|
Vintage Audio (record players, hi-fi etc) Amplifiers, speakers, gramophones and other audio equipment. |
|
Thread Tools |
4th Aug 2021, 8:12 am | #1 |
Dekatron
Join Date: Sep 2015
Location: Southwold, Suffolk, UK.
Posts: 8,302
|
The (sometimes controversial) ECL86
I went with my Father to the launch of the ECL86 in the early Summer of 1962 at Mullard House, Torrington place, London WC1. Their USP for this new valve was that it combined the ECC83 with the EL84 as an effective triode-pentode amplifier within one envelope. At the end of the presentation I (probably annoyingly) asked why they did not demonstrate an ECL86 amp with that of an (1/2) ECC83 and an EL84? I was rather firmly advised that their electrical characteristics were the same. Even so, I would have liked to have actually listened to a comparisom.....I wonder what other Members might have thought or what questions they might have asked?
__________________
Edward. |
4th Aug 2021, 9:43 am | #2 |
Tetrode
Join Date: Jun 2021
Location: Leatherhead, Surrey, UK.
Posts: 60
|
Re: The (sometimes controversial) ECL86
They will sound the same.
Any differences will be mostly down to any differences in output transformers. |
4th Aug 2021, 9:56 am | #3 |
Heptode
Join Date: May 2017
Location: Konongo, Ghana
Posts: 510
|
Re: The (sometimes controversial) ECL86
The maximum anode dissipation of the EL84 is 12 Watt, while that of the L-part in the ECL86 is 9 Watt. Looking at it like that, the L-part of the ECL86 is much closer to the EL41 than to the EL84.
Besides that, the L-part in the ECL86 is a beam power tube (with beam forming plates), while the EL84 (like the EL41) is a true power pentode (with suppressor grid). I have built a guitar amplifier with 2 x ECL86, see: https://www.vintage-radio.net/forum/...d.php?t=147961
__________________
Robert Last edited by Robert Gribnau; 4th Aug 2021 at 10:08 am. Reason: Addition |
4th Aug 2021, 10:20 am | #4 |
Nonode
Join Date: Nov 2016
Location: Aberaeron, Ceredigion, Wales, UK.
Posts: 2,869
|
Re: The (sometimes controversial) ECL86
I think your question Edward was very sensible, I’m surprised they did not have a demonstration.
John |
4th Aug 2021, 10:53 am | #5 |
Nonode
Join Date: May 2007
Location: Preston, Lancashire, UK.
Posts: 2,510
|
Re: The (sometimes controversial) ECL86
Did they demonstrate anything?
|
4th Aug 2021, 11:21 am | #6 |
Dekatron
Join Date: Sep 2007
Location: Oxfordshire, UK.
Posts: 4,310
|
Re: The (sometimes controversial) ECL86
I would have been 4 years old in 1962, so I can only speculate, but I wonder what the general attitude to 'listening tests' was back then ?
Perhaps for speakers it wasn't very different to the way it is now - basically the job of a stereo hifi system is to fool the brain into thinking that the sound, which is actually coming from two smallish sources, is instead coming from a distributed group of musicians (or sometimes just one) located at points in space other than where the speakers are. Success at that task can be very hard to predict from measurements alone and things like a speaker output's directionality, interaction with the room it's in, small timing differences between different drivers, perhaps the impact of crossovers on damping and transient handling etc might all matter, making a listening test worthwhile. But for amplifiers the measurements really ought to be good indicators of 'sound', neglecting any psychological effect caused simply by knowing that this is an amp you're expecting to like, or expecting not to. The wisdom at the high-end of the market seemed to be that once you'd satisfied the hifi criteria - adequate power, output impedance less than 10% of speaker impedance, frequency response flat from 20Hz-20kHz, THD at a small fraction of a percent, hum and noise unobtrusive - then listeners wouldn't be able to tell amps of different design or manufacturer apart. Push-pull amps in this category, based on a pair of ECL86s and an EF86 say, would measure pretty much like the Mullard 5-10 which was based on EL84s, an ECC83 and an EF86 of course, and so would sound the same. At the sub-hifi end of the market, say with a single output valve being driven by a single small-signal stage, or with a push-pull PA circuit optimised for maximum power over limited bandwidth, performance inevitably would be worse. I might expect these amps to sound different from one another. But now I'd have thought the sound would be dominated by things like the circuit architecture (single-ended versus push-pull, lots of feedback versus little) or the amount spent on the output transformer (in the end, size can matter) rather than on the particular valve. Just my two penn'orth of course. I don't know if I'd have asked in public, and maybe they already answered the question in their presentation, but in the case of the ECL86 I'd have wanted to know how durable it was. The EL84 was already known as being a lot of valve in a little bottle and Leak's experience with it showed that there was essentially no headroom for pushing it beyond its ratings. Adding another element in almost the same sized bottle must have raised the thermal load even more, and there would have been more metalwork that would have needed outgassing at pump-out. Cheers, GJ
__________________
http://www.ampregen.com |
4th Aug 2021, 1:47 pm | #7 |
Tetrode
Join Date: Jun 2021
Location: Leatherhead, Surrey, UK.
Posts: 60
|
Re: The (sometimes controversial) ECL86
Agree with Grim there.
Interestingly I've just run up an EL84 and ECL86 together on the curve tracer and whilst the '84 shows as expected more deflection being bigger, the actual curve shapes were pretty much identical, so given the same transformer and no neg feedback they will sound virtually identical. Ps, hi Graeme. |
4th Aug 2021, 2:50 pm | #8 |
Dekatron
Join Date: Nov 2009
Location: Lincolnshire, UK.
Posts: 4,985
|
Re: The (sometimes controversial) ECL86
A good valve, I think, until they get old and well used, particularly with bad coupling capacitors, which tend to finish them off and give them grid emission all of their own. I note that there's a radiogram thread running at the moment, which is displaying typical characteristics of a pair of bad ECL86s, although in this case the fault may well turn out to be a failing HT supply, but I wouldn't like to put a large bet on it!
|
4th Aug 2021, 4:27 pm | #9 |
Octode
Join Date: Dec 2018
Location: Llandeilo, West Wales, UK.
Posts: 1,092
|
Re: The (sometimes controversial) ECL86
I'm not sure why they would 'demonstrate' it?
Really depends on what type of launch release it was I suppose. Being 1962 the obvious question would be how long will it be supported with the new fangled transistors coming along so fast after. That is the only real problem with the ECL86, it had a comparatively short production run and when it does wear out and no one makes a current replacement. You have to use NOS or used examples so pay the price...
__________________
Never Leave Well Enough Alone... |
4th Aug 2021, 5:24 pm | #10 |
Octode
Join Date: Aug 2006
Location: Stoke-on-Trent, Staffordshire, UK.
Posts: 1,082
|
Re: The (sometimes controversial) ECL86
some very rated amplifiers used these Rogers used loads of them in the hg 88 and cadet series
These have been highly regarded in the 8-10 watt class Armstrong also used them and to be fair they did sound good Trev |
4th Aug 2021, 5:37 pm | #11 |
Octode
Join Date: May 2009
Location: Stevenage, Herts. UK.
Posts: 1,515
|
Re: The (sometimes controversial) ECL86
I had to change one in my Heathkit S-33H a couple of years back. I think maybe Mullard pushed the envelope too much. It's noticeable that there's no modern production, whereas there is a modern ECL82.
|
4th Aug 2021, 8:37 pm | #12 |
Dekatron
Join Date: Sep 2015
Location: Southwold, Suffolk, UK.
Posts: 8,302
|
Re: The (sometimes controversial) ECL86
Thanks for all the many and varied responses here. Indeed, and disappiontingly, I did not know that there was now no new production of ECL86s. I wonder what their "future" now holds?
__________________
Edward. |
5th Aug 2021, 8:23 am | #13 |
Heptode
Join Date: May 2017
Location: Konongo, Ghana
Posts: 510
|
Re: The (sometimes controversial) ECL86
Although there is no new production of the PCL86 either, there are still many of them around for reasonable prices. With a voltage doubler for the filament supply (helped by the somewhat higher mains voltages we have nowadays) they could serve as a replacement.
__________________
Robert |
6th Aug 2021, 3:45 am | #14 |
Nonode
Join Date: Jan 2009
Location: Papamoa Beach, Bay of Plenty, New Zealand
Posts: 2,943
|
Re: The (sometimes controversial) ECL86
I think that GJ pretty much nailed it in post#6.
Back in those days, a competent hi-fi amplifier designer would take pains to ensure that the characteristics (and normal production variations thereof) of the chosen valves did not materially interfere with the objective of producing an essentially transparent gain device. I doubt that the valve makers wanted their output valves to be known for imparting a particular kind of sound. Rather they would probably have preferred that they were seen as relatively benign devices that did not make more difficult the amplifier designer’s task. Mullard’s primary sales target would have been the equipment makers who built amplifiers and other items, such as radio receivers, that included amplifiers, rather than the end users of such equipment. I doubt that this audience would have been asking “how does the ECL86 sound", given that their own amplifier designs were the major controlling factor there. Rather, they would have been more interested in its utility, price, durability and so on. Stating – with supporting data - that the triode was one half of an ECC83, and that the pentode mirrored the characteristics (up to the 9 W anode dissipation point) of the EL84 would I think have been key information, creating the impression that the ECL86 provided an easy pathway to valve envelope economy in many applications where the EL84 was already used. Or, in other words, Mullard was selling the valve, not the amplifiers in which it would be used. An actual sound demonstration might have been nice to have, but it would have been much more a comparison of two amplifiers than of the valves themselves, and so it would not have provided the desired QED. That had been done by showing the similarity of characteristics. Effectively, Mullard was saying to designers – you know how to get the results you want when using the EL84; you can directly transfer that knowledge to using the ECL86. Had I been there, I am not sure what question I might have asked. I am not sure that I can recall with any accuracy how much I actually knew back then, although I am fairly sure that I was aware of the EL84. With today’s knowledge, I think my question would have been what factors caused the anode dissipation of the pentode to be limited to 9 W, and were the electrodes sized accordingly, or were they the same as for the EL84. Cheers, |
7th Aug 2021, 9:05 am | #15 |
Octode
Join Date: May 2009
Location: Stevenage, Herts. UK.
Posts: 1,515
|
Re: The (sometimes controversial) ECL86
This 1962 date must've been the public release, release to favoured OEMs must've been earlier - the Grundig TK14 uses one and was apparently released in August 1961. In this simple application I don't really see any advantage over an ECL82.
|
7th Aug 2021, 12:20 pm | #16 |
Dekatron
Join Date: Sep 2015
Location: Southwold, Suffolk, UK.
Posts: 8,302
|
Re: The (sometimes controversial) ECL86
Don't fully agree regarding the role of an ECL82 (released 1956) against an ECL86. There is sufficient triode stage gain, and given the high-slope pentode/tetrode, to insert a Baxendall type tone control circuit in some applications. Philips and Defiant used this topology in their portable record players.
__________________
Edward. |
7th Aug 2021, 2:16 pm | #17 |
Octode
Join Date: Nov 2005
Location: Surrey, UK.
Posts: 1,869
|
Re: The (sometimes controversial) ECL86
Lower hum levels in the '86. Compare the screening of the triodes in each type.
Leon. |
7th Aug 2021, 6:40 pm | #18 |
Octode
Join Date: Nov 2005
Location: Surrey, UK.
Posts: 1,869
|
Re: The (sometimes controversial) ECL86
A few other observations for ECL/PCL86 users:
The PCL and ECL86 are identical apart from the heater. The Brimar data book No. 9 helpfully gives curves for both types - the same. Philips, in their usual furtive manner give the differing characteristics at differing voltages. They also describe the output device as a "pentode" (as usual). It's a beam tetrode. The Mazda Belvu data also states that the PCL and ECL have the same characteristics. The Telefunken data sheet very helpfully gives a diagram of the device structure - including the beam plates. The stated heater voltage of the PCL86 differs according to supplier - Brimar 14.5V; Mullard 13.3V; Philips 13V etc. I have 2 PCL86 valves to hand, and fed the heater of each from a constant 300mA source. A "Thorn UK" valve measured 13.6V and an unmarked specimen 12.7V. Clearly, there's some variation, which matters if the PCL is used in a parallel fed circuit to replace an ECL86. Leon. |
7th Aug 2021, 8:04 pm | #19 |
Dekatron
Join Date: Sep 2007
Location: Oxfordshire, UK.
Posts: 4,310
|
Re: The (sometimes controversial) ECL86
To add a few more PCL86 heater data points to Leon's, I've got three here, all slightly different inside in terms of getter, shielding details etc. Two are clean and in HRS branded boxes, so may be unused. One of these has a 'seamed' top to the glass and a Philips type/factory/date code indicating it came from the Ei plant in Nis, Yugoslavia. Once its temperature had stabilised it needed 14.9V across it to maintain 300mA ! The second of the HRS ones is unbranded on the glass but has a little square metal tag, stamped 86, just above the top mica. It needed 13.4V. The last one looks like it might be a pull. The main labelling on the glass is rubbed away, and I can't find any code, but there are traces of the MADE IN ENGLAND BVA text on it. It needed just 12.5V for 300mA.
Cheers, GJ
__________________
http://www.ampregen.com |
7th Aug 2021, 8:31 pm | #20 |
Dekatron
Join Date: Apr 2012
Location: Wiltshire, UK.
Posts: 13,953
|
Re: The (sometimes controversial) ECL86
Pragmatically, Mullard were out to make money - if they could do that by selling valves, that was good news for their shareholders.
I always thought of the [x]CL86 valves as being 'a bit hefiter than the similar valves with smaller last-digits' which could be advantageous - but they were not really spectacular. As a youngster I built a three-valve transmitter for use on the Army Cadets HF radio nets. One ECL86 used the triode as a crystal-oscillator and the pentode as power-amp; the other ECL86 was the audio-amplifier, fed from a carbon mic. A push-pull output-transformer acted as modulation-transformer to couple the RF and audio sides together. The result worked a lot better than the WS62 - I reckon it was good for 10 Watts of RF out. I'm kinda tempted to re-create it and put the result on 1.8MHz. Maybe using a QQV03/10 only because I've got lots of them. |