UK Vintage Radio Repair and Restoration Powered By Google Custom Search Vintage Radio Service Data

Go Back   UK Vintage Radio Repair and Restoration Discussion Forum > General Vintage Technology > Components and Circuits

Notices

Components and Circuits For discussions about component types, alternatives and availability, circuit configurations and modifications etc. Discussions here should be of a general nature and not about specific sets.

Reply
 
Thread Tools
Old 30th Jul 2019, 10:09 am   #41
turretslug
Nonode
 
Join Date: Nov 2011
Location: Surrey, UK.
Posts: 2,903
Default Re: EF91 to EF80

There's probably a world history influence here as well- the predecessor EF50 with its large 9-pin base was a Dutch-origin valve, this 9-pin base enabled the development of the improved VHF performance EF54 with no less than 4 cathode connection pins- this feature may have been late in the day by comparison with the RCA 6SG7/6SH7 but was seized with gusto once apparent. Immediately post-war, UK valve procurement would have been dominated by US production/export/availability and the B7g base, apparently developed in the late '30s US for battery portable use, was then exploited for a new generation of compact high-performance VHF valves that were in high demand for military use. There would have been pressure to use this proven investment in B7g production lines, rather than start work on a base with more pins. Hence lots of 6AM6s everywhere in the late '40s, the desire for a separate suppressor connection limiting cathode connection to a single pin. Once post-war investment in valve production got under way, a B9a EF91 successor with duplicate cathode pins would be seen as an early good idea, even necessity.
turretslug is offline   Reply With Quote
Old 30th Jul 2019, 6:22 pm   #42
Mike. Watterson
Heptode
 
Join Date: Dec 2014
Location: Limerick, Ireland.
Posts: 878
Default Re: EF91 to EF80

RCA, Philips, Tungsram and Telefunken all seem to have been developing button bases to replace the pinch.
Reasons:
Automation (saving money)
Performance: Shorter leads = less inductance, better RF
Size: The Acorn though proved to be a mistake. Too hard to make with too many rejects.

Here is a rather poor article with some good links that I did.
https://www.radiomuseum.org/forum/fr...to_button.html

I don't think the B7G was solely for battery valves, though RCA's 1940 (radios in 1941 before Pearl Harbour) were a miniaturised version of Sylvania's 1938 first "All Dry" 1.4V valves using 50mA filaments. Telefunken instead optimised for parallel battery filaments in the Y8 low profile metal case with button base. These were used during WWII and into early 1950s, killed off by Philips 25mA types.

So Dutch, Americans and Germans simultaneously doing button bases to reduce manufacturing costs and improve RF performance.
The Dutch Philips EF50 9 pin similar to Sylvania loctal and the Philips B8G 8 pin with pip to replace B7G and Octal on mains and battery valves. DK40 good example vs 1L6.

The EF91 was inevitably to be replaced by a Noval.
The 7 pins and 8 pins wasn't enough. Though there was an B7G triode hexode for mains using one heater pin for cathode and the US sets tended to use Heptodes/Octodes and not worry about valve count.
https://www.radiomuseum.org/forum/ea..._innovals.html
The Noval 9 pin base was inevitable for Europe/UK to avoid top caps, screened RF Pentodes, ECH81, EABC80 etc. Also more power than B7G.
RCA wasn't concerned with valve count or Triode-Hexodes, hence 7 pins on B7G rather than 8 or 9. Also allowed a smaller package for the military gear; shell & bomb fuzes, communications, walkie-talkies. The battery valve personal sets was an obvious publicity for the miniature tubes.
https://www.radiomuseum.org/forum/co...es_valves.html

Very many sets before USA entered WWII. But certainly one goal was smaller, higher performance military gear.
https://www.radiomuseum.org/forum/rc...article.html#3
Mike. Watterson is offline   Reply With Quote
Old 30th Jul 2019, 6:53 pm   #43
G6Tanuki
Dekatron
 
G6Tanuki's Avatar
 
Join Date: Apr 2012
Location: Wiltshire, UK.
Posts: 7,768
Default Re: EF91 to EF80

I always thought that the first B7G valves were the 9001/9002/9003

http://www.r-type.org/exhib/aab0161.htm
http://www.r-type.org/exhib/aaa0254.htm
http://www.r-type.org/exhib/aaa0253.htm

which were essentially Acorns re-engineered to avoid the seal-cracking-and-loss-of-vacuum issues that beset the Acorns.

The 717A

http://www.r-type.org/exhib/aab0012.htm

was also an intriguing interim valve - though it used an Octal base it minimised leads-inside-the-envelope issues by turning the electrode assembly on its side so it could be fitted as close as possible to the base.

Equally, I have always liked the 6AK5/EF95

http://www.r-type.org/exhib/aaa0122.htm

that was used in a number of 1940s/1950s-era civilian-air-band [100-150MHz] receivers and some similar-timescale civilian mobile-radios. Small, efficient, alas it was not good for use in VFOs because of the suppressor-grid-to-cathode strapping.
G6Tanuki is offline   Reply With Quote
Old 31st Jul 2019, 9:17 am   #44
Mike. Watterson
Heptode
 
Join Date: Dec 2014
Location: Limerick, Ireland.
Posts: 878
Default Re: EF91 to EF80

Quote:
Originally Posted by G6Tanuki View Post
I always thought that the first B7G valves were the 9001/9002/9003

http://www.r-type.org/exhib/aab0161.htm
http://www.r-type.org/exhib/aaa0254.htm
http://www.r-type.org/exhib/aaa0253.htm

which were essentially Acorns re-engineered to avoid the seal-cracking-and-loss-of-vacuum issues that beset the Acorns.
The Acorn was a bad idea, both to make and in service, so logical some re-engineered. However B7G battery valves were demonstrated in 1940 with very many radios in 1941.
The re-engineered Acorns were a couple of years later, 1943.
The B7G valves only imported to UK during WWII for military and covert applications. Two radios (a TRF, Sweetheart and Superhet) were built in the UK during WWII. No domestic sets used the B7G in UK till 1946.
https://www.radiomuseum.org/r/simons...heart_311.html

This is like a military version of a US 1941 Personal Radio.
https://www.radiomuseum.org/r/mil_gb_op_3_type_301.html
Except the US Models only had MW (Broadcast).
UK models using B7G battery valves adding LW (or SW for export) appeared in 1948.

I don't know when the first mains sets in UK using B7G valves appeared. Possibly ITT-KB and EMI (HMV) would have been first.

Last edited by Mike. Watterson; 31st Jul 2019 at 9:31 am.
Mike. Watterson is offline   Reply With Quote
Old 31st Jul 2019, 11:08 am   #45
Mike. Watterson
Heptode
 
Join Date: Dec 2014
Location: Limerick, Ireland.
Posts: 878
Default Re: EF91 to EF80

The USA 6AM6 (EF91) and 6AK6 (EF95) are probably the originals and seem to be Jan 1949 and 1942 (though registered in May 1943). The 6AK6 seems to have been for military applications at first.

The USA preferred Heptodes / Octodes etc to Triode-Hexodes in radios even in 1930s. Is the 1945 6BE6 = EK90 = 6А2П = X77 = X727 = CV453 = HM04 = 6H31, the first B7G mains type? The first B7G was the battery 1R5 = DK91 = 1А1П = VT-171 = X17 = 1A2_1R5 in 1940.

So perhaps the mains/indirect versions of the B7G were two years after the battery valve developments with a full domestic radio line-up not available till 1945.

In any case the EF91 was sooner rather than later going to be obsolete in Europe.
Mike. Watterson is offline   Reply With Quote
Old 8th Aug 2019, 5:35 am   #46
Synchrodyne
Nonode
 
Join Date: Jan 2009
Location: Mt. Maunganui, New Zealand
Posts: 2,448
Default Re: EF91 to EF80

Quote:
Originally Posted by Mike. Watterson View Post
The USA 6AM6 (EF91) and 6AK6 (EF95) are probably the originals and seem to be Jan 1949 and 1942 (though registered in May 1943).
I think as already stated, the 6AM6 was not the original of the EF91. Rather it was the American designation for the Z77 et al, dating from 1947, the registration having been done by Cossor. Once this was done, Brimar switched to the 6AM6 moniker, having previously used 8D3.

I am not aware that the 6AM6 was ever used in American equipment. The VHF pentode in the initial (1945) American miniature domestic receiving range was the 6AG5, drawn from the existing WWII range. Early updates were the 6BC5 (c.1949) and the 6CB6 (1950), still on the B7G base.

The early history of B7G valves is well-covered in these two RCA articles:

Battery Miniatures from RCA Review 194004.pdf Miniature Tubes in War and Peace from RCA Review 194706.pdf

I don’t know when the EF91 became obsolete, but as already noted, it was still nominated for new applications in the early 1960s, presumably in applications where the EF80 or other later valves offered no advantages, or perhaps even where the EF91 was directionally better. Neither had the B7G base become obsolete in British equipment by that time. For example, a look at the valve lines-up for Eddystone HF receivers that were new in the 1960s shows a B7G majority, B9A minority in each case. I am not sure that the available information really supports the notion that the EF91 would inevitably be replaced by a noval. In fact there was not a direct noval replacement. The EF91 and the EF80 certainly overlapped, but were enough different that they co-existed for most of the latter part of the valve era, and anyway, as already observed, the EF80 was conceived as a (mostly) improved derivative of the EF42. That the latter was of the Rimlock type meant that it was inevitably replaced by a noval.


Cheers,
Synchrodyne is offline   Reply With Quote
Old 8th Aug 2019, 5:37 am   #47
Synchrodyne
Nonode
 
Join Date: Jan 2009
Location: Mt. Maunganui, New Zealand
Posts: 2,448
Default Re: EF91 to EF80

Another perspective on the EF91 vs. EF80 comparison for FM receiver applications was provided in respect of the Amos and Johnstone (BBC) tuner circuit published in WW 1955 April. In respect of the pentode choice, it was said: “There is considerable latitude in the choice of valves. The four pentodes are of the same type; the authors have successfully used B7G-pentodes of the EF91 type and B9A-pentodes of the EF80 type.” Another comment, in respect of the RF amplifier, was: “ .and the valve input resistance is approximately 2,000Ω for an EF91 and 4,000Ω for an EF80.” That suggests that more gain could be had from the EF80 than the EF91.

The full valve line-up for this tuner was : EF91 or EF80, RF amplifier (with AGC from the limiter grid); EF91 or EF80, mixer; ECC81, oscillator and AF cathode follower; EF91 or EF80, IF amplifier; EF91 or EF80 IF amplifier/limiter; crystal diode ratio detector; EM34, tuning indicator arranged to provide a centre-channel indication.

The apparently odd valve line-up of the first Lowther FM tuner has already been noted. It does seem possible that this was derived from another BBC design by J.G. Spencer for a simple FM receiver, published in WW 1951 November. This had the following valve line-up: 6F12 (EF91), RF amplifier; X81, mixer-oscillator; 6F12 (EF91), IF amplifier; EF42, limiter; 6T8, discriminator and AF amplifier; KT61, output. The X81 was soon replaced by the X79. The use of a triode heptode frequency changer was “of its time”. For example, of 5 German AM-FM receivers examined by the BBC as part of its Research Department Report G052, 1953/02, two had an ECH42 frequency changer for both FM and AM, another used the ECH81 for the same purpose, one used the ECH81 triode as a self-oscillating frequency changer on FM, and one used an EF80 as a self-oscillating FM frequency changer. So 60% of that population used a triode hexode/heptode for FM frequency changing.

In this light, one may reconsider Lowther’s valve line-up as not ab initio choices, but as stemming from addressing the question – which of the newer valves might offer improved performance as compared with the BBC originals. Thus the 6BW7 replaced the EF91 as RF amplifier, but the EF80 was its replacement for the IF amplifier. Whilst the EF80 would have been better than the EF91 as RF amplifier, the higher input impedance of the 6BW7 probably allowed the realization of an even higher gain increment. On the other hand, in the IF position, whilst the EF80 was slightly better than the EF91, possibly stability considerations allowed no further material improvement through the use of the 6BW7, hence the EF80 was chosen. The EF42 was left in place as the limiter, which tends to confirm that it was excellent in this role – and perhaps that it was still current in 1953.

Once the EF42 was non-current, the modal fallback choice for limiter appeared to have been the EF80, with some use of the ECF80 pentode section, and rarely, the 6BW7. Towards the end of the valve era, the EF184 was also used, for example in the Pye HFT109 and HFT113. I haven’t seen the schematic for the HFT109, but apparently the 3-stage IF stripped used an EF89, EF80, EF184 combination. The EF89 is consistent with the Philips view as to which was the best valve for an FM IF gain stage. The second IF stage might have been a combined amplifier/limiter, with the third a full limiter. If so one might deduce that the EF184 made an excellent limiter, but was less suitable for an FM IF gain stage, perhaps for stability reasons, hence the choice of EF80 for the second stage. On the other hand, I recall some NZ valve-era TV sets that had a single EF194 as 5.5 MHz intercarrier sound amplifier, which would have been a relatively narrowband application, at least as compared with a vision IF stage. So maybe there were other factors at work in the Pye HFT109 case.

In 1955, Weyrad offered an FM-AM front end that had a 6AM6 (EF91) FM RF amplifier and a 6BE6 frequency changer for both bands, as shown in the following WW item.

Click image for larger version

Name:	WW 195506 p.260 Weymouth.jpg
Views:	21
Size:	78.2 KB
ID:	188030



Cheers,
Synchrodyne is offline   Reply With Quote
Old 30th Sep 2019, 12:00 am   #48
Synchrodyne
Nonode
 
Join Date: Jan 2009
Location: Mt. Maunganui, New Zealand
Posts: 2,448
Default Re: EF91 to EF80

A point of interest here is that the EF91 was not mentioned in any of the Philips books of the era about receiving valves. Book IIIA covered the radio Rimlocks, the B7G battery miniatures and the very early radio novals. The EF42 and UF42 were included, but not the EF91. Book IIIB covered the additional FM-AM era valves, mostly novals, including the ECH81 and EF85, but not the EF91. Book IIIC covered the initial range of TV valves, including the EF80, but not the EF91. I also have the impression that the EF91 was probably rare in European FM and TV equipment, and that the EF80, and before it, the E/UF42 used where high-slope VHF pentodes were required.

It would appear then that Philips saw the EF91 as belonging solely to its B7G industrial range, and so did not offer it for domestic receiving applications. Mullard obviously took a different tack with the EF91, although others in the Philips/Mullard B7G industrial range, such as the EAC91 and ECC91, stayed in their places and were seldom used in domestic equipment.

Possibly Mullard was influenced by the competitive landscape in the UK at the beginning of the miniature era, wherein Osram, Brimar and Cossor had all gone the B7G route, and all offered essentially the same high-slope pentode (Osram Z77, thought to be the original, Brimar 8D7 and Cossor 6AM6) for both TV and industrial applications. Thus Mullard felt it necessary to offer the EF91 as well as the EF42 for TV applications. The same was true of Mazda, who I think offered its Z77 clone, the 6F12, as well as its Rimlock 6F13 for TV applications.


Cheers,
Synchrodyne is offline   Reply With Quote
Old 30th Sep 2019, 9:00 am   #49
kalee20
Dekatron
 
Join Date: Feb 2007
Location: Lynton, N. Devon, UK.
Posts: 5,138
Default Re: EF91 to EF80

Quote:
Originally Posted by Nuvistor View Post
One problem I found was the EF91 was not as reliable as the EF80.
Quote:
Originally Posted by AC/HL View Post
The EF80 is slightly later, and is fully screened internally, so doesn't need a can.
I'll bet the two are connected: a canned EF91 is bound to run hotter than a naked EF80, other things being equal.

Quote:
Originally Posted by Synchrodyne View Post
In the Mullard VATA for the EF91, attached to post #16 above, it was said:

“In a high gain I.F. amplifier the important factors are slope and input and output capacitances. The ratio gm/(Cin +Cout) is a useful figure of merit for I.F. valves, since it is a measure of the gain and bandwidth product. The figure of merit for the EF91 is higher than for any other high-slope pentode.”

The EF91 had a nominal gm of 7.65 mA/V, Cin of 7.0 pF and Cout of 2.0 pF. That gave it a figure of merit of 7.65/(7.0 + 2.0) = 0.85.

The EF80, still in the future, but only just, when that VATA was published (1949 September), had a nominal gm of 7.4 mA/V, Cin of 7.5 pF and Cout of 3.3 pF, for a figure of merit of 0.69, essentially the same as the EF42.
I'm a bit surprised by this. The EF91 is smaller so would be expected to have smaller capacitances. But I wonder if the comparisons are 'real, for instance, was the EF91 Cout measured with or without a screening can? The EF80 is stuck with its internal screen, whereas the EF91 does have some flexibility in that respect, and if push came to shove a large-diameter screening enclosure could be used, to increase distance from anode and reduce Cout. (Of course, if you need a 2" screen around your valve, some of the benefits of miniaturisation disappear anyway!).
kalee20 is offline   Reply With Quote
Old 30th Sep 2019, 8:34 pm   #50
PJL
Dekatron
 
Join Date: Sep 2005
Location: Seaford, East Sussex, UK.
Posts: 4,246
Default Re: EF91 to EF80

EF80's are indestructible and EF91's may not have even made it to the end of warranty. Hardly surprising Philips were not promoting it.
PJL is offline   Reply With Quote
Old 30th Sep 2019, 11:06 pm   #51
Synchrodyne
Nonode
 
Join Date: Jan 2009
Location: Mt. Maunganui, New Zealand
Posts: 2,448
Default Re: EF91 to EF80

Quote:
Originally Posted by kalee20 View Post
I'm a bit surprised by this. The EF91 is smaller so would be expected to have smaller capacitances. But I wonder if the comparisons are 'real, for instance, was the EF91 Cout measured with or without a screening can? The EF80 is stuck with its internal screen, whereas the EF91 does have some flexibility in that respect, and if push came to shove a large-diameter screening enclosure could be used, to increase distance from anode and reduce Cout. (Of course, if you need a 2" screen around your valve, some of the benefits of miniaturisation disappear anyway!).
As Mullard didn’t state the conditions under which the measurements were made, we can but guess. Also, the statement about the EF91 being the best along this vector was made before the EF80 was released.

The Philips data sheet for the EF91 quoted the capacitances with a screening can as Cg1 7.3 pF, Ca 3.4 pF and Cag1 <0.01 pF. That does suggest that the Mullard numbers, Cin 7.0 pF, Cout 2.0 pF and Cag1 < 0.008 pF were measured without the screening can; the Mullard datasheet is mute in this regard.

A perspective on the EF80 vs. EF91 performance was provided by the quote from Jason in post #39:

Quote:
Originally Posted by Synchrodyne View Post
An interesting commentary on the EF80 as compared with the EF91 for the FM case was provided by Jason in respect of its Jasonkit FMT2 of 1958. The FMT2 was essentially a self-powered version of the FMT1 in the same case as used for the then-new FMT3. The FMT1 was in 1958 the new name for the original Jason/Jasonkit local-area FM tuner circuit that used four Z77 (EF91) valves, by then available only in Jasonkit form. The FMT2 had essentially the same circuit but used four EF80 valves. (The Jasonkit FMT3 also of 1958 used five EF80 and one ECC81; it was different to the Jason (built version) FMT3.)

[I]“The valves have been changed to EF80, and this results in an improved sensitivity. In the RF and FC stages the fact that there are two cathode leads reduces the feedback which is experienced at this point and gives a slight increase in gain. In the IF stage advantage has been taken of the slightly lower anode-to-grid capacity which is found when the valveholder capacities are also included. Therefore less damping may be used across the coils and more gain results.”
This I think better aligns with your expectations.


Cheers,
Attached Files
File Type: pdf EF91 Philips 19990629.pdf (31.4 KB, 6 views)
File Type: pdf EF91 Mullard Issue 1.pdf (238.9 KB, 4 views)
Synchrodyne is offline   Reply With Quote
Old 1st Oct 2019, 12:52 am   #52
Synchrodyne
Nonode
 
Join Date: Jan 2009
Location: Mt. Maunganui, New Zealand
Posts: 2,448
Default Re: EF91 to EF80

The apparent Philips demarcation between the EF80 as a domestic receiving valve and the EF91 as an industrial valve turns up in the following Pye examples. During the 1950s, Pye used many Mullard valves in its domestic equipment. The FenMan II FM-AM radio receiver of mid-1955 was an example, with 2 x EF80, 1 x ECF80, 1 x ECH81, 1 x EF85, 1 x EABC80, 1 x EBC41, 1 x EL84, 1 x EZ80, 1 x EM80. These were all from the Philips/Mullard noval domestic series except for the Rimlock EBC41, probably used because the European release of the EBC81 was held back, as noted in this thread: https://www.vintage-radio.net/forum/...d.php?t=159120. In particular the EF80 was used in the positions (FM RF amplifier and FM limiter) where a high-slope, sharp cutoff VHF pentode was required.

On the other hand, on the industrial product side, the Pye Telecommunications PTC145 VHF monitor receiver of the same period, covering 68 to 174 MHz, had the valve line-up (AM version) was: EC91, 2 x EF91, ECF82, 2 x EF92, ECC82, EZ90. Here the EF91 was used in the positions (1st mixer and 2nd IF amplifier) where a high-slope, sharp cutoff VHF pentode was required. Now it could be that the EF91 was preferred over the EF80 for technical reasons – perhaps it might have been better in very narrow-band applications – but the overall valve line-up suggests a preference for B7G types except where B9A was unavoidable.


Cheers,
Synchrodyne is offline   Reply With Quote
Old 1st Oct 2019, 11:28 pm   #53
Synchrodyne
Nonode
 
Join Date: Jan 2009
Location: Mt. Maunganui, New Zealand
Posts: 2,448
Default Re: EF91 to EF80

Quote:
Originally Posted by PJL View Post
EF80's are indestructible and EF91's may not have even made it to the end of warranty. Hardly surprising Philips were not promoting it.
Assuming that assertion to be statistically supported, one may wonder whether the valve makers put any effort into obtaining longer life. The EF91 was used quite extensively in industrial equipment, where durability would have been important. For example, the Marconi HR24 dual-diversity point-to-point SSB receiver valve complement included 25 EF91 out of 70 total. The EF91 was used in most oscillator positions, where one would think that retention of high gm was critical. (The 70 valves included two octals (both voltage stabilizers) and 4 novals (AFC motor drivers); the rest were all of the B7G type.)

Receivers in the HR24 class had built-in valve checking equipment, so that failing valves could be detected before they caused any interruptions. Thus the short life of the EF91 would have been quite visible, and one might have expected Marconi to either ask for better from its suppliers or change to a more durable valve type. Or maybe the EF91 was heavily discounted and users were encouraged to keep a large stock of spares?


Cheers,
Synchrodyne is offline   Reply With Quote
Old 2nd Oct 2019, 1:28 am   #54
Synchrodyne
Nonode
 
Join Date: Jan 2009
Location: Mt. Maunganui, New Zealand
Posts: 2,448
Default Re: EF91 to EF80

In the domestic equipment sphere, the EF91 and EF80 were usually found in TV receivers and FM receivers and tuners, where their apparent relative merits have already been discussed.

But to a lesser extent, they were also found in AM receivers and tuners, and also in AF roles. Following are a few examples.

There was a viewpoint that for domestic receivers that included SW bands, lower noise at the higher frequencies was achievable by using a high-slope VHF pentode as RF amplifier rather than a conventional HF remote cutoff pentode. One example was the Ambassador Viscount (original version), which had an EF80 RF amplifier in what was a mostly Rimlock valve line-up. Others using this technique were Murphy (Mazda 6F1 in the TA160) and Armstrong (Mazda 6F13 in the EXP119 and early BS125.) In these cases I suspect that the valve choice had less to do with the subtleties of the various high-slope valves than with what each setmaker’s established valve suppliers were offering at the time. Ambassador appeared to be in the Mullard camp, and was probably already using the EF80 in its TV receivers. Murphy was in the Mazda camp, and the timing of the TA160 (later 1951) was such that by then the improved 6F1 was available. Armstrong was Mazda oriented at the time it moved to miniature small-signal valves late in 1949, and the EXP119 was probably developed when the 6F13 was its high-slope offering, and just before the 6F1 became available.

On the industrial side, it may be noted that the same technique was also favoured by Marconi, who used an EF91 1st RF amplifier (with an EF92 2nd RF amplifier) in many of its point-to-point receivers.

Another approach to lower noise in the higher SW frequencies was to use an aperiodic grounded grid triode input stage. Dynatron did this with its T139 tuner/control unit of 1954, using a triode-strapped Z77 (EF91). Dynatron had moved to Osram valves when it went to miniatures with the T99 in 1951, and was already using the Z77 elsewhere in its equipment. So commonality was probably a reason for its choice in this case. Not only that, the Z77 was specified for triode as well as pentode operation. The T139 was designed for use with an outboard FM tuner, the FM1, which was released at the same time or very soon after the T139. The FM1 used no fewer than 5 Z719 (EF80) valves. So had Dynatron seen the Z719 as a better choice than the Z77 for the T139, it would not have been a logistical problem.

As an aside, it could be that the aperiodic grounded-grid RF input stage, as well as being beneficial from a noise viewpoint, also helped with selectivity, in that the initial tuned circuit (following it) would not need to be such a trade-off between noise and selectivity, nor would it need to be configured to accommodate a quasi-random range of aerial impedances.

Dynatron also used triode-strapped Z77 in both the driver and phase splitter positions in its LF512, LF612 and LF613 audio power amplifiers. The T139 also included a triode-strapped Z77 in its tone control stage, and I think that the same applied to the T99. The T139 also had a triode-strapped Z729 (EF86) for the gramophone input stage. One could infer from this deployment that the triode-strapped Z77 made a competent “line” level AF amplifier, suitable for high quality work, such that it was not necessary to go to the Z729 outside of low signal level situations.

Rogers used a 6AM6 (EF91) as AGC side-chain IF amplifier in its RD Junior AM tuner of 1953. The main IF amplifier was a 9D6 (EF92). In this case the choice of 6AM6 probably came about because Rogers was using Brimar valves for this unit, and the 6AM6 was then a mainstream Brimar valve.

I don’t think that the foregoing helps much with the basic EF91 vs. EF80 question, but if nothing else, it does demonstrate the versatility of the EF91. Probably the EF80 would also have fitted the AM and AF applications for the EF91 delineated above, but perhaps without real advantage and it might also have been seen as overkill.


Cheers,
Synchrodyne is offline   Reply With Quote
Old 2nd Oct 2019, 8:37 am   #55
kalee20
Dekatron
 
Join Date: Feb 2007
Location: Lynton, N. Devon, UK.
Posts: 5,138
Default Re: EF91 to EF80

Two AF uses for both valves:

The Soundmirror tape recorder uses two EF91's, one as second amplifying valve (playback) or record output stage to drive the head; the other is used as a 28kHz bias oscillator on record only.

Philips have an amplifier circuit which, rather oddly to me, uses two EF80's as phase-splitter to drive the output valves! I can't recall the circuit in full detail (hopefully someone may); but given that they could have used one of their own double-triodes, why they would use two high-slope RF pentodes beats me.

EF91's and their equivalents - I can't comment on inherently shorter life. PJL's remark about barely making it to end of warranty period might be less a reflection on expected-life-in-service, and more a comment that infant mortality was higher, maybe quality control of the new-fangled miniature all-glass valves was not fully developed, and the later EF80 reaped the benefits? In which case, burn-in screening for industrial valves would have helped: those that 'died' within the first 500 hours never left the factory; those that passed were defect-free and were shipped, secure in the knowledge than many tens of thousands of hours of performance could be expected.
kalee20 is offline   Reply With Quote
Old 2nd Oct 2019, 9:53 am   #56
Nuvistor
Dekatron
 
Nuvistor's Avatar
 
Join Date: Aug 2013
Location: Wigan, Greater Manchester, UK.
Posts: 7,200
Default Re: EF91 to EF80

Perhaps the industrial equipment used more conservative ratings for the valves whereas domestic equipment ran the valves much harder.
__________________
Frank
Nuvistor is offline   Reply With Quote
Old 4th Oct 2019, 3:42 am   #57
Synchrodyne
Nonode
 
Join Date: Jan 2009
Location: Mt. Maunganui, New Zealand
Posts: 2,448
Default Re: EF91 to EF80

The Radiomuseum.org site allows one to do a “reverse lookup” on valve types to see which equipment each was used in. A cursory use of this facility shows that the EF91 was not very common at all in European domestic receivers, although it was found in some instruments. But it was quite common in British domestic receivers. On the other hand, the E/UF42 was widely used in European domestic receivers, but less so in British models. Of course, the EF80, once it was available, was very widely used in Europe and the UK.

This limited survey does at least directionally support the notion that Mullard did differently to the parent organization. There might have been an element of doing differently to headquarters simply for its own sake. A possible example of that was its (semantically illogical) renaming of the EQ80 enneode as a nonode. As already mentioned, the UK competitive scene might have been a factor, but also perhaps what might be called the “non-competitive” scene that arose from the cartel of the time. That may have fostered a situation where it was desirable that all suppliers offered a common high-slope pentode, namely the Z77 and its clones. If so, Mullard could well have deemphasized the E/UF 42. On the other hand, it took the lead with the EF80 generation, with most of the other UK valve makers cloning it, e.g. the Osram Z719 and Brimar 6BX6.

Quote:
Originally Posted by G8HQP Dave View Post
I read somewhere that the EF80 could provide performance with 170V supply rail which an EF91 needed 250V to deliver. The two cathode pins means that you can optimise either in-out isolation or input conductance, whichever is most important for your design.
Looking at the numbers, those for the EF42 and EF91 were quoted only for an anode voltage of 250, at which the respective slopes were 9 and 7.65 mA/V both at 10 mA anode current.

The UF42 was the same as the EF42 except for its heater, and its performance was quoted at an anode voltage of 170, at which its slope was 8 mA/V, still at 10 mA anode current, a bit lower than what the EF42 achieved at 250 volts. That suggests that the valve was optimized for 250 volts, and that the fall-off – not that large - when operated at 170 volts was simply accepted. By extension, the EF42 at 170 volts would have given 8 mA/V.

EF80 data was quoted at 170, 200 and 250 anode volts, with respective slopes of 7.4, 7.1 and 6.8 mA/V, all at 10 mA anode current. So clearly it was optimized for the 170-volt case, typical of AC-DC chassis that might have to operate on supply voltages as low as 200.

For the 6AM6 Brimar quoted a slope of 7.5 mA/V at both 200 and 250 anode volts and respective anode currents of 9 and 10 mA. Osram gave 7.5 mA/V at 250 volts, 10 mA for the Z77, the same when triode-strapped at 250 volts, when the µ was 75.


Cheers,
Synchrodyne is offline   Reply With Quote
Old 4th Oct 2019, 7:49 am   #58
Radio Wrangler
Dekatron
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 12,607
Default Re: EF91 to EF80

Back in the sixties, I remember that the EF80 was looked upon as a 'telly valve' and if one turned up in any other application it was looked askance at and dismissed as a cheapening exercise or maybe just what the designer had in his junk box.

This was probably grossly unfair on the EF80.

At the time, I was making myself progressively better oscilloscopes, and I'd bought a tea-chest of surplus CV138 (EF91/Z77/6AM6....) from Jim Fish at his shop on Chapel Hill. So I guess I fell into the 'it's what he had in his junk box' camp. I also had plenty of EF80s in the junk box, but I'd never have thought of using them back then.

I guess the EF80 suffered typecasting and an image problem. The nasty staining seen inside old ones didn't give an aura of high quality.

The EF80 is also still so cheap that the audiophiles won't latch onto it. Zero, possibly negative, esotericity!

If I ever do a valve data book, I think I'll put a column in the tables labelled 'esotericity' maybe defined as a factor of today's market price divided by original price when new.

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is online now   Reply With Quote
Old 14th Oct 2019, 11:09 pm   #59
Synchrodyne
Nonode
 
Join Date: Jan 2009
Location: Mt. Maunganui, New Zealand
Posts: 2,448
Default Re: EF91 to EF80

I suppose as a summary answer to the original question one could say that in the late 1940s, Mullard offered both the EF42/UF42 and the EF91 for TV and other applications requiring high-slope pentodes, whereas Philips generally offered only the EF42/UF42. The Mullard duality probably stemmed from the UK competitive situation, wherein several majors were offering complete B7G receiving ranges.

Philips developed the EF80 as an improved/optimized successor to the EF42/UF42, with a fairly quick turnover from one to the other. Likewise Mullard also adopted the EF80 from its initial availability, quickly replacing the EF42/UF42. The EF91 remained in the range for industrial and miscellaneous purposes, so was still available for TV and radio applications. But for TV (and some radio) applications, the EF80 was the better choice, so there was a turnover from the EF91 to this valve, although at a slower pace than was the case with the turnover from the EF42/UF42.

The Mullard late 1940s duality was also apparent in its simultaneous offering of the EB41 and EB91 for receiving applications, as recorded in the concurrent thread Why Both EAA91 and EB91? (https://www.vintage-radio.net/forum/...d.php?t=160476).


Cheers,
Synchrodyne is offline   Reply With Quote
Old 16th Oct 2019, 12:40 am   #60
Synchrodyne
Nonode
 
Join Date: Jan 2009
Location: Mt. Maunganui, New Zealand
Posts: 2,448
Default Re: EF91 to EF80

Something I have not yet found is “hard” information as to the origin of the EF91 valve group. My best estimate is that the Osram Z77 was the parent valve, the circumstantial evidence supporting this being follows.

In the post-WWII period, the major UK valve makers, Mullard (Philips), Osram, Mazda and Brimar were faced with the need to introduce miniature domestic receiving valve ranges, including high-slope pentodes suitable for TV applications.

First though was the choice of miniature type, which at the time was between the American B7G and Philips Rimlock. Mullard, as a Philips subsidiary, chose Rimlock, as did Mazda. Brimar, whose valve range was based upon American practice, chose the B7G form. Osram also opted for the B7G form.

In respect of the “TV” pentodes, Mullard had the Philips EF42 from the Rimlock range, and Mazda developed the similar 6F13.

Osram’s TV pentode offering was the Z77, as part of its “77” series (which included some “78” suffix valves.) This range included some clones of American types, such as the DH77 (6AT6) and X77 (6BE6), and some own designs, such as the W77 and X78. The Z77 had no counterpart in the American series, and so was either an own-design or a clone of another American or European valve.

Brimar mostly chose American types, but did develop its own, or clone British/European types where there was no American type suitable for the job. Its B7G radio receiving range was essentially American (6BA6, 6BE6, 6AT6, etc.), but the American TV pentode of the time, the 6AG5 (slope around 5 mA/V), may have been seen as not being fully competitive with what was expected from the other British and European valvemakers. (Probably the general features of the Philips EF42, including its very high slope, were known by late 1946, even if production was yet to start.) Thus Brimar offered the 8D3 (later renamed as the 6AM6) as its TV pentode, this being essentially the same valve as the Z77.

The two previous paragraphs point to the Z77/8D3 as being either an Osram or a Brimar development, cloned whichever of the two did not develop it.

Notwithstanding that Mullard and Mazda opted for the Rimlock type for their respective domestic receiving ranges, it appears to have been the case that industrial/commercial and military users would not adopt the Rimlock for general use, but preferred the B7G type. This meant that Mullard and Mazda would need to offer an adequate range of B7G types additional to their Rimlock ranges, including high-slope pentodes.

The minimum effort approach would been to simply develop B7G versions of their respective Rimlock valves, the EF42 and 6F13. On the other hand, the military market in particular liked standardization, and so there was a benefit in cloning whatever their B7G-oriented competitors were offering as high-slope valves. In that case the Mullard EF91 and Mazda 6F12 would have been more-or-less copies of the Z77/8D3.

Had either Mullard or Mazda been the lead developer, therefore with the Z77/8D3 characteristics unknown, it might have been expected that they would have closely followed their Rimlock prototypes, rather than develop valves with somewhat different characteristics. So on the balance of probabilities, I’d say that they were the followers in this case.

That leaves us with either Osram or Brimar as the developer. Osram generally developed its own valves except where American or industry types were a good fit, so it probably expected to have to develop its own TV pentode both for domestic receiving and (Marconi) studio/transmitter equipment (in which I think the Z77 was widely used). Z77 development might well have been the major activity in the whole “77” programme. Brimar on the other hand generally followed American practice, with own development being the exception. It may well have earmarked the 6AG5 as its TV pentode until it saw what its UK competitors were doing. Combined with the military preference for standardization, cloning could well have been seen as the optimum pathway.

So again on balance of probabilities, the Osram Z77 looks more likely as the parent valve than the Brimar 8D3.

Presumably the BVA would have facilitated the interchange of information necessary for cloning. Apparently there was tentative agreement on a new standard receiving range – including high-slope TV pentodes) most on the Rimlock (B8A) base, reported in Wireless World 1946 November, p.375,but events did not turn out quite as planned.

First mentions in Wireless World (that I can find) for some of the above TV pentodes were:

Mazda 6F13 - 1947 June p.228
Osram Z77 - 1947 June p.199
Mullard EF42 & EF91 - 1947 November p.435
Mazda 6F12 – 1949 January p.13 (I think that this was somewhat after it was released.)

American registration for the Z77/EF91, as the 6AM6, was done by Cossor in 1949. (What designation, if any, Cossor may have used before this I don’t know). Brimar changed from 8D3 to 6AM6 soon thereafter.

Although the Mullard EF91 and Mazda 6F12 might have been intended for commercial/industrial and military applications, they were also used in domestic receivers and so each of these valvemakers had two horses, as it were, in the TV pentode stakes of the late 1940s, one B7G and one Rimlock.


Cheers,
Synchrodyne is offline   Reply With Quote
Reply

Thread Tools



All times are GMT +1. The time now is 7:49 am.


All information and advice on this forum is subject to the WARNING AND DISCLAIMER located at https://www.vintage-radio.net/rules.html.
Failure to heed this warning may result in death or serious injury to yourself and/or others.


Powered by vBulletin®
Copyright ©2000 - 2019, vBulletin Solutions, Inc.
Copyright ©2002 - 2019, Paul Stenning.