Leak Troughline 1. Are they really this insensitive?

[paulsherwin]

Originally Posted by Sideband

Replacing the lot may cure it BUT it's poking and hoping and you won't learn anything! The two EF80's are probably IF amps but with three ECF80's these could be used as IF amps as well. I'm going to make a guess and say the two EF80's are RF amp and self oscillating mixer (or maybe just a mixer with one of the ECF80 triodes used as oscillator. If that is the case then the EF80's need to be in top-notch condition so I would at least try those first. If you have a meter, check the HT supply as the EZ80 may be tired causing low HT which will also cause poor sensitivity. Perhaps someone with a Troughline 1 circuit can clarify which valves do what (I only have the Troughline II circuit).

EF80s are notorious for soldiering on indefinitely with low but functional emission. Given that they are pretty much the cheapest NOS valves around, it would certainly make sense to replace them, especially if the current ones have a lot of miles on the clock.

[Radio Wrangler]

That may be a matter of them being mostly used in TV IFs where there is plenty of AGC range and gain to spare. Band II FM is at a significantly higher frequency, incoming signal levels are lower than for TV service, so the EF80 might be being relied on rather a lot more.

Withe the pentode RF and mixer, this is very much an early generation FM tuner. The pentodes were dropped very rapidly once VHF triodes appeared on the scene.

David

[Sideband]

Yes that is the case. EF80's do soldier on for many years in TV's. However when they started to be used in FM tuner front-ends, they were being pushed to their limits at the higher frequencies (one of the reasons why these early tuners only tune to about 100Mhz.....OK at the time. However it does mean that after years of use, they will struggle at the higher frequencies. I would certainly try these two first.

[unclemanly]

Many thanks for the advice. I must say, having previously dealt with valves almost entirely in the context of HiFi amplifiers, I was astonished to to see that I could revalve the entire Troughline with NOS parts for less than the cost of a single NOS audio valve. I'll start off with the EF80s though.

[Synchrodyne]

Re the pentode vs. triode issue in respect of FM receiver and tuner RF amplifiers, the cascode was the logical step-up from the pentode, and the development of the requisite double triode valves and circuitry for VHF TV purposes from 1951 made it quite easy for the equipment makers to adopt this for higher performance FM equipment. That happened in the USA, but the UK makers seemed to have been more reluctant, even though the cascode was well-established, with European valves available for it, by 1954. Leak was one of the few who did make the move, although not right away. The Troughline II of 1959 had a 3-gang front end with ECC84 cascode RF amplifier and one triode of an ECC85 as mixer. The original Troughline had been 2-gang, with an EF80 RF amplifier and the pentode side of an ECF80 as mixer. So the Troughline II should have been at least a little better in terms of lower noise and improved front-end selectivity. For the Troughline Stereo, Leak moved up to the frame grid ECC88 for the cascode, ostensibly for higher gain.

Beneath the cascode case though, the empirical evidence is that the choice between a pentode and a single triode as FM RF amplifier does not seem to have been clear cut, with the pentode being remarkably persistent, and into the 1960s. The single-valve double triode FM front end was a German invention, initially with the ECC85. I suspect that sometimes, perhaps often, it was used as much for economy as for the lower noise level possible from the triode RF amplifier. Given the lowish gain available from a single triode RF amplifier, a triode mixer, for lower noise, was indicated, hence the double triode and not, say, a triode-pentode. Aiding the economy side was that complete single-valve FM front ends were readily available from third party makers, often the same as who offered complete TV tuners assemblies. Almost always the single-valve front ends were 2-gang. This approach, well-established by the time that FM arrived in the UK, evidently appealed to many of the UK setmakers.

An interesting example is that of Pye, who in 1955, had adopted a two-tiered approach with its domestic FM receivers. The lower-priced FenMan I had an ECC85 front end. The more expensive and more elaborate FenMan II had a two-valve, 3-gang front end, with an EF80 RF amplifier and an ECF80 frequency. One may reasonably deduce that the putative noise disadvantage of the pentode was not considered to be serious enough to detract from the overall higher performance positioning of the FenMan II as compared with the FenMan I, and that the ECC85 was used in the later primarily for economy reasons.

Another example was Dynatron. It had used a pentode RF amplifier (Z719/EF80) in its FM1 and FM2 tuners, both notably used by the BBC for rebroadcasting and monitoring purposes. But it had changed to an ECC85-based front end for its T10 FM-AM tuner. However, that was soon superseded by the T10A, in which a two-valve (EF80 and ECF80), 3-gang front end was used. Evidently Dynatron had perceived that the two-valve front end was overall better, even if the noise performance was not quite as good.

The single-valve front end, as a third party unit, was quite late in getting to the USA – late 1950s -where, below the cascode, the pentode had been the dominant type of FM RF amplifier in domestic receivers. Some established pentode users, such as Zenith, eventually made the change, but others, like GE, did not. There were also some further developments of the single-valve itself. These included the use of a frame grid triode for the RF amplifier side (e.g. 6JK8), the double tetrode (e.g. 6C9/17C9), and the triple triode, used variously in RF/autodyne MX/AFC, RF/MX/Oscillator, and MX/Oscillator/AFC combinations. Then there was the 6/12JN8 pentode triode, in which the pentode was the RF amplifier, and the triode was the autodyne mixer, for those who definitely preferred the pentode RF amplifier. For example, GE used the 12JN8 in its T2000 stereo table receiver of the early 1960s, whereas Zenith used the 6JK8 in its comparable MJ1035. Both of these makers I suspect had a clue or two about what was required for good reception of the then-new Zenith-GE FM stereo system. And there was not a unanimous choice in respect of the RF amplifier.

On the basis of the empirical evidence then, I should not be too worried about the use of pentode RF amplifiers in FM receiving equipment from the valve era. Sure, a cascode could and should be a bit better, but such were not abundant in British equipment. And the Leak Troughline matched its contemporaries of the period, who were mostly in the 2-gang/pentode camp. These included Lowther, Quad and RCA (UK). Quad used a circuit originally developed by G. Horn c.1952 and apparently derived somewhat from the Brimar playbook. RCA’s New Orthophonic FM tuner had an EF95 (6AK5) RF amplifier, so it chose a very low noise pentode, even though it was the first developer of the series cascode (and its associated valve) for VHF RF work (although the cascode per se had prior origins.) At a lower level vis-à-vis FM tuners, such as Armstrong, one did find the ECC85 being used.

I’d say that on this battlefield, the pentode/triode issue played out as a small-margin points decision, not as a Bosworth-style clean sweep for the triode, unlike say the TV Band III case.


Cheers,

[Radio Wrangler]

Thanks for that. It played out somewhat before my time. Pentodes are inevitably noisier than triodes can be, but then triodes need either neutralisation or some other help in order to stop Cag spoiling the gain. Neutralisation means having a fairly nasty adjustment to get the best gain and still leave the thing stable. The cascode arrangement is a good solution to this problem but needs a device with good heater-cathode insulation.

Another way of dodging neutralisation would be to use a grounded-grid single triode. This gets the noise advantage of the triode and allows more gain without neutralisation, if the grid is an effective enough screen. Usually, high performance triodes intended for grounded grid use feature multiple grid pins, a trend which ended up with disc-seal triodes (too expensive by far for domestic receivers) But for a grounded grid RF amp, the input impednce is much lower than the grounded cathode configuration, so a different arrangement of tapping the RF tuned circuit would be needed.

It's interesting to read how history played this out, but this is one area where the transistor really did shine once FETs had come along.

David

[Nuvistor]

Accepting the ECC85/UCC85 were probably introduced for economy, they worked surprisingly well in the environment of the time. In the UK it was mono only with three powerful stations mostly 2.2Mhz apart. Fitted into table radios and radiograms the the wire aerial supplied often worked well. The area I covered was about 35 miles from Holme Moss and rarely was an outside aerial required.
When stereo came along the radios I dealt with were transistor, bipolar at first. An outside aerial was really required with the stereo signal, not the transistors fault just the nature of the system. The ECC85 type would probably not have proved as useful in that case.

However it worked well in its time, it may have been designed for economic reasons but it did the job.

[John_BS]

I would suggest a three way investigation:
1. Swap the two EF80's and see if there's any change.
2. Introduce a 10dB attenuator in the feed from the loft antenna: I've known tuners with no or little RF selectivity ahead of the first active device to appear "deaf" due to blocking / intermods.
3. If neither the above results in any improvement, and assuming your DC voltages are about right (?), it may be that the LO/RF stage tracking has been misaligned or drifted. That might account for only one station / frequency performing reasonably well.
I can check the latter for you if you don't mind posting.
John

[unclemanly]

Thats a very kind offer John, thanks.
Following earlier advice from other members, I've got a couple of EF80s on order. If they don't do the trick and aerial attenuation doesn't either, I'll PM you.

[mark2collection]

It's worth checking the condition of the valve pins, valve sockets and the aerial socket/plug.

The MKI Troughline I have, had quite badly tarnished valve/socket contacts, add to the electrically leaky wax paper capacitors and a heavily tarnished aerial socket, it's a wonder it worked at all.

Contact cleaner was used on the sockets/aerial plug, and a rubberised 'emery' block was used carefully, on the valve pins. Wax paper capacitors were replaced with modern equivalents.

Also worth checking to see if your tuner has the de-emphasis capacitor fitted. Cannot remember the values, it'll be written on the schematic. My tuner never had this modification though does now, and after all this work, the tuner is an excellent performer.

I also fitted a cheeky/reversible modification, where the audio out socket can be used as an MPX output. Connecting this to my home-made stereo decoder, you'd never know you're listening to a 50's tuner(!)

Checking the resistors is worth a shot, that said, all the one's in my tuner were within spec.

I also have a MKII & Troughline Stereo, both had wildly out of spec resistors and an electrolytic in the discriminator circuit which was resistive.

Stereofetics are fussy if the PSU is not setup correctly or the lamp has failed (since this forms a PSU load). Phasy stereo, distortion, poor station 'pick-up' and drifting have been things I've seen. The internal supply rails from the regulator PCB have to be in spec.

Hope this helps.

Mark

[unclemanly]

Well, replacement EF80s, ECF80s and EZ80 arrived. I replaced the original valves one at a time, testing the tuner after each substitution. No single change made a huge difference but cumulatively the new valves have improved reception. Its now fine across the bottom half of the frequency range (88-94MHz) and acceptable, if not great, across the top half (94-100MHz).

I'm going to leave it at that. Tinkering around with stuff I don't really understand is unlikely to be productive.

Thanks to everybody who offered advice and help in this thread.

Rob

fine across the bottom half of the frequency range (88-94MHz) and acceptable, if not great, across the top half (94-100MHz).

Tinkering around with stuff I don't really understand is unlikely to be productive

A proper engineering solution, all it has to be is good enough.