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

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

Notices

General Vintage Technology Discussions For general discussions about vintage radio and other vintage electronics etc.

Closed Thread
 
Thread Tools
Old 19th Jan 2012, 12:24 am   #161
G8HQP Dave
Rest in Peace
 
Join Date: Sep 2008
Location: Solihull, West Midlands, UK.
Posts: 4,872
Default Re: Valve Questions

I thought the ECC82 was aimed at multivibrators for TV line and frame oscillators. It can also be used as a low power push-pull audio output stge.
G8HQP Dave is offline  
Old 19th Jan 2012, 1:44 am   #162
Synchrodyne
Nonode
 
Join Date: Jan 2009
Location: Papamoa Beach, Bay of Plenty, New Zealand
Posts: 2,943
Default Re: Valve Questions

That's certainly where the ECC82 ended up, along with pulse applications as mentioned by kalee20, but it's not clear that Philips/Mullard so positioned it in the early days. Perhaps the original Philips documentation at the time of its release would clarify, but that is something I cannot find.

In the UK at least, Brimar with the 12AU7 was ahead of Mullard, and whilst its early advertising described it as a TV and industrial valve and mentioned oscillator applciations, it didn't specifically include multivibrators, although that could be inferred as a subset of oscillators.

Cheers,
Attached Thumbnails
Click image for larger version

Name:	WW 1950-01 p.A43.jpg
Views:	531
Size:	54.7 KB
ID:	61399  
Synchrodyne is offline  
Old 19th Jan 2012, 7:20 am   #163
Synchrodyne
Nonode
 
Join Date: Jan 2009
Location: Papamoa Beach, Bay of Plenty, New Zealand
Posts: 2,943
Default Re: Valve Questions

And what makes the apparent ECC82 situation even more strange is that a diversity of applications, including multivibrators, was noted in respect of the earlier ECC40, such as in the attached snip from Philips Book IIIA. The ECC40 was one of the “miscellaneous” Rimlock valves, along with the EF40, EF42 and UF42. It did not have a direct successor in the noval series, but de facto the ECC83 had that role in respect of audio circuits, whilst the ECC82 certainly looked to be the most likely choice for pulse, etc., circuits.

I can’t find Philips Book IIIB on the web, but perhaps it has something about the ECC82. Book IIIC on TV valves includes the ECC81, but not the ECC82.

Still, the ECC40 seems to have had properties that made it attractive for some applications even in the noval era. For example the Pamphonic 2001A preamplifier used a pair of ECC40s, even though the corresponding and contemporary 1002 power amplifier included an ECC81 and an ECC83.

Cheers,
Attached Thumbnails
Click image for larger version

Name:	ECC40 Philips IIIA.jpg
Views:	518
Size:	129.4 KB
ID:	61402  
Synchrodyne is offline  
Old 5th Feb 2012, 2:12 am   #164
Synchrodyne
Nonode
 
Join Date: Jan 2009
Location: Papamoa Beach, Bay of Plenty, New Zealand
Posts: 2,943
Default Re: Valve Questions

Suppressor grid control of pentodes was discussed recently in this thread. There is an interesting item on this topic at: http://www.radiomuseum.org/forum/dua..._pentodes.html, which links to sets of curves of suppressor grid curves at: http://www.radiomuseum.org/forum/sup...surements.html. It seems – not surprisingly, I think - that at higher positive voltages on the suppressor grid, one gets kinked tetrode-like curves.

Now back to an earlier issue discussed here, namely triode vs. pentode wherein the triode is favoured where lowest noise is required. There is an apparent - but no doubt explainable - anomaly in that for low-level audio work, the pentode (EF40, EF86) was preferred over the triode (ECC40, ECC83). This is clear in the Philips/Mullard literature, that for the EF86 stating: “A high gain pentode of special design, the EF86 is particularly suitable for preamplifier and input stages, in which hum, noise and microphony must be kept to a minimum.” And Philips/Mullard example circuits and valve lines-up involving gramophone amplifiers, tape deck amplifiers, film projector amplifiers, etc., usually included an EF37A, EF40 or EF86 as the first stage.

I would in infer from this that hum was the limiting factor, and that for whatever reasons, it was easier to build a low-hum pentode than a low-hum triode, hence the effort to make the pentode as good as possible in the role of 1st low-level audio amplifier. The 1st pentode stage was often followed by triode stages. This is the opposite of say the VHF case, where the preference was to precede a pentode amplifier chain with one or two triode stages, the cascode being the outcome of a search for the optimum arrangement where two triode stages were required.

In favour of the pentode as 1st AF stage was its high gain, and the fact that at lower signal levels, potentially poorer linearity as compared with triodes was less of an issue. The high pentode gain also allowed significant input equalization to be applied via shunt feedback whilst still obtaining significant overall gain. With triode-based first stages, a cascade pair would have been needed to obtain the requisite gain in many cases, with series feedback then being used. The latter would require an unbypassed or partly unbypassed input stage cathode resistor, and this would have worsened the hum transfer problem.

But the triode hum issue could be addressed by the use of DC heating, after which the triode cascade pair – likely an ECC83 - was better than the EF86 in terms of noise and hum. Such configurations seem to have appeared in the late 1950s/early 1960s, more I think for high quality tape deck applications as replay head amplifiers than for hi-fi amplifiers. I recall reading somewhere that late in the valve era – say mid-1960s, the conventional wisdom was that in respect of magnetic cartridge input stages, and EF86 with shunt feedback would produce acceptable signal-to-noise ratio for input sensitivities down to around 4 mV (for rated output), whereas 2 to 2.5 mV could be achieved with a DC-heated ECC83 with series feedback. And by the nominal end of the valve era for audio equipment (let’s say circa 1968), 4 mV sensitivity was no longer sufficient for the latest generation magnetic cartridges.

This relativity supports the fact that good audio triodes are generally quieter than good audio pentodes, but tend to suffer from greater hum transfer. But nowhere can I find any commentary as to why that should be. Presumably the answer lies somewhere in their relative structures, interelectrode dimensions and maybe the effects of the screen grid in the pentode case.

Given that the EF86 could be used as a triode, did it then retain its low hum characteristic? Or did strapping the screen to the anode (and the suppressor to the anode or cathode) make it more like the ECC83 in respect of hum?

Improved triodes, such as the Brimar 13D7, did arrive later. This commentary was provided in Wireless World for October, 1961: “Another Brimar valve, the double triode 13D7, is designed for use in low-level low-noise preamplifiers. It has an amplification factor of 140 and can provide an average grid-hum level of only 3 µV – slightly more than low-noise A.F. pentodes – with a total valve gain of over 5,000.” Nothing was said though about how the lower hum (as compared with previous triodes) was achieved. I am not sure whether the 13D7 was later renamed as the ECC807, or whether the latter was a further improved derivative. I don’t think that the 13D7 significantly displaced the ECC83, despite its better performance.

Cheers,
Synchrodyne is offline  
Old 5th Feb 2012, 12:48 pm   #165
Leon Crampin
Octode
 
Join Date: Nov 2005
Location: Surrey, UK.
Posts: 1,869
Default Re: Valve Questions

Two features of the EF86 which contributed to its low levels of hum were a spiral wound heater (inherited from the EF37A and probably the EF40) and the positioning of the g1 connection on pin 9 which is equidistant from the two heater pins.

The 6BR7 was capable of very similar performance (and is now cheaper to buy, but unfortunately is not pin-compatible) but the ultimate accolade for low hum and noise perhaps should be awarded to the 6BS7 which is a 6BR7 with a top cap. Requiring a fully screened top cap connection to g1, this valve was clearly not an "entertainment type" and was sometimes found as the first stage Y amplifier in old Solartron 'scopes. It is claimed to be usable with an input as low as 10 microvolts.

I believe there was also a version of the ECC83 (probably in 12AX7 guise) with a spiral wound heater to minimise hum - does anyone know its type number?

Although DC heating of triodes might seem an easy option now to reduce hum, it's worth remembering that in the days of selenium rectifiers and less than wonderful high capacity electrolytics, the provision of a low ripple DC heater supply was not a straightforward matter. I have seen some professional amplifiers which use an HF supply to the heater (as was used for the exciter lamp supply in film projectors) to minimise hum.

Leon.

Last edited by Leon Crampin; 5th Feb 2012 at 12:53 pm.
Leon Crampin is offline  
Old 5th Feb 2012, 4:57 pm   #166
G8HQP Dave
Rest in Peace
 
Join Date: Sep 2008
Location: Solihull, West Midlands, UK.
Posts: 4,872
Default Re: Valve Questions

The spiral-wound heater version of 12AX7 is ECC83; the 12AX7 originally had a conventional heater. The Americans then copied this European improvement to make the 7025. Later versions of 12AX7 probably then had the improved heater too. Among modern valves, it may be pot luck whether one marked ECC83 has a spiral heater, as nowadays people regard ECC83 and 12AX7 as being identical.

The issue about pentode vs triode noise is simple for RF: the pentode has partition noise due to the screen grid. At audio frequencies it is more complicated, as the dominant noise is 1/f or flicker noise. The EF86 is quiet because it has low screen current, and possibly a special cathode with low flicker noise. It is also designed to have low microphony. There have been internet rumours that European cathodes were generally quieter than American ones, but I have no idea if this is true.

With two triodes in an envelope there may be less choice about where to put the pins to minimise hum.
G8HQP Dave is offline  
Old 7th Feb 2012, 10:46 am   #167
BBRQ229
Retired Dormant Member
 
BBRQ229's Avatar
 
Join Date: Aug 2005
Location: Devon, UK.
Posts: 71
Default Re: Valve Questions

I hope it is ok to interrupt a theme to ask a couple of basic questions about valves and valve performance - triggered by my attempts to repair an Invicta radio (I hope I am in the right thread).

One of a pair of EL33s tests with 50+ mA anode current against spec value of 36 mA. The mA / volt reading is however low. I am fairly confident that the readings are correct as the other one (EL33) tests exactly to spec on current, but a bit low on gm.

Why would Ia be high ?

and why can mA/V be low when emission is either ok or above spec value?

Is there a list anywhere of valve types and the date each came into use?

Probably naive questions, but I have many more!
BBRQ229 is offline  
Old 7th Feb 2012, 1:01 pm   #168
G8HQP Dave
Rest in Peace
 
Join Date: Sep 2008
Location: Solihull, West Midlands, UK.
Posts: 4,872
Default Re: Valve Questions

The first part is rather off-topic for this thread, so maybe the mods will move it.

The question about date of introduction is on-topic. I'm sure I saw a list somewhere, but accuracy can't always be guaranteed. There is an attempt at this here.
G8HQP Dave is offline  
Old 7th Feb 2012, 6:46 pm   #169
BBRQ229
Retired Dormant Member
 
BBRQ229's Avatar
 
Join Date: Aug 2005
Location: Devon, UK.
Posts: 71
Default Re: Valve Questions

Dave - that's really helpful - better than I expected. Thanks
BBRQ229 is offline  
Old 22nd May 2012, 3:33 am   #170
Synchrodyne
Nonode
 
Join Date: Jan 2009
Location: Papamoa Beach, Bay of Plenty, New Zealand
Posts: 2,943
Default Re: Valve Questions

Curiously this thread was started with a question about the EF89 introduction date!

Turning back to a topic noted previously but not really explored, another question is what was the rationale for using sharp-cutoff valves rather than the remote cut-off type in AGC-controlled RF and IF amplifiers, particularly in TV receivers.

It is evident that Philips intended that the EF80 be so used right from the start, when it was a founder member of the TV World Series of valves. So presumably it was designed with AGC in mind. As previously mentioned, the initial series did not include a variable-mu pentode, the EF85 being added quietly a little later. I should guess that was because some set makers wanted it, whereas Philips seemed to be happy to continue using the EF80 for controlled stages.

I had wondered whether the lower bias magnitude required to take a sharp cut-off pentode down to minimum gain was a factor, this translating to less gain required in the agc system. But the pertinent Mullard papers (noted below) suggest that there is not much in it when the rf delay, etc., is factored in.

Looking at voltage magnitudes, not signs, for the negative modulation case, Mullard assumes a PCC88 RF amplifier, requiring 5 maximum agc volts with a 3 V delay, and 3 EF80 IF stages of which two are controlled and also require 5 maximum agc volts. An additional requirement, to minimize cross-modulation, is that the RF and controlled IF stages should both reach maximum agc volts, namely 5 V, at the same time. With the resistor network required to achieve the correct delay and proportioning, this then back-calculates to a required 12 V from the agc system.

For the positive modulation case, Mullard assumes a PCC84 RF amplifier requiring 10 maximum agc volts with 3.5 V delay, an EF85 first IF stage requiring 13.5 maximum agc volts, followed by a pair of EF80s, the first of which is controlled and requires 5 maximum agc volts. The agc system is required to supply 13.5 V maximum. Mullard also provides the calculations for a PCC84 RF amplifier, in this case with 2 V delay, and an all-EF80 IF strip; here 15 V is required from the agc system to provide 10 V to the PCC84 and 5 V to the two EF80s at the same time.

So reduced agc gain requirement does not seem to be a material factor here. Mullard makes the comment: “With a negative modulation system, using frequency modulation for the sound signal, the problem of cross modulation in the receiver is not so serious as for systems with a.m. sound. Sharp cut-off pentode valves (EF80) are normally used in the i.f. amplifier.” But that does not seem to have deterred Philips, whose valved Belgian multistandard receivers, having to deal with three different positive modulation transmission systems but just one negative modulation system, seem to have used EF80s throughout the IF system, including for controlled AM sound stages.

Mullard’s mention of the PCC88 in the negative modulation case is also interesting. When the PCC84 variable-mu cascode amplifier was added to the TV World Series in 1953, it did not appear to have a sharp cutoff counterpart, so one might have inferred that Philips had decided that variable mu was indeed a better choice for the RF amplifier. But the frame-grid age ushered in both the PCC88 (sharp cutoff) and PCC89 (variable mu), so it looks as if there was a reversion to the original thinking.

So I remain rather puzzled, as to why the EF80 was used - by Philips at least - in agc-controlled applications once the EF85 was available as a standard and widely distributed valve type, particularly once it had been adopted as a key type in the FM-AM radio receiver range. Presumably there was not a material cost difference between the EF80 and EF85. Perhaps the versatility of the EF80 had something to do with it, given that it was originally noted for use as an IF amplifier, RF amplifier (Band I), frequency changer, video amplifier and sync separator. With quite a few likely to be used in a TV receiver of the 1950s, maybe it was just easier to also use them in the gain controlled stages rather than introduce another type. Or was there was a good technical reason that I am missing?

Cheers,



The Mullard TV AGC papers referred to above:

P.L. Mothersole, A.G.C. Circuits for Positive Modulation Television Receivers, Mullard Technical Communications Vol. 3 No.27, December 1957, p.214ff.

P.L. Mothersole, Automatic Gain Control Circuits in Television Receivers for Negative Modulation Systems, Journal Brit.l.R.E., May 1958, p.307ff
Synchrodyne is offline  
Old 22nd May 2012, 11:20 am   #171
G8HQP Dave
Rest in Peace
 
Join Date: Sep 2008
Location: Solihull, West Midlands, UK.
Posts: 4,872
Default Re: Valve Questions

I'm not sure if this answers the question, but when running at full gain a straight pentode is likely to slightly more linear than the corresponding variable mu pentode. As soon as AGC is applied, the variable mu valve wins because its performance does not deteriorate by nearly so much. Unfortunately Philips don't give cross-modulation curves for their straight valves so it is difficult to compare. A straight valve will also have slightly more gain and slightly less noise.

For TV reception you could assume, at least early on, that you had just one station with a fairly well-defined signal strength. If it was too weak you needed a bigger aerial. If it was too strong you could put an attenuator in the cable. The AGC then made the final adjustments. Later, when Band 3 was used for ITV, it was quite possible that the two stations would have different signal strengths so better AGC was needed. Remember that the wide bandwidth of TV means that the total dynamic range between 'too noisy' and 'overloading' is narrower than for sound radio, so AGC has less work to do for TV than radio.

Something which puzzles me is why many Heathkit receivers applied AGC to a straight valve in the IF amplifier.
G8HQP Dave is offline  
Old 1st Jun 2012, 10:34 am   #172
Synchrodyne
Nonode
 
Join Date: Jan 2009
Location: Papamoa Beach, Bay of Plenty, New Zealand
Posts: 2,943
Default Re: Valve Questions

Hi Dave: Thanks for that, which does help. I think that it is pointing to the fact that Philips (and perhaps others) did it that way (AGC on sharp cutoff valves) at least in part simply because it could be done that way for TV receivers, whereas it probably would not have been a very good idea with radio (LF/MF/HF AM) receivers. Somehow I don’t think that an EF80 final IF stage biased to the bottom of the curve because of a strong signal would have been very helpful in avoiding modulation rise in an AM receiver.

TV AGC systems, as well as operating over a narrower range typically seem to have reasonable gain when measured against the video signal level right after demodulation. Some of that gain comes from the video amplifier, and some (maybe the majority?) from the gating valve. On the other hand, such gain in AM receivers was not common; I suppose that the Chapman AM tuners were amongst the best known examples with of amplified AGC in British practice post-WWII.

Relating TV agc systems to valves used for the purpose, European practice appears to have been to use types from the “standard” inventory, such as EF80 and ECC82, whereas the American valve makers developed specialized types, such as the 6GY6 and 6BU8.

A derivative topic, in the sense that we have been talking about the use of the EF80 in situations where at first glance an EF85 might seem more logical, is the existence of pentode “pairs”, one sharp cutoff and the other remote cutoff, but otherwise generally similar. The EF80 and EF85 fall into this category, as do their frame grid successors EF184 and EF183. And the same looks to be true of their predecessors EF42 and EF43. Another pair is the 6BH6 and 6BJ6. On the other hand, the EF41, and its EF89 successor, both remote cutoff, seem not to have sharp cutoff counterparts, mostly I should guess because such were of little utility for MF/HF applications. The 6BA6 (EF93) and 6AU6 (EF94) I am not so sure about; they could well have been unrelated, with the 6AU6 initially intended more for TV IF applications, whereas the 6BA6 was “the” AM RF and IF valve in the initial American domestic application B7G range.

Cheers,
Synchrodyne is offline  
Old 1st Jul 2012, 1:01 am   #173
Synchrodyne
Nonode
 
Join Date: Jan 2009
Location: Papamoa Beach, Bay of Plenty, New Zealand
Posts: 2,943
Default Re: Valve Questions

As well as the above-mentioned pentode “pairs”, one sharp cut-off, one remote cut-off, there appears to have been at least one similar cascode double triode pair, namely the ECC88 and ECC89.

Talking of cascode circuits and valves, I recently came across an article (*) in which the original Wallman circuit (shunt cascode) is referred to as a cascode, but the series cascode version is referred to as a “driven grounded grid”. The latter is attributed to RCA in early 1951, using the 6BQ7 valve. Likely it took a while for the idea to reach the European valve makers, which might help explain why the initial Philips/Mullard World Series of TV valves did not include a cascode double triode, although RF amplifier circuits using the ECC81 as a shunt cascode amplifier were suggested.

The article includes the circuit of the original RCA turret tuner using the 6BQ7. It had a double-triode mixer-oscillator, unidentified, but given that it had a common cathode, probably a 6J6. The mixer was neutralized, necessary I think for the 45.75 MHz IF, of which this must have been an early example. It is always risky to extrapolate from a single datapoint – although one does have the luxury of being able to head in any direction – but it would seem that in US practice, the triode pentode TV mixer oscillator valve came after the double triode series cascode valve. Perhaps once the latter was available, it was seen to provide sufficient RF gain to allow the use of a pentode mixer at Band III frequencies without an unacceptable noise penalty. And although it was not otherwise desirable at these frequencies, the pentode mixer at a stroke solved the Band I Miller feedback problem when highish IFs were nudging the lower Band I channels. This situation was an oddity in the whole pentode vs. triode question, I think. The Wallman cascode was the answer to the question of how to best deploy two triode stages ahead of a pentode amplifier when one stage was not enough to discount the pentode noise. But I am not sure that it its use in front of (an even noisier) pentode mixer was originally envisaged.

By way of references, the article confirms that Philips introduced the PCC84 and PCF80 in 1953. It also notes that the ECC84 preceded the PCC84.

(*) The article is a January, 1955 British IRE paper, A Survey of Tuner Designs for Multi-Channel
Television Reception, by D. J. Fewings and S. L. Fife, both of English Electric.

Some references from the paper (which I haven’t seen) are:

R. M. Cohen, Use of new low-noise twin triode in television tuners, R.C.A. Review, 12, pp. 3-25, March 1951.

The PCC 84 double triode, Electronic Applications Bulletin (Philips), 14, No. 8/9, 1953.
The PCF 80 triode-pentode, Electronic Applications Bulletin (Philips), 15, No. 1/2, 1953.

Cheers,
Synchrodyne is offline  
Old 13th Aug 2012, 10:12 am   #174
Synchrodyne
Nonode
 
Join Date: Jan 2009
Location: Papamoa Beach, Bay of Plenty, New Zealand
Posts: 2,943
Default Re: Valve Questions

Quote:
Originally Posted by G8HQP Dave View Post
I think I saw it argued somewhere that there is a continuum between pure pentode and pure beam tetrode construction. It all depends on what produces the low potential between the screen grid and anode: electrodes in the electron stream connected to the cathode (pentode), or bunching of electrons due to beaming partly caused by electrodes just outside the electron stream (also connected to the cathode - beam tetrode). I don't see any reason why you can't have aligned grids (like a beam valve) in a pentode, but I don't know if this was ever done.
Here is something I missed previously, from the April, 1958 Mullard Technical Communications article “Transformerless Audio Output Stages”. In terms of suitable valves, that is those that could operate with low DC line voltages, it was said:

“Two Mullard pentode types, the EL86 and UL84, are particularly suitable. In each type (they are identical but for heater conditions) µg1g2 is 8, and an anode current of 43 mA is available when anode and screen grid are at only 100V. Screen grid current and dissipation are minimised by the use of shadowed grids. Both types have sufficiently high cathode-to-heater voltage rating.”

I assume that shadowed grids are tantamount to aligned grids, with the screen lying in the shadow (as viewed from the cathode) of the control grid. Perhaps the suppressor was in the shadow, too?

Now, if the EL86 and UL84 had shadowed grids, one might reasonably expect that the EL84 had the same. But I do not find it mentioned in any of the EL84 literature that I have looked at.

Be that as it may, I guess the answer to the question is yes, there were pentodes with aligned grids.

Cheers,
Synchrodyne is offline  
Old 13th Aug 2012, 11:10 am   #175
G8HQP Dave
Rest in Peace
 
Join Date: Sep 2008
Location: Solihull, West Midlands, UK.
Posts: 4,872
Default Re: Valve Questions

I think the EL86 and its cousins came out a bit later than the EL84? Comparing the datasheet currents for normal quiescent Class A:
EL84 (at 250V a, g2): Ia 48mA, Ig2 5.5mA - ratio 11.5%
EL86 (at 200V a, g2): Ia 65mA, Ig2 3.2mA - ratio 4.9%

I think that means the EL84 could have partially aligned grids, as a typical RF pentode might have a ratio nearer 20%. Note that later RF pentodes had smaller g2 currents to reduce partition noise, while an output pentode benefits from smaller g2 currents for reasons of heat dissipation.

The suppressor grid can't be aligned because it necessarily has a much wider pitch, to stop it getting in the way of the electron stream. A possible exception would be dual-control pentodes such as 6F33 or 6AS6 but I don't know if these used aligned grids or not.
G8HQP Dave is offline  
Old 7th Oct 2012, 9:14 am   #176
Synchrodyne
Nonode
 
Join Date: Jan 2009
Location: Papamoa Beach, Bay of Plenty, New Zealand
Posts: 2,943
Default Re: Valve Questions

Another output pentode reportedly with shadowed grids was the PCL85. It was also of the frame grid type. The PCL85 also represented a departure in that its 8 W pentode was intended solely for – and presumably optimized for – TV field output applications. Previous members in the series, namely the PCL83, PCL82 and ECL80, had pentodes that were intended both for field output and audio output applications. The same duality also applied to the PL84 and PL82 pentodes, and de facto for the UL46, which was allegedly just a more robust version of the UL41. The later PL508 was designated as a field output valve. Of course that did not stop alternative deployments.

The UL46 was also designated as a video output valve. But here the functional separation came early, as the PL83 was the designated video output valve in the following noval series, whereas this function was not mentioned in respect of the PL82, which looks to have been very similar to the UL46. This specialization was maintained with the PCL84, PFL200 and PL802.

As a sidebar here, the PCL84, PFL200 and PL802 respectively had 4, 5 and 6 W anode dissipations. Against that background, it seems possible that the 9 W of the PL83 was more than the application really needed. But perhaps it was the case that the Philips/Mullard available and relatively recent high slope miniature pentode happened to be of 9 W capacity, and it was preferred simply to “tweak” it somewhat to improve its capabilities as a video output amplifier than design a new valve.

Returning to the PCL series, the PCL86 was also monofunctional, with its pentode evidently intended for audio output applications only, and with a high µ triode that looked like half of an ECC83. It also represented a return to the 9 W anode dissipation level of the UL41 and its brethren.

Cheers,
Synchrodyne is offline  
Old 27th Oct 2012, 9:28 am   #177
Synchrodyne
Nonode
 
Join Date: Jan 2009
Location: Papamoa Beach, Bay of Plenty, New Zealand
Posts: 2,943
Default Re: Valve Questions

Earlier in this thread, posts #79 through 81, we discussed the possibility of using a cascode as a low-noise mixer, although no such actual examples were forthcoming. Recently though, I have come across the attached circuit, taken from the BBC R&D Report “Colour Television: The Adaptation of the N.T.S.C. System to U.K. Standards; Part 4: Coding and Decoding”, available at: http://www.bbc.co.uk/rd/publications..._1958_15.shtml.

Cascode synchronous demodulators were used in both phase detectors in the quadricorrelator colour subcarrier regeneration circuit of the decoder. The circuit appears to be a multiplicative mixer with a pentode input stage in shunt cascode with a grounded grid triode, with burst (“signal”) applied to the pentode grid and subcarrier local oscillator applied to the triode grid.

But I think that the triode grid would not be grounded at subcarrier frequency, which is also burst (signal) frequency, so that looked at from the signal input, maybe it is not a full cascode circuit? It has me puzzled. Is it something that will work only for the case that signal and local oscillator frequencies are the same, or would it be more generally applicable to cases where signal and oscillator frequencies were quite different? In this case the triode cathode is sitting on a parallel tuned circuit that I imagine would be at burst/subcarrier frequency. But for a generalized case, one might want the cathode bypassed for oscillator frequency only, but otherwise not bypassed for RF over the signal frequency range.

Cheers,
Attached Thumbnails
Click image for larger version

Name:	Cascode Synchronous Demodulator.jpg
Views:	2053
Size:	31.3 KB
ID:	71899  
Synchrodyne is offline  
Old 27th Oct 2012, 11:43 am   #178
G8HQP Dave
Rest in Peace
 
Join Date: Sep 2008
Location: Solihull, West Midlands, UK.
Posts: 4,872
Default Re: Valve Questions

It is not clear to me how that circuit can work. The pentode sets the signal current flowing in the triode, irrespective of the voltage on the triode grid. Unless the triode grid voltage changes so much that the pentode is forced down into its saturation region, where the anode impedance falls as cathode current is diverted to g2 because the anode voltage is so low. If so, this may be fine for a frequency control loop but not too good for a receiver mixer?
G8HQP Dave is offline  
Old 22nd Dec 2012, 5:30 am   #179
Synchrodyne
Nonode
 
Join Date: Jan 2009
Location: Papamoa Beach, Bay of Plenty, New Zealand
Posts: 2,943
Default Re: Valve Questions

Possibly the triode is swung down to cutoff? The subcarrier input level is probably high enough to do this. Then it would seem to be a case of switching the pentode on and off. I guess we’d need to find the original article on this circuit to find out what was in the designer’s mind and how it actually worked.

Talking of cascode circuits, I have since found this previously mentioned article:

R. M. Cohen, Use of new low-noise twin triode in television tuners, R.C.A. Review, 12, pp. 3-25, March 1951.

This covers the development and application of the 6BQ7 valve, which I think was the first double triode designed specifically for cascode amplifier use. The author makes the following early comment: “The low-noise features of triodes have been recognized generally, but stability difficulties and other problems associated with the use of triode tubes in conventional circuits have limited their extensive application in television tuners. Consequently pentodes have been used in the radio-frequency stages of most receivers despite their higher noise. The development of the 6BQ7 and its associated circuits offers the possibility of a change in this situation.” With hindsight one can see that that prediction was slightly understated, as the cascode became the standard TV VHF tuner RF amplifier through the 1950s.

Cohen also refers to the original Wallman cascode amplifier, suggesting that it was difficult to use for other than single-frequency amplifiers. Thus was developed what was then called the direct-coupled driven-grounded-grid circuit, which later came to be known as the series cascode circuit. Each triode in the 6BQ7 was identical, and the inter-unit screen had its own pinout, but keeping the input leads short led to preferences as to which was deployed in which role. It was inherently a sharp cutoff valve, but the series connection conferred remote cutoff characteristics.

The Philips/Mullard PCC84 of 1953 seemed to follow the same pattern of being a sharp cutoff valve. In this case the inter-unit screen was connected to the grid of triode 2, which thus more-or-less necessarily became the grounded grid output stage, whilst the pinout so freed up was used for a second cathode connection to triode 1, thus following the precedent of the EF80 and EF85 pentodes. Prior to the advent of the PCC84, Philips had suggested using the ECC81 as a shunt-cascode TV RF amplifier.

Later though, the PCC88 and PCC89 formed sharp cutoff and remote cutoff counterparts. I should guess that the PCC89 followed conventional remote cutoff valve practice and had a graduated grid pitch, at least for the input grounded cathode unit. Presumably in the case of the PCC88, means were found to prevent or minimize the widening of the grid base caused by series connection.

Cheers,
Synchrodyne is offline  
Old 22nd Dec 2012, 12:09 pm   #180
turretslug
Dekatron
 
turretslug's Avatar
 
Join Date: Nov 2011
Location: Surrey, UK.
Posts: 4,385
Default Re: Valve Questions

Re. multiple cathode connections, it long struck me as odd that the EF50, an otherwise advanced design of the time (and with the relative kuxury of a 9-pin base and (I think usually) a metal spigot) only had one pin assigned for the cathode. An approximate US counterpart and contemporary, the 6SH7 (8 pins and plastic spigot) had two cathode connected pins, it seems that low cathode connection inductance at high HF/low VHF loomed sufficiently large in some designers as to make it a high priority by the mid-thirties.

The EF54 went for multiple cathode connections with a vengeance, though...

Last edited by turretslug; 22nd Dec 2012 at 12:30 pm. Reason: Supplement.
turretslug is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 2:51 pm.


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 - 2024, vBulletin Solutions, Inc.
Copyright ©2002 - 2023, Paul Stenning.