|
General Vintage Technology Discussions For general discussions about vintage radio and other vintage electronics etc. |
|
Thread Tools |
15th Mar 2009, 6:48 am | #1 |
Nonode
Join Date: Jan 2009
Location: Papamoa Beach, Bay of Plenty, New Zealand
Posts: 2,943
|
Valve Questions
Does anyone know when the EF89 valve was released?
As best I can determine, it may have been a year or two later than the circa 1953 “crop” of B9A valves that were developed for FM/AM receivers, including the ECH81 and EF85. The ECH81 seems to have been intended not only to do the same basic job as the ECH42, but the heptode was also designed to be a credible FM IF amplifier. And the EF85 had a higher slope than the EF41, ostensibly to better suit it for use as an FM IF amplifier. I recall reading somewhere that it was more than needed for AM IF applications, which was why the EF89 was developed, with slope intermediate between those of the EF41 and the EF85. Also, one might infer from the datasheet that the EF89 was also designed to work alternatively with no standing bias, thus allowing some economy in domestic radio receiver circuits through the elimination of the cathode bias resistor and its bypass capacitor. And did the EBF89 appear at the same time as the EF89? It seems to be, to a first approximation at least, an EF89 with two diodes added. The earlier EBF80 looks as if it were more-or-less an EF41 with two added diodes – or maybe an EAF42 with a second diode added – and although of necessity it had a B9A base, it was effectively a member of the B8A family, and used as such. E.g., the early Chapman AM tuners such as the S4 and S5 used an ECH42, EF41, EBF80 line-up. Thanks in advance. Cheers, |
15th Mar 2009, 7:12 am | #2 |
Heptode
Join Date: Jan 2007
Location: Parkes, NSW Australia
Posts: 877
|
Re: Valve Questions
1954 for both according the details on radiomuseum.org.
Peter |
15th Mar 2009, 12:06 pm | #3 |
Octode
Join Date: Jan 2003
Location: Ware, Herts. UK.
Posts: 1,082
|
Re: Valve Questions
In the UK at least, earlier FM equipped sets used the EF85 as the IF amplifier. The EF89 started to appear in UK sets around 1957 or '58. Some European sets changed over to the EF89 a bit earlier, around 1955 or '56. My 1956 Grundig 5080 has an EF89 as does this set
http://vintage-radio.net/forum/showt...highlight=ef89 John |
16th Mar 2009, 11:29 pm | #4 |
Rest in Peace
Join Date: Sep 2008
Location: Solihull, West Midlands, UK.
Posts: 4,872
|
Re: Valve Questions
The EF85 was intended mainly for TV IFs, although it was also used for FM and AM IFs. The EF89 was intended for FM and AM IFs. It has lower gm, but also lower capacitance and higher anode impedance. This means that in a narrow-band IF it can give greater gain and better stability than an EF85. A wide band IF, such as TV, benefits from the higher gm of the EF85, as the tuned circuits will have lower impedance. The EF89 also has a particularly smooth AGC characteristic, which might mean that it has lower crossmodulation than EF85.
Although it needs slightly different bias, an EF89 is almost a B9A version of the EF93/6BA6. |
17th Mar 2009, 9:34 am | #5 |
Dekatron
Join Date: Feb 2007
Location: Lynton, N. Devon, UK.
Posts: 7,061
|
Re: Valve Questions
G8HQP Dave, that's a fantastic comparison, succinct, yet very relevant!
|
17th Mar 2009, 9:49 am | #6 |
Octode
Join Date: Apr 2007
Location: Hampshire, UK.
Posts: 1,574
|
Re: Valve Questions
That's an interesting point about the EBF80 being part of the B8A family. It would seem that the UBF80 (same thing but with a 100mA series heater) was the first "U" series valve with a B9A base, it is certainly the only one in the 1951 (I think) Mullard book.
The UBF80 in my set (1952 model year) is the only one in there with a B9A base, the rest are variously Loctals (U-20 series), B8A and side contact (!). The method of construction is also closer to B8A (seperate welded-on disc shaped button base) than B9A (typically a smoother, more fused together appearence). |
20th Mar 2009, 11:24 am | #7 |
Nonode
Join Date: Jan 2009
Location: Papamoa Beach, Bay of Plenty, New Zealand
Posts: 2,943
|
Re: Valve Questions
Thanks G8HQP Dave for the elegant comparison.
Re the AGC characteristics, I had a look at the curves in “Vade Mecum” and the EF89 certainly looks a lot smoother than say the EF85 and EF183, and is relatively straight at the bottom end of the gain range. As well as helping with cross-modulation, it probably also helps with minimizing modulation rise, perhaps useful in a simple domestic receiver (ECH81 plus EF89, etc.) where the need for adequate gain control range would not easily allow the text-book (Langford-Smith) solution of partial control voltage application to the IF stage. The connection between the EF89 and EF93 is interesting, too, and I had missed that previously. The EF93 seems to have remained popular for commercial and communications receivers into the 1960s, e.g. it was used in the late Eddystone valve models (e.g. 880, 830, 940, 850 series) in most RF and IF positions, although the 670C and 840C retained B8A valves, by then becoming unusual. On the other hand, the EF89 seems to have been more of a domestic receiver valve. I have since remembered that the 1955 edition of Radio Servicing Pocket Book includes a listing of valve lines-up for a wide range of British domestic receivers. EF41s, ECH42s, EBC41s and EL41s abound, and now and then amongst them is found an EBF80. FM (and FM-AM) receivers are listed separately, and there are found mostly B9A types, with both the EF85 and EF89 represented, although with the EF85 then more prevalent. Cheers, |
20th Mar 2009, 9:58 pm | #8 |
Rest in Peace
Join Date: Sep 2008
Location: Solihull, West Midlands, UK.
Posts: 4,872
|
Re: Valve Questions
Wild guess coming up: UK communications receiver manufacturers often used EF93 as RF or IF because the 6BA6 was popular in the US (good AGC); UK domestic receiver manufacturers preferred to simplify stock holdings by using all B9A valves, or were more heavily influenced by Philips/Mullard?
Interestingly, the US always seemed to prefer miniature 7-pin RF pentodes with external screening can, while Europe preferred 9-pin with internal screen. The UK, as always, was somewhere between! The EBF89 is an EF89 with diodes. The EBF80 is almost an EF92 with diodes. The EF92 was used for a while in B7G line-ups, as it behaves like the last generation of octal pentodes (gm 2mA/V) so manufacturers could miniaturise without a major redesign. My Pye P45 has an EBF80 among B8A valves - did it come out before the EBF41? |
22nd Mar 2009, 4:40 am | #9 |
Nonode
Join Date: Jan 2009
Location: Papamoa Beach, Bay of Plenty, New Zealand
Posts: 2,943
|
Re: Valve Questions
According to the site kindly advised by petervk2mlg, above, the EBF80 was released in 1950. That was still during the B8A period for British domestic radio applications.
Would it be reasonable to assume that valves like the EF41, ECH41/42, EAF41/42 were also designed to behave more like their octal predecessors? Whilst the commonly used B9A series valves typically delivered improved performance and/or functionality, there were also valves like the EF81 and ECH80 that were essentially B9A versions of the EF41 and ECH42 respectively, although very little used as far as I know. There is evidently an article on IF amplifiers for FM/AM receivers in Mullard Technical Communication 21 of February, 1957. It is listed in the contents for Volume 3, 1957 to April, 1958, of which I have a bound copy, but for whatever reason my copy is missing No. 21. (Evidently simply not bound in rather than subsequently removed. I acquired my volume second-hand quite a few years back from Chevet Supplies when it was still in business, but these Mullard publications don’t seem to around these days, at least not visible on ABE or eBay.) According to the index, the said article mentions the following valves: EBF89, ECH81, EF41, EF85, EF89, EABC80, ECC85, EL84, so it is probably quite interesting generally as well as being pertinent to some of this issues raised in this thread. Cheers, |
25th Mar 2009, 2:03 pm | #10 |
Retired Dormant Member
Join Date: Jun 2006
Posts: 1,798
|
Re: Valve Questions
I have learnt more about these valves by reading this thread than all the data books
EBF89, ECH81, EF41, EF85, EF89, EABC80, ECC85, EL84, All of these apart from the EF41 are still listed in my 1974 Mullard data book. Geof |
25th Mar 2009, 2:47 pm | #11 |
Moderator
Join Date: Jun 2003
Location: Oxford, UK
Posts: 27,787
|
Re: Valve Questions
I suspect the EBF80 was developed specifically for short superhet radios. Previous to its introduction manufacturers had built these using EBL31 output valves, but these were obsolescent by the late 40s. Output valves were a tight squeeze in B8A or B9A envelopes and it was difficult to fit two diodes inside, so Philips decided to put them inside the IF amp valve. This required 9 pins unless internal connections were to be used..The alternative B8A EAF41 only contained a single diode resulting in poor AGC performance.
Philips must have done this through gritted teeth, as they were pushing the B8A standard hard in 1950. Early 50s short superhets almost all have an ECH42 / EBF80 / EL41 lineup. Paul |
25th Mar 2009, 3:41 pm | #12 | |
Dekatron
Join Date: Jan 2003
Location: Bradford on Avon, Wiltshire, UK.
Posts: 3,301
|
Re: Valve Questions
Quote:
__________________
"Nothing is as dangerous as being too modern;one is apt to grow old fashioned quite suddenly." |
|
27th Mar 2009, 12:18 am | #13 |
Dekatron
Join Date: May 2006
Location: Invercargill, New Zealand
Posts: 3,440
|
Re: Valve Questions
|
29th Mar 2009, 4:08 am | #14 | |
Nonode
Join Date: Jan 2009
Location: Papamoa Beach, Bay of Plenty, New Zealand
Posts: 2,943
|
Re: Valve Questions
Quote:
It’s interesting to note, though, that within a few short years, receiver manufacturers were apparently quite happy to abandon delayed agc for the AM sections of FM/AM receivers, using just one diode in the EABC80 for AM detection and agc rectification. I have recently found a September, 1954 “Wireless World” article on FM/AM receivers (pp 431 to 436 of that issue) that throws some light on the subject. Some short excerpts: “A special series of valves has been made available for combination receivers. The Mullard range includes the ECH81 triode-heptode, the EF85 variable-mu pentode with a mutual conduction of 6 mA/V, and the EABC80 triple-diode triode. With the latter, two diodes are used for the f.m. demodulator and the third for detection and a.g.c. on the broadcast bands. Delayed a.g.c. cannot be provided with this arrangement, but as a great deal more gain can be obtained from the i.f. amplifier than is usual in broadcast receivers, this should be of little consequence.” “With standard 470-kc/s i.f. transformers, the EF85 stage gain will be far too high and instability will result unless this is reduced to manageable proportions. This can be achieved either by using i.f. transformers with a suitably low dynamic resistance or, possibly more simply, by switching in supplementary cathode resistance.” It seems, though, that some manufacturers at least, saw that a better solution in using the EF89 instead of the EF85, once it was available. Cheers, Last edited by Synchrodyne; 29th Mar 2009 at 4:23 am. |
|
30th Mar 2009, 9:41 am | #15 |
Dekatron
Join Date: May 2008
Location: Derby, UK.
Posts: 7,735
|
Re: Valve Questions
Am I being thick here? Surely a pentode and two diodes need only 8 pins. I make it two for the heater; four for the pentode (anode, screen grid, control grid and cathode / suppressor grid); and two more for the diodes' anodes?
__________________
If I have seen further than others, it is because I was standing on a pile of failed experiments. |
30th Mar 2009, 10:43 am | #16 | |
Dekatron
Join Date: Jan 2003
Location: Bradford on Avon, Wiltshire, UK.
Posts: 3,301
|
Re: Valve Questions
Quote:
__________________
"Nothing is as dangerous as being too modern;one is apt to grow old fashioned quite suddenly." |
|
30th Mar 2009, 11:39 am | #17 |
Nonode
Join Date: Jan 2009
Location: Papamoa Beach, Bay of Plenty, New Zealand
Posts: 2,943
|
Re: Valve Questions
In the EBF80 and EBF89 the pentode and diode pair shared a common cathode. However, the pentode suppressor grid had its own pin, hence the need for 9 pins. It was normal for RF pentodes to have the suppressor grid brought out to a separate pin, rather than being connected to the cathode internally. I must admit I've forgotten exactly why, but it was considered to be good practice.
Cheers, |
30th Mar 2009, 12:06 pm | #18 | |
Dekatron
Join Date: May 2008
Location: Derby, UK.
Posts: 7,735
|
Re: Valve Questions
Quote:
I'll have a look through some schematics of some old sets (I've got the service data DVD, and can thoroughly recommend it). If there's a good reason for bringing out G3 separately, I'm sure it'll become clear.
__________________
If I have seen further than others, it is because I was standing on a pile of failed experiments. |
|
30th Mar 2009, 12:07 pm | #19 | |
Moderator
Join Date: Jun 2003
Location: Oxford, UK
Posts: 27,787
|
Re: Valve Questions
Quote:
I suppose that, once the EBF80 was available, manufacturers would use it even if it gave a very small improvement if the cost was similar. Paul |
|
30th Mar 2009, 1:56 pm | #20 | |
Retired Dormant Member
Join Date: Aug 2005
Location: Keyworth nr. Nottingham, UK
Posts: 140
|
Re: Valve Questions
Quote:
That might be one reason why so many AM/FM sets are pretty mediocre on AM. The use of the EABC80 implies economic rather than technical considerations were paramount. Not only does it deny the use of delayed AGC, it also limits the ratio detector to the unbalanced form, thereby also removing any chance of deriving an AFC control voltage from this source. A better option would be the rarer EBC81 and an EB91 or germanium diodes, but that costs more . . . . Cheers, Neil |
|