I'm pretty sure they were not.
I have an old journal article somewhere in my collection of magazines and journals that indicates that frame grid construction was an American invention. Originally anounced by Bell Labs/Western Electric about 1949.
Philips managed to make some very high gm wideband tubes back in the late 1930's (eg EF50) using very fine tunsten wire for the grid, and had some post-war improvements. But they were not frame grid construction, and while they performed like a frame grid, they were more microphonic and were more prone to failure.
Philips usually were up with the latest American developments though, and sometimes originated USA developments - via & with their American subsidury Amperex.
Last edited:
Jac v/d Walle measured a small number of ECC83 variations:
Unbenanntes Dokument
The Tesla E83CC is the best, with the TFK ECC83 not even coming close. Sadly there are no measurements of the TFK ECC803S or the Valvo E83CC to compare. Jac sees these three as the best ECC83 versions.
Philips didn't invent the frame grid. In Europe Siemens was the first with the introduction of the C3g in 1952. Not sure about the US companies.
Have done a bit of looking up.
Theres' frame grids and there's frame grids.... Oil's ain't oils....
We need to clearly define what is a frame grid. Here's my definition - A frame grid is a structure constisting of rigid vertical members who's width and separation determines the area enclosed by the grid, and these vertical members are held apart by lateral bars at top and bottom. Electron control grid wires are held in tension by the rigid frame, and so these grid wires run straight across.
By contrast, in a standard grid tube, the grid side rods are held in position by the micas, and the area enclosed by the grid is determined by the grid wire, which is not in tension, being formed into loops and relying on grid wire stiffness.
In short, a normal grid's enclosed area is set by the grid wire; in a frame grid the area enclosed is set entirely by the frame.
Modern frame grid tubes began with TV tubes like the 6ES8 and these went into mass production in the late 1950's. They were designed by Amperex and RCA.
I found that a patent 1465381, which describes a frame grid clearly conforming to my definition above, was filed by a Russel F Trimble in 1918, and was assigned to Western Electric Co.
So, like almost all vacuum tube innovations, the frame grid is American.
The later mass production of frame grid tubes was driven by the television market, particularly the advent of UHF channels and colour TV in the USA, both of which would have made the number of tubes excessive and the fault rate too high, without the use of frame grids and other tube innovations.
The mass production at low cost was made possibly not by any new basic structural principle but by innovations in assembling and welding the frame grid assembly, eg US patent 3161799 assigned to Sylvania, which is about using vertical frame members of rectangular cross section with round ends to facilate assembly in micas made on standard punching machines making roud holes.
Theres' frame grids and there's frame grids.... Oil's ain't oils....
We need to clearly define what is a frame grid. Here's my definition - A frame grid is a structure constisting of rigid vertical members who's width and separation determines the area enclosed by the grid, and these vertical members are held apart by lateral bars at top and bottom. Electron control grid wires are held in tension by the rigid frame, and so these grid wires run straight across.
By contrast, in a standard grid tube, the grid side rods are held in position by the micas, and the area enclosed by the grid is determined by the grid wire, which is not in tension, being formed into loops and relying on grid wire stiffness.
In short, a normal grid's enclosed area is set by the grid wire; in a frame grid the area enclosed is set entirely by the frame.
Modern frame grid tubes began with TV tubes like the 6ES8 and these went into mass production in the late 1950's. They were designed by Amperex and RCA.
I found that a patent 1465381, which describes a frame grid clearly conforming to my definition above, was filed by a Russel F Trimble in 1918, and was assigned to Western Electric Co.
So, like almost all vacuum tube innovations, the frame grid is American.
The later mass production of frame grid tubes was driven by the television market, particularly the advent of UHF channels and colour TV in the USA, both of which would have made the number of tubes excessive and the fault rate too high, without the use of frame grids and other tube innovations.
The mass production at low cost was made possibly not by any new basic structural principle but by innovations in assembling and welding the frame grid assembly, eg US patent 3161799 assigned to Sylvania, which is about using vertical frame members of rectangular cross section with round ends to facilate assembly in micas made on standard punching machines making roud holes.
Last edited:
Wasn't Amperex merely a brand name used by Philips in the US? Almost all Philips R&D took place in the Netherlands.Keit said:
The Radio Museum mentions 6ES8 as appearing in 1973, and ECC189 in 1963. Similarly, 6DJ8 in 1973, and PCC88 in 1957 (it doesn't give a date for ECC88). This suggests that they were European designs which were later registered in the US.
A US patent granted to two UK employees(?) of EMI Ltd refers to a method of making a frame grid, and refers to earlier patents by RCA and Bell Labs. Later patents from several US and European firms refer to it. My reading of all this is that the frame grid was invented in the US, but then developed and first widely used by domestic TV valves in Europe. In the latter days of valve TV sets we had the EF183 and EF184 frame grid pentodes but nothing similar was developed in the US - their high gain TV IF valves used a conventional grid?
Hi,
Thanks for the extra info, Keit. I'd been reading the same stuff and it's still not really clear who invented what and when.
Cheers, 😉
Thanks for the extra info, Keit. I'd been reading the same stuff and it's still not really clear who invented what and when.
Cheers, 😉
Hi,
That's also why I mentioned Philips as the frame grid type construction used in the E83CC is the exact same as the ones first used in the ECC88 (I assume the birth date for the ECC88 and PCC88 is the same).
Going by memory of what I read in telephone book sized data books from Philips it certainly sounds as if they'd developed it.
Cheers. 😉
That's also why I mentioned Philips as the frame grid type construction used in the E83CC is the exact same as the ones first used in the ECC88 (I assume the birth date for the ECC88 and PCC88 is the same).
Going by memory of what I read in telephone book sized data books from Philips it certainly sounds as if they'd developed it.
Cheers. 😉
I started to lose myself in browsing through old patents, before realising that there is so much more than just 'the invention of the frame grid'. You have the basic concept, the 'idea': take a frame and put grid wires across, like a weavers' loom you might say. Is that the invention of the frame grid? Nothing working yet. Next you need to figure out what to use it for and design a prototype. Building a prototype leads to challenges that need to be overcome. When you have prototype, tweak it until it has the desired specs. How to put it in (mass) production? Is there a market, is it economically viable? Lots of patents and companies involved.
As Keit said, Philips managed to produce high gm types with conventional grids, so why switch to a more expensive technique when there is no need for it (yet)? When the market asked for higher gm values and lower miller C's the frame grid could deliver its potential: the rigid frame didn't need the structural support from the gridwire itself as the traditional grids do, so much finer wire could be used. The wire can also be strung on a high tension, allowing more precise positioning, closer to the cathode. Added advantage of both thinner wire and higher tension is that microphonic resonance frequencies move up considerably.
So who invented the frame grid? Many companies did 😉 Compare it to the beam tetrode history. RCA was the first company to release a beam tetrode, the 6L6 in 1936. EMI developed the first beam tetrode (patent from 1933) but couldn't (wouldn't?) put it in mass production. Instead the design was shared with RCA. Who was the inventor? Both have put a lot of work in it.
But still I am curious to what was the first US frame grid valve put in production.
As Keit said, Philips managed to produce high gm types with conventional grids, so why switch to a more expensive technique when there is no need for it (yet)? When the market asked for higher gm values and lower miller C's the frame grid could deliver its potential: the rigid frame didn't need the structural support from the gridwire itself as the traditional grids do, so much finer wire could be used. The wire can also be strung on a high tension, allowing more precise positioning, closer to the cathode. Added advantage of both thinner wire and higher tension is that microphonic resonance frequencies move up considerably.
So who invented the frame grid? Many companies did 😉 Compare it to the beam tetrode history. RCA was the first company to release a beam tetrode, the 6L6 in 1936. EMI developed the first beam tetrode (patent from 1933) but couldn't (wouldn't?) put it in mass production. Instead the design was shared with RCA. Who was the inventor? Both have put a lot of work in it.
But still I am curious to what was the first US frame grid valve put in production.
Hmm, why didn't Philips use the frame grid technology for their own SQ ECC83's? Or perhaps I should ask why did TFK and Tesla? Surely a low gm type like the ECC83 doesn't need a frame grid, and there are cheaper ways to reduce microphony. The Valvo E83CC conformed to the same very strict specs of the TFK ECC803S, but with a regular grid.
Hi,
Well, they did. The E283CC is one example. Not sure it was ever used on the E83cc though.
Cheers, 😉
Well, they did. The E283CC is one example. Not sure it was ever used on the E83cc though.
Cheers, 😉
Tesla flipped the names around in that:
a Tesla E83CC is modeled after a TFK ECC803s
(Short plates and frame grid)
&
a Tesla ECC803s is modeled after a TFK E83CC
(long plates and not frame grid)
I know that's common knowledge, but it seems like it needs mentioning as it can be very confusing.
a Tesla E83CC is modeled after a TFK ECC803s
(Short plates and frame grid)
&
a Tesla ECC803s is modeled after a TFK E83CC
(long plates and not frame grid)
I know that's common knowledge, but it seems like it needs mentioning as it can be very confusing.
Hi,
AFAIK there's no such thing as a TFK E83CC.
Their naming convention is just different from the Philips SQ one in that they insert a 0 and add the S (which does not mean Spanngitter, frame grid btw but Sicherheit, Security)
As an example, there were ECC801S and 802S versions, none of which had a frame grid. There's a lot of BS on the interweb.
Cheers, 😉
AFAIK there's no such thing as a TFK E83CC.
Their naming convention is just different from the Philips SQ one in that they insert a 0 and add the S (which does not mean Spanngitter, frame grid btw but Sicherheit, Security)
As an example, there were ECC801S and 802S versions, none of which had a frame grid. There's a lot of BS on the interweb.
Cheers, 😉
Amperex was a US tube manufacturer that Philips purchased in order to do two things:-
# Make money selling tubes in the USA;
# Get access to the rather incestuous tube patent pool of the US tube makers.
How tubes were made is a fascinating story from a business point of view as well as a technical innovation point of view. There were several US tube manufactuers. Each made tubes for each other. Each did not make a full range. On each manufcturer's production line, the labelling machines would be sequentially set up to apply the other manufactuer's brands.
Each US manufacturer originated new tube types, roughly in proportion to overall market share, and the new tube types were sometimes made by other and sometimes not, but they all sold them.
The Australian tube manufacturer AWV, for a considerable time the largest manufactuer in the southern hemisphere, was part of this US-based co-operative arangement. I have only been able to identify about 7 types that were originated in AWV design labs though. AWV was like Amperex in that they had a similar co-operative arrangement with Philips/Mullard, though there was no company share ownership relationship.
So, a tube originated by (say) RCA, could be made by AWV or Amperex and sold by Mullard. A tube made by Philips could be sold by RCA, though this would have happened often.
I always have a laugh when I see someone has wriiten that a Brand-X tube of the same type gave a different sound than a Brand-Y. Not only does the sound mostly depend on the circuit, not the tube, both probably came out of the same factory anyway.
Well, those 6ES8's I changed in TV sets I repaired when I started in electronics as a TV repairman must have been transported from the future then! Australian TV production changed over to all-solid state beginning in 1966 and all brands dropped tube models by 1967.
The 6DJ8 was not used in Australia, but RadioMuseum's page you linked to has a set of links at the bottom for sets using it. The first use of the 6DJ8 listed by Radiomuseum was a Tektronix CRO in 1957.
You need to do more patent seaching. The patent you linked was filed 1953. RCA filed a clearly modern-style frame grid patent (2549551) in 1948). The more you search US, British, and Euro patents, the more you'll realise that frame grids are like most tube innovations - driven by Americans.
Commercial use of frame grid types was more or less coincidental and made possible by the cooperative arrangments between manufacturers.
Last edited:
As I said, it depends one what you define as a frame grid. Just a grid on a frame, or fine grid wires held in tension by a strong frame that does not depend on the micas for side-rod separation?
If the latter, you have the 1928 Western Electric patent, but the fisrt tube in quantity production as I recall was a Westen Electric type announced in 1949.
As someone esle alluded to, there wasn;t a commercial need until then.
Amperex was a US company purchased lock, stock, and barrel by Philips. As far as tube R&D is concerned, a greater proportion was done in Holland. A significant amount was done in the UK. And a bit done by Amperex. And small bits at other places eg Valvo in Germany.
Philips were an interesting company in many ways - not the least because of their approach to R&D. Until some business mistakes made in the 1980's and 90's, Philips approach was to have a significant presence in as many countries as possible, even if certain countries didn't ake any money. For example Philips was a major manufacturer in Australia, but never made any money there.
They made their subsidury in each country a "centre of excellence" for something. For example, Australia was the CofE for 2-way radio, with backup at Cambridge in the UK. Australian 2-way radio designs were made in Australia (for local and export), Cambridge, and Taiwan. Blackburn and another location in the UK was the CofE for television receivers, with backup in Holland - devices like line output coils, tubes, CRT's, transistors for TV were designed and engineered in the UK, and made at various plants around the world. Holland was the CofE for high quality audio, although all Philips Australian audio products were local designs.
I once worked for Philips. They were unique in many ways. We were all very proud to be working for such a significant company. They were focussed on excellence and they were very interested in having happy workers. It's very sad that their factories have closed and they are no longer of such significance. And it wasn't so much the Japanese that killed them. It was stupid decisions in Holland like trying to adapt American 5ESS telecomns to theg World market.
Last edited:
Sorry about the typo. What I was trying to say is the Tesla E83CC is made of the same construction as TFK ECC803s (short plate & frame grid). Which is counter-intuitive, because the Tesla ECC803s is not modeled after the TFK ECC803s but is instead modeled after the TFK ECC83 (long plate & not frame grid).
So, if I've got this right, as far as construction goes:
Frame Grid & Short Plate:
TFK ECC803S = Tesla E83CC
Not Frame Grid & Long Plate:
TFK ECC83 = Tesla ECC803S
Here's the source giving me this information: ECC803S TESLA FAKE TUBES If I've misunderstood something on this page, please let me know.
I'm not saying the Telefunken and Tesla versions sound exactly same (honestly, I wouldn't know first-hand, although I'd love to 🙂 ), but their construction is nearly identical, even if their naming conventions are reversed. If someone is looking for the Tesla tube most similar to a TFK ECC803S, he/she wants a Tesla E83CC, not a Tesla ECC803S.
So, if I've got this right, as far as construction goes:
Frame Grid & Short Plate:
TFK ECC803S = Tesla E83CC
Not Frame Grid & Long Plate:
TFK ECC83 = Tesla ECC803S
Here's the source giving me this information: ECC803S TESLA FAKE TUBES If I've misunderstood something on this page, please let me know.
I'm not saying the Telefunken and Tesla versions sound exactly same (honestly, I wouldn't know first-hand, although I'd love to 🙂 ), but their construction is nearly identical, even if their naming conventions are reversed. If someone is looking for the Tesla tube most similar to a TFK ECC803S, he/she wants a Tesla E83CC, not a Tesla ECC803S.
I'm sure you will correct me if I am wrong, but my impression is that Australian TV etc. used mainly European type valves but bearing US style names. I don't think they used the 6ES8 much in America, as they had their own designs of remote-cutoff cascode twin triodes.Keit said:
The late date given by the Radio Museum for 6ES8 doesn't mean it wasn't available until then, but it could mean it wasn't mentioned in US valve data books until then.
It could be that the 6DJ8 was developed in the US, then copied in Europe as the ECC88. A remote-cutoff version (ECC189) was then made here, and then registered as 6ES8.
Two things to note: a US name does not necessarily mean a US design (e.g. 6BW6), and one side of the Atlantic often developed and improved valves coming from the other side (e.g. 12AX7 became ECC83 when it acquired a spiral heater in Europe).
The manufacture of TV sets in Australia is an interesting story in itself.
There was essentially one wholely owned Australian electronics manufacturer - AWA, whose subsidiary AWV made vacuum tubes, CRT's, and transistors starting from raw materials, and MSP, who made a full range of passive components (resistors, capacitors, transfomers, speakers, everything) from raw materials. There was also a loudspeaker manufactuer, Rola, not in an ownership sense related to Rola/Celstion in teh UK, and Rola USA.
AWA and AWV had close technical ties to RCA. Essentially, AWV's tube designs, tooling, and know-how was partly their own but mostly RCA. AWA/AWV had the largest market share.
When TV came to Australia in 1957, AWA developed its own TVs around RCA tube designs, but see below. All other manufactures in Australia had overseas parents - the most important were as follows:-
Philips - Philips Holland
Kriesler - Philips
Astor - Philips
Pye - Pye UK which was owned by Philips
Admiral - Admiral USA
GE - GE USA
HMV - EMI UK
When TV started, each campany with an overseas parent got more or less told by their parent "Here is our current model - make that".
However, Australian had decided on the then latest variant of 625 line negative modulation, FM sound, same as Europe. USA had a similar standard but 525 line. UK had a completely different standard, 405 line positive modulation, AM sound.
Also, Australia had much tighter electrical safety standards than the UK, and somewhat different standards to the USA.
Australia has a hot climate. This means that for the same as built quality, reliability will not be as good.
The UK market was dominated by small screens - 17 inch is about a large as was wanted in the small British loungrooms and in any case the very poor detail in 405 line TV makes a large screen a bit pointless. The Australian market was dominsted by a deamnd for large screeens 21, 23, 25 inch. Our picture was considerably sharper and our rooms considerably bigger.
What this all meant was:-
AWA came up with their own product based on US tubes
The companies with US parents adapted their parent's US models as that was entirely practical.
The companies that traced their ownership back to Philips Holland adapted Philips euro designs.
HMV - their local engineers realised immediately that modifying a UK model just would not do. It was cheaper to start from scratch and design/engineer a local design.
As the tube making side was dominated by AWV, supported by sales to companies using US designs, nearly all tubes made here were US types. So HMV sets used the same tubes, and the Philips sets used the same tubes.
There were some quirks. When US makers moved to later tube types, AWV did not want to spend money retooling. So where they could get away with it, they made minor alterations to exsiting tube types and uprated them. So later US types we never saw.
Australian TV's that did not use the US designed 6DQ6 used the 6CM5 as the line output tube. I have seen on the internet that the 6CM5 was a British design. Perhaps - I have no evidence either way.
The short story is that Australian TV set used tube types almost entirely American in origin.
The 6CM5 is a 6.3V heater version of the PL36, which was used in many European monochrome TV sets. I believe the PL36 was a Philips design, although its US name of 25E5 may indicate the contrary. Near equivalents were produced by other European manufacturers. As screen sizes and deflection angles grew it was superseded by the PL500/504, then PL508 for colour sets.
A rule-of-thumb which I use is to assume that a European valve with an 'early' US name was probably a US design, but one with a 'late' US name was probably a European design. On this basis, for example, I assume that the 6U8/ECF82 was a US design while 6BL8/ECF80 was European. Finding the truth is difficult because in their advertising most makers glossed over questions of development priority; they would announce a new valve and 'forget' to say that it was a copy from someone else (in some cases just 'badge engineering').
A rule-of-thumb which I use is to assume that a European valve with an 'early' US name was probably a US design, but one with a 'late' US name was probably a European design. On this basis, for example, I assume that the 6U8/ECF82 was a US design while 6BL8/ECF80 was European. Finding the truth is difficult because in their advertising most makers glossed over questions of development priority; they would announce a new valve and 'forget' to say that it was a copy from someone else (in some cases just 'badge engineering').
Other factors complicate whether or not a given tube type was of US origin, British origin, or whatever:-
1. The type number (6AU6 or whatever) identifies the specification. It does not, in general, identify the internal design and construction.
A great many tube types are of RCA origin. If there is sufficient demand, in teh US or elsewhere, another manufacturer might decide to make it. They have a choice: a) obtain from RCA the manufacturing details (dimensions of internals parts, specifications of alloys, etc), and buy the same tooling; b) obtain the RCA manufacturing details, and modify it to suit the tooling they already have. AWV did that a lot. c) work out a completely new design that conforms to the data in the registered data sheet.
So long as the tube does work as set out in the data sheet, it got the same type number regardless of whether it was the same internally or not.
So, a British tubemaker may come up with a new tube type, and register it to get an American type number. If their is a deamnd, and American firm might have decided to make tubes of the same type number - with their own, different internal design. It amy be better than the original British tube or it may not. Is this variant then a British type or an American type? You could argue either way.
2.
All American tube manufactuers, including Amperex, were in a co-operative arrangment where they engineered tube for each other, and made tubes for each other. Philips had access to American know-how though owning Amperex - one of the reasons why they bought Amperex. So if Philips take an RCA design tube they had access to via Amperex, and put into production in the UK with Mullard adapting it to their tooling, its it an RCA type, a Philips type, or a Mullard type?
1. The type number (6AU6 or whatever) identifies the specification. It does not, in general, identify the internal design and construction.
A great many tube types are of RCA origin. If there is sufficient demand, in teh US or elsewhere, another manufacturer might decide to make it. They have a choice: a) obtain from RCA the manufacturing details (dimensions of internals parts, specifications of alloys, etc), and buy the same tooling; b) obtain the RCA manufacturing details, and modify it to suit the tooling they already have. AWV did that a lot. c) work out a completely new design that conforms to the data in the registered data sheet.
So long as the tube does work as set out in the data sheet, it got the same type number regardless of whether it was the same internally or not.
So, a British tubemaker may come up with a new tube type, and register it to get an American type number. If their is a deamnd, and American firm might have decided to make tubes of the same type number - with their own, different internal design. It amy be better than the original British tube or it may not. Is this variant then a British type or an American type? You could argue either way.
2.
All American tube manufactuers, including Amperex, were in a co-operative arrangment where they engineered tube for each other, and made tubes for each other. Philips had access to American know-how though owning Amperex - one of the reasons why they bought Amperex. So if Philips take an RCA design tube they had access to via Amperex, and put into production in the UK with Mullard adapting it to their tooling, its it an RCA type, a Philips type, or a Mullard type?
The 6CM5 was used regardless of screen size and deflection angle. In Australia, you saw the 6DQ6 in sets made prior to 1960. From 1962 until tube-based TV production ceased, it was always a 6CM5 in the line output socket. And a very reliable durable tube it was. The only exception were the GE portables, which used US-sourced compactron tubes.
I've seen those websites that claim the 6CM5 is a 6.3V version of the PL36 too, but I don't necessarity believe them. I DO believe the old AVW publications I have that describe how the tube was engineered to cope with 110 degree 23 inch TV service. The PL36 (EL36 for 6.3V heater chains) was designed for 90 deg 17 inch sets. Howver a 6CM5 will work in an EL36 socket. A PL36 will work, but not for long perhaps, in a 6CM5 socket.
It is a complete waste of time trying to work out the origin of tubes, or anything, from looking at databooks. Each manufacturer sold a complete line, each type as soon as everybody else. But each manufacturer made only the types they decided on.
The same thing happened with semiconductors. I recall somewhat of a revolution in Australia when Philips and Mullard simultaneously announced in 1966 the production of the silicon planar transistors BC107/BC108.BC109. With every indication they were local Philips made - but not actually clearly stating so. They were a huge leap forward in performance over the then current germanium type AC126, which was indeed a Philips type 100%. In fact they were buying the BC107/8/9 wafers by contract from a couple of US manufacturers. All Philips did was separate them and put them into their TO39 cases.
Last edited: