes. The U19 line was the only line handling top-cap bulbs at the time. The girls lost some of their hourly productivity rate doing this. Us teenagers had to pay a high price! (4 dates and 4 cinema tickets! My, how we suffered!) Aged/burned in, (not just the tubes!) we took them to the Dev. Lab. Say no more, Pat Cundy et al took over, and that was the gestation of the TT21 and its sisters. Mike and I were both moved immediately to different sections. Me to DET24s, then magnetrons and pulse radar thyratrons/ hv rectifiers. They're not so nice to listen to. But they do light up fluorescent tubes at 10feet.

Getters

These are fired by rf induction during the very last stages of pumping. They clear up any remaining gases in the envelope vacuum. We are already down to 10 to the minus6 mmHg here, unless you're having a bad day. The shapes and sizes and numbers are a reflection of life-testing showing up production problems. A single getter may look rinky-dink compared with a fat cup, or two Ds, two rings or two squares. After its first job of getting the vacuum down to f-all of a mmHg, it stays active chemically, ad/absorbing?, whatever, soaking up any gases released when you cook the screen grid/anode 'cos you've turned the wick up too high. Don't forget, manufacturing costs could make two getters cheaper than one big one, and squares and Ds were all the same to 25kW of 454kc/s rf in the induction coils. The firm that made them is just out of reach of my memory-- ( it was a liquid/unlikely name, like GM making pizzas! (oops, maybe they will be!) Just recalled, Union Carbide.

If the getter patch on the bulb is silvery, that's very good. Some slight browning is fine, too, it's doing its job. When it's browny black, it's done its job. The bigger the better. If it is tiny, isn't there, or its white-- get another one!
Just remembered; if the tube failed on test because of "backlash" ie negative grid current(ions, rather than electrons) , they were given a 2nd. blast of rf induction across the getter to clear up the gas. Many a ring finger was burned, many a watch-wearer found the hairspring a silver ball rolling around the face! These gave a getter patch that was both silvery and brownish. But they were fine, life tests came back with no problem.

A re-cap on Dag-80-- M-O V exclusively used magnesium to coat the bulbs of Z77/6AM6s. This was fired like a getter, and provided some rf screening, and a reduction in secondary emission. It has a very open construction, the cathode "seeing" a huge amount of the glass. The cheaper Dag-80 carbon coating was tried, and was a disaster; production line failure rate was huge, and life was reduced. A blue coating (can't remember its name) was also tried, but was not successful.

12AU7, 12AT7, 12AX7

Yes , we made these at Gateshead. We made B309(12AT7), B709, B319, B329(12AU7), B339(12AX7). The B319 was a vhf/uhf tv tuner type. I see B329, B729 and B739 are given as equivalents to the 12AU7 in some references. All three would have been different, in some way. And B339, B759 as 12AX7. I think that they may be bought in from another maker, and numbered slightly differently from the M-O V types.

We also made the Z729/EF86/6267 pre-amp valve.

Octal Bases

These developed/improved over the years. Costs played an important part. With octal bases being used for AF and HF there were times when the more costly HF types were "cheaper" than the AF ones. Materials were variations of Bakelite and resin plastics. Metal shells were stronger, but costly. used on KT88 and TT21 types with their higher anode dissapation. I guess they helped screening, too, at RF. There was no discernable differences on test.

E3375

I see in a past post I thought that it was similar to an EL36/360, but looking again at the size of the anode, I feel that the suggestion it is "an EL38 equivalent" is right. I'm surprised that an E-type is available in commercial quantities, as it would have been given the prefix A when it went into production. 1974 looks like the manufacture date on the pictures I've seen.

KT88 Amplifiers

One web page shows several KT88 amps, including one using 10 in the output stage. I saw the prototype of this in the Dev Lab at Hammersmith, built on the proverbial "breadboard". It was used for the factory PA system for years, until "it sounds a bit distorted"--no wonder, only 2valves on one side were functioning, and 3 on the other!

I also obtained a copy of the original Williamson design brief and circuit, in its original blue duplicated form. (Gestetner?) Long gone now, after several house moves, and a divorce!

Cheers, Happy Holiday to you and yours.

Bryan



Subject: RE: Re MOV/GEC/Osram codes and more
Date: Mon, 1 Dec 2008 21:44:43 -0600


Hi Bryan-

Thanks again for writing me back the other day.

Yeah, the idea of the number across from the date codes (within the box printed on the glass) as indicating factory was just a hypothesis, in part because the started putting “Z” there in just that position in 1959, and that code clearly could be seen as a factory designator.

I have seen “3” and “9” (the latter for Gateshead, I take it) , maybe others, on other GEC valve types, but I don’t recall those on KT66 and not on any of the ‘40s –‘50s ones I have at hand from accumulating over the years.

In addition to Hammersmith, I know that (older bottle style) KT66s were made at Shaw during WWII per Vyse & Jessop (2000), there’s even a photo of one from wartime Shaw display (p290). I have felt confident that post-WWII KT66s were indeed manufactured at Dover facility as well for awhile, on the basis of having seen several with British mil acceptance coding, i.e., pieces marked “KB/DZ”. There’s an example here: https://www.tubeworld.com/kt66gen.jpg The “LJ” indicates 9/55 for the acceptance date on that one, so manufacture would have been at least a bit earlier.

My understanding has been that the factory code in this (mil/postal acceptance) system is supposed to reflect the site where the final pump out was completed.

The “4”, “6”, “10” coding is now perhaps more mysterious to me than ever, based on reexamining some codes since you wrote. I had compiled some lists of KT66 code markings over the years, but not very well organized. When I attempted to do this with some available pieces this week, I notice that the number codes seem to be bunched temporally, at least on these pieces. Here are some examples I can verify first hand:

DH 6 (Aug 1948)

GK 6

HA 6

HL 6

JC 6 (Mar ’53)

KC 10 (Mar ’54)

KK 10

LD 10

LF 10

LL 10

LM 10

MD 10

MF 10

MK 10 (Oct ’56)

NH 4 (Aug ’57)

NL 4

PD 4

PJ 4 (Sept ’58)

QB Z (Feb ’59)

RD Z

Now, I could have sworn that I had seen “overlap”, i.e., different number codes for the same date codes on KT66s, but I’ll be damned if I can find it right here and now with these!

Also, I feel quite certain that I’ve seen these number codes on other valve types (non-KT66) outside of the above time frames, e.g., “4” codes on late ‘40s- early ‘50s U52s, etc., so that may have fueled my assumptions too.

I will say that I have seen specimens with both "PF KB/Z" and "PF 4", FWthatsW.

The date coding is explicated several places online and off, but the following I find to be a useful reference site: http://www.tubecollector.org/cv-valves.htm

Thanks for interesting comments on the coated vs. clear KT66 versions. I had thought about asking you why they “went clear” in my original message, but didn’t want to press my luck! Is that aquadag or equivalent applied to the envelope? Over the years I’ve observed that the change (from coated to clear envelope) appeared quite abruptly with all grey through 1961 and clear starting without exception on 1/62 and all later-coded, at least this holds up over dozens of examples seen. Interestingly, have seen numerous examples dated from second half of 1961 that have a uniquely curtailed coating, all of the top “dome” is clear, everything below is grey. Very different looking, I guess these are “transition” pieces.

Some other observations about the post-WWII KT66s, with the “modern” (1950’s) bulb shape, in all cases dating by the letter date codes:

+The 8/48 one listed above is the oldest I have. I seem to recall on ebay seeing possibly as early as ’46 with this bulb, but not certain.

+Bases seem to be uniformly black in late ‘40s and after through first few months of 1953, then shades of familiar reddish/tannish brown later in 1953 and thereafter.

+The appearance of the “Z” as the third character was approx the beginning of 1959 and thereafter.

+Change from “smoked”/grey/ coated to clear bulb starting in 1/62, per above.

+Single “inverted cup” for the getter throughout the grey era, changed in shape a bit but still just one “cup” at the bottom in first three years of clear versions. Changed to double halo getters (one on each side at the bottom) at end of 1964.

+Earliest “Gold Lion” branded KT66s I have or have seen are early 1960 (several). These of course are grey coated.

+Date codes changed to 4 digits starting 1969, after all letters were run through including “Z” for 1968.

+Latest date code I’ve seen on a KT66 is 1983.

***So that brings up another question I wanted to ask: do you recall just when the letter date codes were applied? I’m assuming it was in the factory, very shortly after manufacture, so they would be very close to a “finishing” date for the particular valve rather than a later shipping or branding date (as on many American EIA date codes) (?).

***Another question: You mentioned every piece was tested, were the “Gold Lions” or “Gold Monarchs” any different from bog standard other than packaging and promotion (and price)?

***Finally, it’s a non-sequitur here but . . . did GEC actually manufacture any ECC83/12AX7 equivalents (as opposed to branding other manufacturer’s pieces)? If so, I reckon that would have been at Gateshead? I hear rumors about early B339s and B759s that were actually M-OV manufacture, not just selected Mullards, etc. . . . *As long as I’m on the topic, any insights into how GEC selected a Mullard (or Matsushita, or BEL) made ECC83 to be relabeled and designated a (much more expensive) B759? Many thanks for even entertaining these queries!

You mentioned Wembley, have you seen this: http://www.berytek.btinternet.co.uk/old/osrams/page1.htm
The archive photos and others are great!

I have a nice original M-OV publication on the KT66 with specs and application data from 1949. Would be happy to send you a good quality hard copy if that’s of any interest.

Sorry for rambling ride here . . .

Take care, have fun, best regards-

Jeff W.



To: jawket@hotmail.com
Subject: Re MOV/GEC/Osram codes
Date: Mon, 24 Nov 2008 22:27:08 +0000

Hi, Jeff,

I thought this was going to be an easy one to answer! Then I thought I'd check what was on offer on E-bay and other sellers of KT66s, KT88s and then the valves(tubes) I was involved with at the Gateshead factory.

I am sure that MOV/GEC KT66s with a date code later than 1955 could only have been made at Hammersmith. The production lines at Hammersmith were "multi-type" capable; I guess the numbers 4, 6, 10, refer to their order from one end of the factory floor, E3. (East Block, Floor 3). I never noticed the different codes when I was at Hammersmith! Probably because I was trying to keep the failure rate of some radar HV rectifier below 3%. There was a bombing contest in Canada in the late 50s for NATO bombers------. I cocked up magnificently, producing an extreme G-resistant version, (just the job for a dive bomber, yes?), but hadn't realised that while the valve(tube) specification said "no connection to be made" to a certain pin, the RAF had used it. The radar died at the critical moment. It(the shit!) came back directly to me, via those date and factory codes!

I don't believe any other MOV factory produced valves(tubes) after 1956; certainly Gateshead, (factory name E10, stamp code "9"), closed overnight in July 1956, with the loss of 850 workers, not in 1958 as the books claim. MOV's contract with EMI/Marconiphone/HMV had been cancelled because EMI Group would no longer be making TVs, Record players and Radios. Staff were kept on to do an inventory, then offered a job at Ediswan-Mazda, 15 miles away. I went to London, at 17. The factory was taken over temporarily by Ediswan-Mazda, but as they were already supplying Ferguson TV and Radio,(who had bought out EMI/Marconiphone/HMV), with identical types, it was sold off.

The Gateshead and Dover factories were set up after WWII with equipment from the Shaw wartime military "shadow" factory. Shaw produced many types, including the KT66. The Gateshead factory only made "miniature" valves on the B7G and B9A bases, for Radios, TVs, and the military. It also made "proximity" fuse sub-miniatures for 75mm Ack-Ack, too.
The Octal and 4, 5 pin production must have gone to Dover, and Hammersmith, which did everything else demanded that the other two could'nt fit in, along with its own production lines.
Incidentally, the rotary vacuum pumping stations at Gateshead were lend/lease from the RCA Schenactady(?)New York factory when it updated in 1942, via Shaw, and looked OK to go on for ever, when the factory closed. All the grid, anode and cathode making equipment was ex-Shaw, too.
There were also factories at Perivale, and Wembley. the latter the hub for research and development for MOV.

So, the letter date codes are the identifier, anything post 1956 has to be Hammersmith, in my opinion; if earlier, it could be Dover, or, if really old, Shaw. (These would have a VT number, rather than KT66)

Right, now some inside knowledge!

All MOV/GEC valves were individually tested. Not 10% tested like most other makers at that time.

The KT66s with the black carbon coating on the envelope had a longer life span. (cos it got rid of the heat better) BUT those that were clear had a better peak cathode current. (the aqueous-carbon coating vapourises onto the cathode coating and "cools" it down a few degrees). And, for all valves(tubes), the brighter/silveryer the getter patch, the less gas was present when the valve was sealed-off on the vacuum pump. And it should be a good size, otherwise it is trapping released gas, though that's not a bad thing, really, as it's doing its job.

Sorry, got carried away a bit over this reply!


Probably a good idea if I save this and send it to some site looking for old farts/nerds/berks reminiscences!


Regards, Bryan

[dai h.]

jeff thanks for sharing this you are the man

[David B]

Great stuff Jeff, time for a big cup of caffeine and park the carcass for some reading.
Didn't know the old smoked glass lasted longer, have noticed they seem to run cooler than the clears which I'm speculating is explained by Bryan's comments that clear had a better peak cathode current....btw seen an add from GEC long time ago stating how they individually tested and I think shock tested with a rubber mallet!

[Jeff West]

Received a message today that he's having computer trouble but apparently was trying to send something else. We might wanna ask some further questions if you have some.

Happy holidays, all-

Jeff

[Brian Wallace]

I wish I knew what to ask. Maybe he would post a few things here?

[Jeff West]

I sent him the address and invitation, also a few follow-up questions inc clarification about GEC-made 12AX7, AU7, AT7 equivalent versions. I'm still seeing the conventional wisdom on the boards that GEC never made these.

[Maschine]

Hey Jeff,

Informative post as always. I keep forgetting how much early valve related manufacturing/research went on here in New York, GE and RCA were all over the state. Thought it was kind of neat that he mentioned rotary vacuum pumps on lend lease from RCA, but I'm thinking he meant GE as they still have a presence in Schenectady.


cheers bro,


-Clint

[Jeff West]

Hey Clint- Happy New Year.

Also can't forget Amperex in Brooklyn, Westinghouse in Bath and Elmira, among others. And we won't even go to Jersey!

Yeah, Schenectady home of GE, but RCA would certainly make sense, M-OV and RCA had various ties starting long before WWII via EMI, David Sarnoff had worked for Marconi, etc.. Harrison, NJ plant had been GE years before, prior to assumption by RCA Radiotron way before WWII. I know RCA had post-war receiving tube manufacture in Harrison, Cincinnati, Indianapolis, Woodbridge (NJ) . . . not sure about NY State pre-war.

Interesting to consider that some specific RCA and GEC specimens that we've got might have been finished on the same machine, across time and continents.

[Jeff West]

Happy New Year all. Here's some brand new additional info from Bryan Lockey (following my e-mails).

Note that the Mullard 5-10 and Osram 912 he references were audio amplifiers, not valves! JEFF



RE: !gates‏
From: Jeff West (jawket@hotmail.com)

Sent: Tue 12/23/08 5:48 PM


Hi Bryan- Sorry about your computer troubles. Just so it's clear, I did get your fantastic e-mails of Dec 2 and 19.

As a follow up on 12AX7/ECC83, 12AU7/ECC82, 12AT7/ECC81 equivalents made at Gateshead, in particular, I extract from your message that:

B339 (12AX7), B329 (12AU7) and B309 & B739 (12AT7) by your direct knowledge were definitely made there by GEC even if some equivalent versions then or later were imports from other manufacturers.

Also that you had direct experience at the pump head making B309s and B739s (12AT7 equivalent versions) yourself!

Is that right, and does that cover it for those three equiv types?

B729 is ECC804/6GA8 equiv, right? B319 you described before, and B709 I'm not familiar with offhand.

I believe I've seen GEC B339s that were '50s longplate Mullards with GEC livery, but that's cool if B339s were also made from scratch at Gateshead. I'd like to find one of those!

Still waiting for E3375 to arrive, but I'll let you know.

I guess KT88 and then KT77 actually had their debut during your tenure. I've noticed that the oldest 88s went through progressive gettering changes as you describe.

Many thanks, as always.

I'm looking at a 1939 Osram book here that has some info on KT66 from when they were quite new! Not sure if it's design maximum ratings, but shows 400/300V plates/screens for PP AB tetrode fixed bias, 50W output into <3 Kohm reflected load.

Also have more comprehensive “Osram Valves Radio Receiving, Amplifying, & Small Retifying Types for 1938-1939” by The General Electric Co, Ltd. that discusses using a pair as tetrodes with positive grid drive for up to 55W output.

Best for '09-

Jeff

From: Jeff West (jawket@hotmail.com)

Sent: Wed 12/24/08 10:41 AM

Hi Bryan- Happy Christmas, nearly. Congratulations on your daughter's engagement!

Meant to say too that I posted some of what you sent in the Tubes section on the Marstran forum (Marstran Transformers), which is new this year, some interested and knowledgable vintage electronics/music folks on at least three or four continents:

index.php

If you ever are inclined to post something there to say hello, I know Brian Wallace and the members would be highly interested, just as I am.

Regarding the grey vs. clear KT66s, I had meant to note too that, over the years I've noticed that both audio fanatics and electric guitar wranglers seem to develop preferences for one over the other, ostensibly on the basis of sound. At the risk of overgeneralizing, the audio community often seems to put a premium on the graphite versions. Amongst guitarists who have compared both, you often hear that the grey ones are considered a little richer and mellower, etc. and the clear ones are reknowned for a touch of additional bite and edge. So now I can consider that just maybe they're responding in part to difference between anode materials!

Again, best for 2009 to come-

Jeff

jeffket thought you might like this item on eBay‏
From: eBay Member: jawket@hotmail.com (member@ebay.com)

Sent: Fri 12/26/08 10:56 PM
To: jawket@hotmail.com




An eBay member wants to show you this item


Hey Bryan- I happened to see these, July and Aug 1961, final months of the grey KT66s and with the transition coating with the clear top domes, right before they eliminated the graphite altogether. Ironically, these seem to have longer scratch lines than ever from the mica!!!

Jeff



Genalex KT66 matched pair in original box, tested good

Current price: $102.50
End time: Dec-30-08 22:40:24 PST
Add to watch list | See similar items


Check out this item





From: Jeff West (jawket@hotmail.com)

Sent: Sat 12/27/08 9:10 AM

Hi Bryan- Sorry for all the separate e-mails, I should give you a break for the new year. But . . . regarding the GEC 12AT7 equivalents made in-house, thought I'd send these photo links courtesy of Tube World and Tubemonger vendors' sites. B309 is on the left in the top one- presume that's like you dealt with at Gateshead. The A2900 we hadn't discussed, all the GEC-made ones I've seen have apparently been made at Hammersmith, and seem to have '60s date codes like one here, although maybe there were other variants. Best- JEFF


https://www.tubeworld.com/b309.jpg

http://www.tubemongerlib.com/gallery2/d ... ngland.jpg

https://www.tubeworld.com/a2900.gif

https://www.tubeworld.com/ecc83gec.gif

B series and stuff.‏
From: Bryan Lockey

Sent: Fri 1/09/09 4:18 PM
To: jawket@hotmail.com
Hi, Jeff,

Thanks for the links, brought a lump to my throat!

Over the holidays, spotted a "quad" of KT66s on ebay, date codes were all in the "A" year, 1945! and factory code KB/L. Saved the page, but it has since vanished under a "microsoft Internet Explorer has founded a series error" which tried to tell me I had 192 viruses, and started a "Scan and repair". Had to do a restore, to get the laptop up and running again. I have AVG 8 which flagged up the nasty, alledgedly quarantined it, but didn't stop it screwing up .

Anyhow, they aren't there on ebay now, were on a buy-it-now at around $600. Now, could KB/L have been the Shaw factory up until MOV closed it, prior to moving the plant to Dover and Gateshead?

The KT66s with scratches on the wide part of the bulb are either clumsy assembly, or, (I hope!) reflections of overhead fluorescent lighting. Can't believe anyone so clumsy was on assembly to shove them up so badly on the piss (UK reference to the staight-line walk test for DUI) as to scratch the fattest part of the bulb.

The link, tubeworld/B309 shows a typical B309 and a !2AU7 type; if MOV, a B329.
and, tubemonger shows "lot #--" . Tubes were boxed in 100s. If you paid your money, MOV ran off a batch--can't see any economic reason for a run less than 5,000. (hmmm, loads of money-- Govt?, yeah, they'd do that)

The A2900/ECC83/12AX7 pictures are interesting. As someone who worked for both MOV and Mullard, I can see the "Philips" difference immediately. The "seal-off" pip on Mullard/Philips/Continental tubes is moulded mechanically. The Mullard/Philips/Continental makers usually used this seal-off crimping method, and used bulbs with those 4 x 90* bars on the top of B7G and B9A envelopes.

Back to that Mullard ECC83/A2900 picture on Tubeworld. The Mullard tube would have failed our 1956 GEC-MOV visual inspection for having a cock-eyed assembly, and a distorted glass base-to-bulb seal-(stress on the structure, affecting life, due to temperature variations affecting mechanical integrity) Look at the mica distortion on the top of the assembly-. But, they would have still worked to spec ! The difference is quality.

Getters

Further to the last e-mail, obviously the visual gettering effect/patches are dependant on the amount of residual gases in the vacuum of the finished tube,and then the amount of gases released by out-gassing of the electrodes in use, principally the anode, especially when the tube is worked really hard. Theoretically, this means the more and the brighter the better.
But, a manufacturer often met its production-line quota of "good" tubes by firing multiple getters to clean-up a crap vacuum production run.

The Best

The getter patches should all be the same colour--silver-ish, as Barium discolours with gaseous contaminants. That doesn't mean an expensive purchase is a load of crap. If they are all the same size patches, you've probably got a good balanced set, vacuum wise. Barium getter patches, if not silvery, gradually fade from brown to whitish.

Getter Shapes

At Gateshead we used nickel wire U shapes with the getter bar across the end, and occasionally all-circular getters. They could be mounted horizontal or angled, some "close" fitting assemblies had the latter to prevent the getter splattering across the top mica. (noise, leakage, etc)

Valves/Tubes which ran the hottest in use, used the circular getters. Doubt if anyone now has, let alone uses, the B9A HN309 triode output pentode!!! Half a 12AX7 and a 3watt output pentode. That used an angled ring getter. Guess these types burned many fingers in the repair trade; the girls on the ageing racks had to use leather gloves to get them out of the surface-mount sockets into the 100-boxes. Think it was probably used in a record player with a crystal pick-up and a U78/6X4. Not quite Hi-Fi! But cheap for the high street dealers.

Mullard 5-10

Probably the most important kick in the ass for MOV/Osram. But someone got their fingure out, as the Willamson was now an old design, and produced the Osram 9-12. Same animal, different breeding, but, look, 9-12 , must be better than a 5-10. Oh, no it's not! Oh! yes it is (pantomime banter here at Christmastime) Then domestic sales increased, and circuits/applications/ KT88, KT77 followed. That probably was the start of the modern audio market for GEC/MOV, moving on from those stalwarts the PX15 and KT66. (Built myself a record player using B339,a N709 and a U78. And a dreadful BSR Monarch 16n2/3rds to 78rpm 12 disc autochanger!)

12AT7, 12AU7 12AX7 (B309, B329 and B339)

We made these at Gateshead in several forms; the ordinary ones, a CV version of the standard tube, a second CV version with tighter specs., and the CV4000 range which were mechanically superior and had very tight limits on characteristic spreads. There were also the flying lead versions, and the "clip" ones, where a PTFE disc was pushed up the wire leads and then the leads were welded to radial strips on the disc.

Sorry for the lag in replying, but over here there is an unpleasant flu-virus grabbing everyone by the throat and laying them low for 7days or so.

Regards, Bryan.

[Jeff West]

ME: GEC 6V6G and stuff.‏
From: Jeff West (jawket@hotmail.com)

Sent: Fri 1/16/09 6:29 AM

Bryan - I had meant to say as well that I have enjoyed using these:

http://www.kcanostubes.com/pictures/222c/NOS-GEC-6V6G--Matched-Pair.jpg

Like a mini KT66 in some regards.

Jeff

************************************************************
From: Bryan Lockey

Sent: Thu 1/22/09 12:04 PM
To: jawket@hotmail.com

Hi, Jeff,


. . . Aah-- Mullard!

I found that at M-O V that blokes who had taken up a University education were paid $150(1960 rates) a year more than me ($550), I had come from a lower level of education, but had reached the same IEEE qualification levels. I had 5 years experience in the factory, they had 4 years theory. I could see I was never going to be on their salary level. I felt that my career was being based on production, rather than technical expertise.

Fancying the world of a Technical Author, I applied for, and was accepted by Mullard's Technical Service Department as a writer for their remarkable series of "handbooks". God, wish I had kept a copy of those tomes! Needless to say, M-O V fired me on the spot when I tried to hand in my leaving notice!

I had a great experience in the office, but found that the work load was increasing but not the salary! The job also involved trawling Electronic magazines for competitors products, and one day (in the Sits. Vac. pages!)I saw a requirement for electronic engineers to work in the up and coming ITV television studios. At 50% higher salary than I was getting at Mullards. No brainer!

There I got embroiled in image orthicons, plumbicons, flying spot scanners , sound desks, Ampex VR1000, 1200 and 2000 VT machines, and a wide range of tubes to keep that lot running.

I am fascinated by your interest in M-O V tubes. Tell me more!

The GEC 6V6Gs have the War Department arrowhead, the 10E stores number, and the ZA cross reference stores number. The CV number from the cross reference is CV509.
(Actually, that arrowhead is the symbol used by the UK's Ordnance Survey mapping service---duh!! of course, Ordnance is a military term!)

KB/L again

KB was introduced by the British Government to designate approved valve/tube factories when the Common Valve system was introduced in 1942, I think, but they don't seem to have got round to adding dates to the markings until 1945. "M-O V" was GEC at the Hirst Research Centre at Wembley, M-O V at the factory at Hammersmith, London, and the war-time shadow factory at Shaw, Lancashire. On this webpage https://www.tubeworld.com/kt66.htm, the 5th picture down shows various KT66s. One is stamped CV321, date code AK, and factory KB/L, another is CV1075, date code LJ, and factory KB/ZD. Shaw factory closed just after WW2, and the miniatures went to Gateshead whilst the big bases went to Dover. This may be confirmation of Shaw being KB/L.

Thanks for the ebay reference for those KT66s.

regards, Bryan

[David B]

GEC made 6V6Gs ! wonder how those would sound in a BFDR ?

[Jeff West]

I've tried them in that, and just about every 6V6 amp I've got. May not fit (physically) in a few applications! They tend to get beefy and crunchy as you turn up, can definitely work to Marshalize other amps somewhat. Rugged and high qual, as you might imagine.

[David B]

Rugged is good, those Deluxe Reverbs can run some high plate voltage. Haven't seen any examples yet on eBay..

[Jeff West]

From: Bryan Lockey
Sent: Fri 2/06/09 12:55 PM
To: jawket@hotmail.com

Thanks, Jeff, for those links. (Note: This refers to the 1930 GEC/Osram films, i.e.,

http://www.youtube.com/watch?v=pnCYefwyd8U

http://www.youtube.com/watch?v=GbsObrKT3ms&feature=related

http://www.youtube.com/watch?v=VLSZkUp2gTs&feature=related )


I wish I knew of a similar video/film shot at a later date. There must have been a promotion/staff induction film. I didn't see one in the 50's; but was subjected to one at Mullards in 1961. Can't see why GEC/MOV were'nt on the publicity/staff promotion bandwagon, too. Aaah, yes, we (MO-V) seemed to be 2 years behind the competition in going for it! Tubes and publicity!

Good lord, that Philips video brought back memories of Hammersmith. In the opening clip, it looked little different from 1956! Just to the left of the opening shot at the beginning, one can see the edge of the "Citroen" building. This was taken over by GEC from the French Citroen Car Company at an early date, don't know when; it was our mechanical workshop, 2 floors, 100yds x 50yds handling everything from machine repairs to lamp and tube pumping and production machines, Moly and Nickel wire stretching/drawing machins or turning out 3 foot high copper blocks/anodes for Royal Navy RADAR 600mhz magnetrons. Your cell phone is on 4x this frequency, and next to your ear! A fork-lift truck got it up to W3! The machinery scared the life out of me when I was in there as part of the apprenticeship, when shown the problem of factory fluorescent lighting and rotating machinery driven by mains/line power. Yep, the lathe chuck looked absolutely stationary, but was rotating at 1400 rpm! That building still bears the Citroen "coat of arms" of 2 inverted Vs to this day. And the main building is now a Supermarket. Like Piggly-Wggly.

The other Philips videos were another lot of tear-jerkers!

[BryanL]

Hi, guys,

I'm Bryan, Jeff's "GEC/M-O V eyewitness". I guess you have all read the e-mails Jeff and I have had over the last 4 months or so. It has been quite therapeutic to recall my employment at MOV 54 years ago!

I can categorically state that M-O V made the 12AT7/AU7/AX7 types at the Gateshead factory up until 1956. Don't forget, the B309/12AT7 was originally designed as an RF amplifier that came good as an audio amplifier. Hence the structural variations between versions. (The 12AU7 also was originally designed as a push-pull rf output tube at 100Mhz!) I find on http://www.r-type.org there are photos of the Marconiphone B309 which are of Mullard manufacture-- see the factory/date code bottom right. Marconiphone was a UK Domestic trade name , used by TV/Radio set makers such as EMI until 1956, after that used by Ferguson/Thorn/Ediswan-Mazda, who bought up EMI's domestic trade name. The MOV one is the only original MOV one in the group. The Marconiphone one shows the typical "Phillips" pip moulding, and the bulb's 90degrees markers. The Brimar ones show a weak/collapsed seal on the pip. (An MOV quality check failure). The Clairtone is obviously a Mullard job.

The KT66 has had a varied life in M-O V's production. The later fat, stepped envelopes have a greater surface area to radiate heat. Getters-- the more, the better, and the brighter the better. Carbon bulb coating-- like I said to Jeff, better heat radiation, but a possibility of lower peak cathode current.

KT88--like the 66, the more, the brighter, the better.

I hope I have provided you with an insight to MOV and its products. I will try my best to respond to any comments you have on MOV products up until 1961.


Regards, Bryan

PS I said in a previous e-mail to Jeff, that I had seen the original Williamson Amplifier circuit blue-prints in the MOV Dev. Lab at Hammersmith. Reading up on the Williamson on the internet, I find I was wrong. The amplifier I saw described a KT66 push-pull job which used a KTZ63 as first stage, followed by a B63 phase splitter. The sheets were dated 1946.