ESLs: How d'you tell if they're 100%?

[Mike Reed]

Since converting to big ESLs (QUAD 2905) I've often wondered how one tells if anything is amiss. With coil speakers, it's easy to simply listen to each speaker unit, but cloth-covered ESLs are not so obvious.

How would one tell, e.g., if one panel is down? Simply by listening close up to six (in my case) different parts of the front cover? This is not definite as it depends upon the music, surely.

I know nothing about if/how a panel can expire or, really, anything about modern ESLs, nor how many bass as opposed to treble panels there are (4 x bass, 2 x treble?). Could a bang from a blown fuse take out all or individual panels, despite supposedly having excellent safety precautions built in?

ESL learning curve starting, hopefully.

 

[Robert]

Hi Mike,

There can only be 2 treble panels in any Quad in order to maintain the point source arrangement. The bass panels sit above and below, singles or doubles.

On the old ESL63 a neon lamp was used on the rear to indicate excessive panel discharge, the usual sign that something is wrong or that things aren't in good order. Do the 2905s have the neon lamp?

You are probably aware of the usual tests, listening for ticks and pops from the panels or even hissing.

Another way to check is to invest in a cheap USB measurement mic and a copy of either REW or perhaps ARTA and measure the speakers. This pretty easy to do and with both Quads in the same position you should get the same response. Take a reference measurement then repeat every 6 months or so. Any panel failure should be be obvious.

 

[Craig B]

Here's a low tech option, Mike.

• Place the two panels out into the middle of the room facing one another quite closely together (i.e. almost, but not quite, touching one another). Shim the rears, as necessary, such that the panels are as parallel as possible.
  
• Wire one ESL out of phase WRT the other.
• Run a mono signal into them both and listen for any obvious range of sound that isn't cancelled.

They should be pretty quite, but for some sound 'leakage'. Just remember not to turn the volume up to beyond normal.

 

[johnfromnorwich]

I had (have) '63s and I used the neon light to check the humidity of the room... I didn't notice any deterioration until a panel was on the way out. When they are about to go, it's pretty clear - white noise or clicking and popping with the source defeated and your ear up close. When they go, you'll definitely know! I wish I could afford to get them re-panelled tbh.

 

[Mike Reed]

Do the 2905s have the neon lamp?

You are probably aware of the usual tests, listening for ticks and pops from the panels or even hissing.

Another way to check is to invest in a cheap USB measurement mic ..../QUOTE]


No the 2905s only have the on/off light. No I didn't know of the usual signs, so thanks for that (I did say I knew nothing about ESLS !) I'm afraid I'm too old and low-tech to get involved with usb mics or other computer facilitated stuff.

• Place the two panels out into the middle of the room facing one another quite closely together (i.e. almost, but not quite, touching one another). Shim the rears, as necessary, such that the panels are as parallel as possible.

• Wire one ESL out of phase WRT the other.

• Run a mono signal into them both and listen for any obvious range of sound that isn't cancelled.

That's sufficiently low-tech for me to accomplish, Craig, should I not be sure, but advice above and below about clicks, pops etc.(which I didn't know) are a good first indicator, I'd think.

When they are about to go, it's pretty clear - white noise or clicking and popping with the source defeated and your ear up close. When they go, you'll definitely know!

Reinforcement of above indications, so thanks (If I remember a conversation many years back, you're either not from my Norwich or it was aeons ago when you were )

I had a fuse go back last December on one of my 100 watt EAR amps and it created quite a bang, but 5 weeks ago another went on the same amp, only this time out of the blue in the middle of playing a record (faulty valve, I was told; okay now with substitute pair). Don't think my wife has recovered yet; it was loud. Had a session recently after amp. was sorted where the music didn't inspire, and I wondered. However, there was sound at each of 6 parts of the panel and no clicks, pops or whatever, so I was probably having an off night.

Good educational stuff for me, gentlemen, and I feel better for it !

 

[naimnut]

Here is my experience on Quad ESLs. I've owned ESL 63s (should have never sold them. Sold them to my cousins husband, before I really appreciated how rare they were to be trouble-free and perfect. They were unmolested and functioned perfectly.) I replaced my 63s with fully serviced 2805s. My experience with the 2805s matches up quite a bit with some of the comments you read around the 'net. To be brief, my speakers perform incredibly 98% of the time. But there are a few frequencies that can trip them up. A few tracks... And then they make me crazy.

For many years I knew nothing about humidity and did nothing to control it. Kent McCollum, of Electrostatic Solutions, who came to my house at my expense to repair a speaker that had been sent to him but still had annoying distortions, advised me to keep the humidity below 55%. I now run a dehumidifier frequently and I try to keep it low. My wife thinks I'm nuts. The source of the noise turned out to be an intermittent solder joint somewhere down in the power supply. Very difficult, no - almost impossible, to track down. Kent was able to do so, after replacing all my panels for the third time on that one speaker. That speaker still has noise and buzzes. Rarely, but it happens.

If I have my speakers serviced again I doubt that I will ship them to have them serviced. I'll see if I can't fly a Tech into town to do it. Shipping them is a recipe for problems.

One of my speakers likes to make noise. The is the one that has only been serviced once and has generally functioned well. Quiet whooshing. Burps. Most of the time it sounds superb.

From what I've read, the current production Quads have a very high failure rate. I don't think it is fair for me to put a number on it. But it's a high rate. It's just a matter of time before they fail. My expectation is that Quad, as a company/technology that will either A) be sold to someone else who wants to try to figure out how to produce Quads now that the people who really know what makes them tick are no longer around, or B) it will gradually simply wither away, with a limited number of dealers and service network and many very rich, very dissatisfied, very hooked, very disappointed owners. Probably they'll all live in China or somewhere in Asia.

The rest of us will struggle on, keeping somewhat recent production as well as vintage Quads alive, and loving them when they perform, with the help of specialists and a lot of patience.

Now. If only Soundlabs weren't so darn huge and expensive. Well, one can dream, can't one?

 

[Jim Audiomisc]

Since converting to big ESLs (QUAD 2905) I've often wondered how one tells if anything is amiss. With coil speakers, it's easy to simply listen to each speaker unit, but cloth-covered ESLs are not so obvious.

How would one tell, e.g., if one panel is down? Simply by listening close up to six (in my case) different parts of the front cover? This is not definite as it depends upon the music, surely.

I know nothing about if/how a panel can expire or, really, anything about modern ESLs, nor how many bass as opposed to treble panels there are (4 x bass, 2 x treble?). Could a bang from a blown fuse take out all or individual panels, despite supposedly having excellent safety precautions built in?

ESL learning curve starting, hopefully.

Your use of the term 'panels' may be a source of confusion.

The 57s had distinct panels. The 63s and their 'descendants' only have one sheet of diaphraghm, but a set of driving electrodes. The signal is fed to the central electrode pair, and then delayed+filtered by a network connected to the 'ring' electrode-pair nearest the center, then again via another network to the next 'ring', etc.

This means the outer panels don't get the same HF as the central ones.

So a failure that only stopped one part of the output would be due to somethigng like a break in wiring, etc, to an individual electrode. Which I suspect is rare.

However If the *panel* goes - i.e. the single charged sheet that moves - then the entire output would be affected.

 

[Robert]

Your use of the term 'panels' may be a source of confusion.

The 57s had distinct panels. The 63s and their 'descendants' only have one sheet of diaphraghm, but a set of driving electrodes. The signal is fed to the central electrode pair, and then delayed+filtered by a network connected to the 'ring' electrode-pair nearest the center, then again via another network to the next 'ring', etc.

This means the outer panels don't get the same HF as the central ones.

So a failure that only stopped one part of the output would be due to somethigng like a break in wiring, etc, to an individual electrode. Which I suspect is rare.

However If the *panel* goes - i.e. the single charged sheet that moves - then the entire output would be affected.

IME partial failure can also occur due to loss of panel coating. This is particularly the case today when coatings often differ from the original Quad mix.
I've had a 63 lose HF where one half of the central pair had started to lose sensitivity. That was a non Quad serviced panel.
A return to Quad for service fixed the problem. They also confirmed that a non stable coating had been used, essentially graphite rubbed onto the film, and it comes off!

 

[Werner]

The 63s and their 'descendants' only have one sheet of diaphraghm, but a set of driving electrodes.

Sorry Jim, but that's not quite correct. The ESL-63 has four entirely separate panels, the most central ones of which contain the point source ring electrodes, fed a full-range signal, the outermost ones being just ordinary panels, fed only bass (IIRC). There really is no single speaker-sized diaphragm.

(Edit: could be I misunderstood you, in which case only the first two words above apply.)

 

[Jim Audiomisc]

Sorry Jim, but that's not quite correct. The ESL-63 has four entirely separate panels,

I sit corrected!

 

[Mike Reed]

he 63s and their 'descendants' only have one sheet of diaphraghm, but a set of driving electrodes.

IME partial failure can also occur due to loss of panel coating.

Sorry Jim, but that's not quite correct. The ESL-63 has four entirely separate panels, t

I started this thread confused and indecisive but now I'm not so sure.

Well, my 2905s either have separate panels or just one diaphragm but whichever, it's unlikely/impossible to simply have one part malfunctioning; i.e. losing bass or whatever. I guess and hope that the safety devices quad supposedly use for these are sufficient to cope with the transient yet loud effects of a fuse blowing.

I have never had this problem (in 55 years) with fuses/output valves (?)/ESLs; in fact, I can't remember when I had a fuse blow (40+ years (?), yet two in 8 months/same amp. Doesn't inspire confidence. Thanks all for your input.

 

[Cesare]

I think the earlier suggestion is the right one. You need to measure the speaker output, using something like REW or FuzzMeasure, so you need a computer, the right software and a measurement mic. At that point, you should be able to measure a speaker, then switch the second speaker into the same measurement position, and measure it. If the two speakers have the same response, either they are both working, or they are both broken in an identical way, and that is unlikely. How close would the responses have to be for you to feel that is the case? Well, that's a different matter, but I think if there was a problem the responses would look significantly different in a way which would suggest something is broken.

By keeping the test setup the same and moving the speakers to the setup, it means that the room interactions should be the same for both speakers, and hence you can ignore the actual response curves, and just look at the deltas between the two speakers to reach a conclusion.

 

[DimitryZ]

Here is my experience on Quad ESLs. I've owned ESL 63s (should have never sold them. Sold them to my cousins husband, before I really appreciated how rare they were to be trouble-free and perfect. They were unmolested and functioned perfectly.) I replaced my 63s with fully serviced 2805s. My experience with the 2805s matches up quite a bit with some of the comments you read around the 'net. To be brief, my speakers perform incredibly 98% of the time. But there are a few frequencies that can trip them up. A few tracks... And then they make me crazy.

For many years I knew nothing about humidity and did nothing to control it. Kent McCollum, of Electrostatic Solutions, who came to my house at my expense to repair a speaker that had been sent to him but still had annoying distortions, advised me to keep the humidity below 55%. I now run a dehumidifier frequently and I try to keep it low. My wife thinks I'm nuts. The source of the noise turned out to be an intermittent solder joint somewhere down in the power supply. Very difficult, no - almost impossible, to track down. Kent was able to do so, after replacing all my panels for the third time on that one speaker. That speaker still has noise and buzzes. Rarely, but it happens.

If I have my speakers serviced again I doubt that I will ship them to have them serviced. I'll see if I can't fly a Tech into town to do it. Shipping them is a recipe for problems.

One of my speakers likes to make noise. The is the one that has only been serviced once and has generally functioned well. Quiet whooshing. Burps. Most of the time it sounds superb.

From what I've read, the current production Quads have a very high failure rate. I don't think it is fair for me to put a number on it. But it's a high rate. It's just a matter of time before they fail. My expectation is that Quad, as a company/technology that will either A) be sold to someone else who wants to try to figure out how to produce Quads now that the people who really know what makes them tick are no longer around, or B) it will gradually simply wither away, with a limited number of dealers and service network and many very rich, very dissatisfied, very hooked, very disappointed owners. Probably they'll all live in China or somewhere in Asia.

The rest of us will struggle on, keeping somewhat recent production as well as vintage Quads alive, and loving them when they perform, with the help of specialists and a lot of patience.

Now. If only Soundlabs weren't so darn huge and expensive. Well, one can dream, can't one?

This is essentially my experience as well.

If I lived near Kent or another US Quad guru, I would still own mine.

Now I am a happy owner of Eminent Technology LFT-8Bs - very similar functionality and they don't go bad.

 

[Craig B]

I started this thread confused and indecisive but now I'm not so sure.

Well, my 2905s either have separate panels or just one diaphragm but whichever, it's unlikely/impossible to simply have one part malfunctioning; i.e. losing bass or whatever. I guess and hope that the safety devices quad supposedly use for these are sufficient to cope with the transient yet loud effects of a fuse blowing.

I have never had this problem (in 55 years) with fuses/output valves (?)/ESLs; in fact, I can't remember when I had a fuse blow (40+ years (?), yet two in 8 months/same amp. Doesn't inspire confidence. Thanks all for your input.

The method described in my post #3 above was long ago devised by one Peter Walker, and mentioned round the time of the release of ESL63 as being how Quad ensured that each loudspeaker performed the same as their lab reference example. It isn't necessarily about a portion of one panel set being out, more a case of each ESL pair sounding the same. IOW, a quick test against the reference to determine whether a given production example may need further investigation, as, even any structural resonances will be revealed once the net output of the pair is cancelled.

 

[Mike Reed]

You need to measure the speaker output, using something like REW or FuzzMeasure,

Thanks for the input, Cesare, but anything to do with digital/computer/measurements/sofware is wat beyond my ken (or interest, t.b.h.)

This is essentially my experience as well.

With 63s, presumably. They don't have a remarkable reliability track record, but them again, so many were sold c/f the 989s, 2905s and 2912s + their smaller counterparts. My amps (EAR 509s) were designed around 63s (I was told by the man). That was 3+ decades ago.

 

[DimitryZ]

Thanks for the input, Cesare, but anything to do with digital/computer/measurements/sofware is wat beyond my ken (or interest, t.b.h.)



With 63s, presumably. They don't have a remarkable reliability track record, but them again, so many were sold c/f the 989s, 2905s and 2912s + their smaller counterparts. My amps (EAR 509s) were designed around 63s (I was told by the man). That was 3+ decades ago.

I owned 57 and 989s.

63s and speakers that followed them were very close cousins. Later versions added more bass panels, more structural bracing, maybe used different parts or PWB layout. But the circuit remained the same.

63 was Peter Walker's definitive and final design.

 

[mandryka]

Since converting to big ESLs (QUAD 2905) I've often wondered how one tells if anything is amiss. With coil speakers, it's easy to simply listen to each speaker unit, but cloth-covered ESLs are not so obvious.

How would one tell, e.g., if one panel is down? Simply by listening close up to six (in my case) different parts of the front cover? This is not definite as it depends upon the music, surely.

I know nothing about if/how a panel can expire or, really, anything about modern ESLs, nor how many bass as opposed to treble panels there are (4 x bass, 2 x treble?). Could a bang from a blown fuse take out all or individual panels, despite supposedly having excellent safety precautions built in?

ESL learning curve starting, hopefully.

I once spoke to Quad about this re the 63 -- I suggest you do the same about your speakers. They basically said that if they weren't performing as they should, they'd be arcing and I'd know. I was ready to send them in for a service but they said wait till you hear clicks and pops.

That was 15 years ago. Prey for me.

(I should add that I think they're worth the trouble! At least if you can get them to be serviced easily.)

 

[rstloup]

They basically said that if they weren't performing as they should, they'd be arcing and I'd know. I was ready to send them in for a service but they said wait till you hear clicks and pops.)

I recently got a pair of 57s and spoke to Sheldon Stokes who rebuilds Quads in the east coast of the US. He said something similar when I asked about rebuilding my pair:if there is no arcing or sign of something wrong, the rebuild will only improve the sound marginally. Rather, it's worth waiting till something goes wrong.

 

[Mike Reed]

I owned 57 and 989s.

As I'd had most types of 'normal' speakers (mostly big but LS35As and other smaller ones), I fancied trying ESLs. Luckily I picked up a pair of 1960s 57s locally and they turned out to be in good nick. I liked the clarity (I had ProAc Response Four behemoths at the time) but
they didn't do scale.

A few weeks later Audio Emotion (?) had 11 month old 2905s in Classique finish (bit like the 63s) so I got them, gaining both the clarity AND the scale (in spades !). Guess they're about 7 or 8 years old now but they've been used on average 1.5 x a week at most and not been thrashed (well, have never had the o/load protection operate anyway).

Okay, I'm about 100 miles from Cambridge but I'd hate to try to fit them into my small estate. Very much doubt I could put all the complex packaging together anyway ! This is stored in 2 sheds, the garage and the loft !!!!!

 

[DimitryZ]

As I'd had most types of 'normal' speakers (mostly big but LS35As and other smaller ones), I fancied trying ESLs. Luckily I picked up a pair of 1960s 57s locally and they turned out to be in good nick. I liked the clarity (I had ProAc Response Four behemoths at the time) but
they didn't do scale.

A few weeks later Audio Emotion (?) had 11 month old 2905s in Classique finish (bit like the 63s) so I got them, gaining both the clarity AND the scale (in spades !). Guess they're about 7 or 8 years old now but they've been used on average 1.5 x a week at most and not been thrashed (well, have never had the o/load protection operate anyway).

Okay, I'm about 100 miles from Cambridge but I'd hate to try to fit them into my small estate. Very much doubt I could put all the complex packaging together anyway ! This is stored in 2 sheds, the garage and the loft !!!!!

When humidity is 60+%, they may start making a self generated sound, like soft rain. As time progresses, they may graduate to ticks and louder pops. Or they may not and be fine for a really long time.

Some dedicated users learn how to repanel them and say that it's not too hard. I felt that they aren't designed for easy owner maintenance.

It's better if you live in a dry climate or have humidity controlled interior.