This thread is dedicated to anyone who wishes to discuss issues relating to Electrostatic, Planar or Ribbon speakers.....be they Quad, Martin Logan, Final, Sound Lab or Magneplanar.
I moved this from the AVR Cable Blind Test to a separate thread as it was beginning to look like a hijack
Wotslice said :
"....The Quads are noted for the height of their presentation but I'll test this by sitting closer to them as you have done. You say that you have your Logans 3' from the side walls but the Quads dont need this so the stereo width is good but I do wish the lounge was another metre or so wider. Note my toe-in is very slight, they are nearly straight on. I've tried various toe-in degrees and all it does is close-up the soundstage."
What sort of vocalist height are you seeing from your normal position Wotslice?
I found the Z (height) dimension was not something that happened automatically for me and for other Users setups. For some time now I've been trying to figure out an explanation for this.
When I initially setup the MLs the soundstaging I observed was flat and of uniform height. So much so that I found myself almost standing up at times to see over the top of the nearest instrument. This was unexpected but pleased to say it didn't last too long! Once the suspensions on the bass units were loosened up I was able to try re-positioning the speakers, using varying angles of tilt etc. The panels are now sitting at an angle of just a few degrees (less than 5 degrees) off vertical and have been like this for about 16 months. With the latter changes the height dimension really took off ! Vocalists at normal listening level were suddenly 6 ft tall, guitars appeared at a consistent height as did drums but the whole soundstage had an incredible volume-dependent coherence and believable relationship.
It's probably better to describe it in the following way : If you imagine a series of projection lines from the Front Wall of the room converging at vanishing point in the manner of an architect's drawing, then regard it from the listener's viewpoint, the loudest vocalists i.e. the closest, will appear 6ft tall, while those further away will appear shorter, in the same way as you would expect from natural perspective from a seated position. The height is entirely volume dependent, so if you reduced volume, all vocalists - including the front man - would seem shorter in height because they are "further away".
Of course, while this is the "standard presentation" vocals do not have to conform as I've found on other recordings or even within the same track. A vocal (indeed any instrument) can emanate from any height, anywhere in the soundstage. My first experience of this was when I was evaluating the speakers at the Hifi Dealers. Hybrids such as Logans were known in the past for having Bass integration problems. It was this thought I had in mind when listening to a Beatles track and John Lennon's voice came from the bottom end of the speaker just outside the RH channel. I assumed I had identified an integration problem, when in fact it was natural for that track. This was confirmed when JL's voice suddenly appeared 6 ft tall front and centred!! Certainly keeps you alert !!!
If I ever needed any evidence the integration problem had been solved, it was when a bass heavy sound appeared above the top of the panel (over 5ft) while a screechy violin sound appeared at the bottom end of the panel (2ft off the ground) Bass & treble had appeared to juxtapose.....
So the burning question is what biological/speaker/room mechanisms cause this perception and what influence does the studio engineer have (if any)? From the Engineer's viewpoint, I think that anything that results from the studio production is a happy accident (although this belies the apparent coherence of the effect), because the Engineer is primarily concerned with mixing left and right, although 2-D ambience is effected either through mic-ing, or by deliberate use of effect/reverb. (Any Studio Engineer who wishes to add an input here is most welcome!)
If you look at the behaviour of a speaker such as the Quad 989, it fires sound directly at the listener in the manner of a laser beam (and also in the rear direction, dipolar fashion, simultaneously - an inverse wave). The Quad being flat-panelled literally generates a rectangular wavefront from the speaker of identical size to the speaker. It is an electostat in the purest sense of the word. By comparison, the Logans use the old esl-57 curvilinear technique to broaden the sweetspot slightly in the horizintal axis, so you end up with a "slice of pie" wavefront (i.e. a flattened torch beam). The important thing to note in either case is that there is no vertical dispersion whatsoever. If you stand up and your ears are higher than the top of the speaker you will lose the high frequencies (apart from room reflections off the back wall when the speaker is tilted back).
Now relate this to the human ear....
Factors that influence height perception are :-
- High frequencies (naturally seen by the brain to be "higher")
- Angle of incidence of the sound and reflected sounds
- Electronic/physical delays in a sound which fool the brain.
Just to complicate matters, a conventional tweeter is a "point source" but our ears are not "point receivers".
Attached to the eardrum is an external waveguide/labyrinth that helps to collect sound. So while a conventional tweeter is producing exactly the same sound regardless of which part of your ear it hits, the Stat is doing something altogether different. Your ear is perceiving sound from different heights emanating from corresponding points on the panel to the ear i.e. the ear is listening to an ear sized chunk of the panel corresponding to it's height, not the entire panel because of the "laser effect described above. Thus a tall panel does not equal tall sound ! In fact, the height of the portion of the panel we are listening to may be lower in height than the average tweeter, and yet the Stat has the capability to produce massive images within an apparent "wall of sound".
It's worth noting also parts of the each panel do not move by the same degree but are travelling at different velocities, caused by the flexure of the membrane, rather akin to the "billowing of sails" on a mast. While this may produce infinitessimal vertical dispersion it's not detectable, but does it have any effect on our perception of the sound's height?
Clearly angling the panel and stimulating room reflection is also a key part of the mechanism. What I find odd is that even if the backwave is heavily damped and only a restricted spectrum is thrown back into the room, the height perception is as strong as ever, encouraging the belief that the primary sound is mainly responsible for the effect but the angle is critical?
Now here's an interesting question.....if you turned the panel sideways (horizontal) would you still get height perception? Or to cite a more ready made and practical example....could the old ESL-57 produce height info ?
That's a hell of a post Bill I've been on a golf break for a few days hence my quietness.
I'll get back you further on the height issues and I'll conduct some tests re toe-in angles. However, I would say that I do not get a 6ft effect as you suggest more like 4-5ft over head hight from a listening position, but with backing vocals certainly "behind" the lead vocal. What I do like is the bass presentation where the bass drum is just above the floor with bass guitar just above it.
I would say that my system does push forward a lead vocal, but whether that is the recording or a trait of my system (i.e. not just the 989s) I dont know. In fact I thought if anything it was a tad OTT, But when I tried some Van den Hul interconnects they somehow raised the other instruments in the mix and made it sound overall more agreeable (I know this is not a cable thread now)
Yes, the Quads do present a duplicate audio from the rear and is quite eerie when first listening to this. Another thing I've noticed is that if you stand at the back of my room (approx 24 ft) and get closer to the 989s until you are virtually on top of them, the volume doesnt seem to get any louder but I've never really understood this.
How are your Logans with dynamics, in my view the Quads excel in this aspect.
From my "serious" listening position (10 ft from the Quads, the height of main vocal presentation is about 4-4.5 ft (the speakers are 52 " tall including power supply). When I sat 5 ft from them the centre presention was higher, in the end I was sitting virtually surrounded by the Quads as if they were a giant pair of headphones !! this raised the height to well over 5 ft but the bass got a bit confused. So if you want height then sit closer and toe them in accordingly.
I've heard the new Sonus Fabers floorstanders (either cremona or concerto) and whilst not too my taste in overall sound, the presentation was amazingly high.
I'm feeling quite mellow since locating a Filling Station with petrol !! I was getting quite worried as had adopted the noble stance of not aggravating a petrol shortage and waited until I had emptied the tank. I drove past one station after another with sold out signs at the entrance.
"Yes, the Quads do present a duplicate audio from the rear and is quite eerie when first listening to this. Another thing I've noticed is that if you stand at the back of my room (approx 24 ft) and get closer to the 989s until you are virtually on top of them, the volume doesnt seem to get any louder but I've never really understood this."
I think the reason for this is conventional drivers have "inverse square" attenuation with distance i.e. they produce a "hemispherical wavefront" which disperses the acoustical power over wider angles within the room, while the attenuation with Quads is linear, but as you say it's quite common for them to seem as loud at the back of a large room as they are at the front
"How are your Logans with dynamics, in my view the Quads excel in this aspect."
I'm sure you would like the MLs...I've heard it said that the Hybrid construction gives them that extra dynamic capability (in the bass region particularly) which can make them seem loud(er) in comparison with non-hybrids. From what I've read, certainly when compared with the newest quads but unsure about the 989s. It's all about tradeoffs though, as the Quads are renowned for absolute resolution and a possibly even superior midband allied to a high level of bass precision. It's a close thing and I wouldn't like to live on the difference!
"...in the end I was sitting virtually surrounded by the Quads as if they were a giant pair of headphones !!"
It's the endless enigma...Near-field listening vs Far-field listening (although despite a lot of users referring to their listening as "Near-field", even the Stat users could never be truly Near-field unless using them as headphones .
Just reading about other people's system setup I've observed what seems to be an unofficial Rule-of-thumb for Logan users (in addition to the other one mentioned earlier about matching front vs rear wall distances) which indicates that the ideal distance from the plane of the speakers is the Speaker Pitch X 1.2 approx with a slight toe-in. (Mine are about 6.75 feet apart centre-to-centre and I'm finding the ideal distance at roughly 8 ft.)
Comb filtering can be greatly alleviated by toe-in but we need to watch out for corners as they can end up firing the reflection back through the speaker again. An extreme test would be to get someone to hold flat mirrors on each wall so that you can tell whether you can see the rear RH or LH edge of the Quads from your listening position - a bit drastic though .
My favoured option was to damp the backwave heavilly, with open, symmetrically arranged curtains while accepting a reduction in sensitivity of just less than 3db....(92db minus 3db). They still sound pretty loud despite this although I admit this is only good if you feel you can afford it.
This also softens the response and tames any room tendency towards emphasis at high frequencies.
It's a bit of a minefield, as I've found that increased toe-out, and focussing on the inner edge, actually increases the HF content (you would think being on-axis would do that but it can be exactly the opposite with Curvilinears.)
Just out of curiosity Wotslice, is it possible to tilt the 989s back just a whisker e.g. by using spacers/adjustable feet or would that de-stabilise them ?
Hi Bill, the short answer is no as the panels effectively "slot in" to the power supply assembly but they have spikes (which I dont use) but I really wouldnt want to tilt them at all really due to the potential risk of damage. Incidentally I have the ESL63s on stands and this raises them about about 9" of the floor and also tilts them <10 degrees or so.
When I first heard the ELS63 they were raised about 2ft off the floor which probably explains the massive image I heard.
Regards
PS when I get a moment I'll read up some on your stats.
Beat you to it........... Several days ago I looked out some Reviews of the Quads.......Very nice indeed and bearing a much stronger generic resemblance to the current ones than I realised........ In fact, if you factor in inflation the 989 is probably even more expensive than the new ones............................................................Actually I think my suggestion about tilting the 989s may have been misguided...... I've been having a think about it and realised if anything they might already might be designed to lean back to avoid over-balancing?........Having said this I'm beginning to think a true vertical is probably nearer to the ideal for stats in general.......Achieving that ideal however is a different matter - even with the use of a spirit level (which invariably has limited accuracy)..............................................................................................................For ultimate results I've found the position, toe-in, and rake angle (or degree of vertical-ness) have to be absolutely identical. Even if the user measures the positions of the bottom corners relative to the wall with a tape measure, the vertical angle is always going to be a tough one (although I think the 989s may have an advantage over everyone else in this respect if they are normally close to a true vertical....it's probably worth checking if they are on a true level surface?).........................................................The Logans are quite helpful in this respect as you can see your reflection at the listening position in corresponding points in the membrane of each speaker, so you can gauge how well centred you are (although I also know where the true centre is by measurement)...................................A lot of scientific methods have been proposed for getting the rake angle matching e.g trigonometrical, Laser level with angle measurement etc. but I have a simpler method which seems to work quite well.....Assuming the User has done the routine measurement of distance from the Front Wall, if you stand off to the side of the system and look across the face of the panel towards the other speaker you can line them up by eye and check the other speaker's angle "shadows" the nearest one - this method is probably accurate to within one degree or so !..
...Certainly from your point of view, it makes perfect sense that the image appears to get taller as you get closer as this seems to be a function of volume (the height of the tweeter i.e. if a true vertical and true mirror - does not change so the only thing that can change is the volume!).................................................................................................................................Getting back to the original subject of soundstage height I've had some radical thoughts on the matter....One thing that is a "constant" regardless of what sounds emanate from a speaker is the behaviour of electric instruments....Electric instruments by nature are devoid of ambience.....This means that an electric guitar coming from a specific channel with no added studio reverb will appear to spring off the surface of the panel directly from the point your ear is focussed on................Regardless of how your system performs and what kind of stats you use it will always do this....In other words, I have never heard an unreverberated electric instrument do anything other than this i.e. it is incapable of appearing at any other height than the strict height represented by the corresponding stat panel....The moment reverb is added to the guitar, it immediately begins to behave differently : The sound may step to the outside of the speaker and further back, allowing the guitar to appear as a solid representation within it's own acoustic space ?..................................The significance of this fact is not lost on me......It means that the primary reason for perceived difference in location is in fact the signal itself...not the room...or the panel's relationship with it.....
.....The best example of reverb-induced positional change I've heard was on one of Carl's favourites.........It followed on from Pupilla-Tommy-Pupilla on the album Moving Waves"............ The guitar started centre stage, slowly panned left to that channel, stopped, rose up the centre of the panel from 2.5ft climbing to 5 ft (top of the Stat), panned across to the other channel slowly at 5ft height, then slowly descended down to 2.5ft again, before panning level height to centre. I was pretty dumbfounded at hearing this...but it was fascinating to observe !!...........Kind regards, Bill.
....Not that I think the Engineer had any Z axis intentions on his part, just that this is how the signal worked out on Stats. When using conventional drivers on this track I was unaware of such an effect - apart from the basic L-R fading of the guitar, so this would also probably be true of the Studio Eng....
I've never heard anything quite like you describe on the Moving Waves album, must have been a bit eerie when you first heard It?
Re the 989s then I guess you've read some of the reviews and that is why I purchased them as well as being in awe of Peter Walker. I have the Nouveau finish which is Blue grey cloth and light grey power supply covers - to me they are beautiful but to others (somehow) they are hideous.
Re tilting, your post has had me bevering away, as I said on an earlier post they do come with spikes which if fitted will tilt them back slightly but as I keep moving them around I've not bothered with the spikes, so earlier I put a small book under each (just less than an inch) which tilted them back and the result ? more volume esp bass, a more focussed image with things being more "positioned" and the vocals whilst not really being any much higher were just "bigger". I listened to some vocal material - Paul Simon and then Lou reed (Magic and Loss - all about death, but I rate the album) which emphasised the difference in the lead vocal and it was the "in the room" experience.
I'll try the spikes out soon to see if this is really how they need to be set up (I assumed they were optional).
I totally agree with you about the exactness needed for rear wall placement and toe-in.
Now you've got me worried Wotslice ...I'm picturing these beautiful (expensive and tall) speakers teetering on a couple of books
Great to exchange feedback. I'm always interested in the results people are getting and I store it away in my internal database in the hope that it adds to my own experience !
Sounds like you are getting very interesting results. One more thing I've noticed which may help. Since mine flank a window recess which contains vertical blinds, I'm able to use the blinds as diffusers by changing the angles? Unsurprisingly, this changes the sound from sharp focussed, to bigger stage more softened. (i.e. corresponding to closed/angled blinds) I like the sharply focussed image best.
When angling I stop short of exposing the window glass to standing waves.
Do you ever use a home cinema receiver through the 989s?
A useful feature of these devices is the White/Pink noise generator (although the more advanced ones even come with Odessey Room Optimisation software & hardware!)
What I find unusual about Electrostatics is they have the capacity to present the Pink Noise hugely differently at all locations (!) i.e. L, Centre and R (using a phantom centre) can sound quite different from each other even though L/R are creating the Centre sound. First time I tried it the L/R channels were tonally imbalanced, lightweight and "hollow" while the centre appeared full bodied and normal ! This had me scratching my head and was one of the things that led to changing the listening position forward - which had the effect of matching all 3.
The ideal of course was to get the Pink Noise to sound identical (at the very least, if not tonally acceptable) at all 3 locations - which was easier said than done! This began a long journey into damping, Room Optimisation, and symmetry. Even if the dining chairs in the background were moved slightly it impacted on the Pink Noise balance.
Unfortunately I don't run the analog system through the processor so I had to come up with another method for vinyl, and it was a lot simpler. Now that the listening position was corrected, the only requirement was that centre stage vocals had to appear dead centre (this had to include the midrange and treble/sibilant component of the voice. If this was successfully achieved then the chances were the 2 speakers were similarly characterised as seen from the listening position?
The good news is that I hardly ever use the Home Cinema through the Stats as it's tantamount to abuse
No I dont use the Quads as home cinema speakers, this is mainly becuase I havent got a pre amp so I cant plug the TV into the system. My CD player is quad CDP2 into a Quad 909 power amp. The CD player can accept digital signals but not analogue, so if I did get a pre-amp I would at least try the speakers via the TV system. My view on Quad equipment generally is that in terms of VFM nothng is better, although syaing that I've not had the opporunuty to compare my cd/amp with others in the same listening space.
What other gear do you use?
I've been listening to a lot of material using the ESLs tilted as my wife had previously bought a number of plastic wedges for another purpose, but she gave them to me. This provides a very small tilt so I've dont need the books now Anyway, the change in the sound is quite amazing considering this is a ridiculously cheap tweak. The image is pin-point, natural and much wider, so off-axis listening is not so much of an issue.
I've got both ESLs balanced so all aspects of a say a lead vocal are in the same space (including any sibilance - although this is one recording trait I hate).
I've checked the manual for the 989s and the recommnedtaion is 2ft from the rear wall and none from side walls with and toed-in although it doesnt recommend the degree of toe-in, As I said earlier mine are slightly toed-in maybe 15 degrees or less.
Wow! Pleased to hear that you've hit such a rich seam of experimentation! It all goes to prove what they say : that small movements of this type of speaker makes for surprising differences. My only regret is that it takes such a long time to get through all the permutations!! .
Sibilance is a revealing thing. It's often said that electrostatics make a meal of any harshness in the Source. This popular misconception was a worry until I found all of my sources were reassuringly smooth and natural, so I needn't have agonised over it. But what I did find was that amplifiers digital displays can make sibilants - and indeed the rest of the signal, sound unnatural. Turning them off is like the system suddenly "stepping out of the way" of the music. It can only be RFI....
The beauty of the 989s is their lateral dispersion is zero, so you can go as close to side walls as you like (probably just as well given their size ). For others not similarly blessed you need at least 1ft (Although...having said that...there is the question of the backwave when toed-inwards....??)
Yes, the 909 is very much "de rigeur" for Quad enthusiasts because it's a great match. In fact it's an excellent idea in general for a speaker manufacturer to create amps that can actually drive these difficult loads, then test the idea to destruction! So I'm pretty sure you've got an ideal amp which will survive the passage of time.
My other gear comprises Valhalla'd & tweaked Linn LP12/Ittok LVII/Asaka MC with matching MC board within a Naim NAC32 pre-amp (now only used as a Phono Stage). This is an old design but don't let that fool you. I did an AB with a proposed replacement "latest" design £1200 phono stage (manufacturer of which shall remain nameless to avoid embarassment) which was so rare according to the main distributor that my loaned unit was one of only 3 in the UK (!) and the Naim outclassed it easily - partly because it was a perfect match for the cart - unlike the switched and tuned Pretender...
So instead of putting the Naim on E-Bay, I ended up returning the expensive phono stage. (Just to add insult to injury, the Naim had been lying disconnected and cold for 2 weeks, while the new Phono had been thoroughly warmed up for the same period. Plugging in the Naim from cold, the Naim sounded like it was the freshly burned-in model ! Sweet, smooth, grain free and totally natural. I was both pleased and disappointed all at once ! I realised it was going to take a lot longer to find a replacement...(still working on it).
This connects to a Simaudio Moon i-7 RS Dual-mono (same age as the Vantages). In comparison with other amps I'd A-Bd, this one really stood out as a bit special. 500VA transformers in each channel, with loads of reservoir capacity, so plenty of power on tap. The Vantages are 92db efficient so in theory don't need it (but you can hear the benefit).
The Yamaha E-800 HC Digital DSP Processor/Amp I may have mentioned earlier is a useful gadget even if you don't listen to HC much. Doesn't cost the earth but does the job of providing rear channels and a centre if desired. Great for adding ambience that CD often takes out - and this is fully exploited by Stats.
Your notes about positioning had me thinking about the 989's backwave. I would never dare to disagree with Quad, but I would think that for this type of speaker 2ft would be seen as an absolute minimum in any given room. I can only think from Quad's point of view this could be a "bass response thing"?
Reason I say this is received wisdom on some recent trials using damping materials or reflectors close- up to the rear of the speaker shows a tendency to reaction. In a way it's likening the membrane to the chest cavity or lung and giving it room to breathe. Effectively the ultra thin lightweight piece of mylar is trying to compress the air between the loudspeaker and the wall during the backwave so the less reaction (or the more volume of unrestricted air) it sees the more comfortable the movement?
Certainly the 2905s I saw at the last exhibition were miles out from both side and rear walls and I was sitting there enjoying the sound, bass response anomalies didn't even enter my head because the sound was fine. (Obviously the Demonstrator thought so too as he set it up this way )
I can't believe the 989s are so different from the 2905s...
(Having said this, probably half the Stat users in the world use this same clearance Quad describe, and are happy! But you still wonder what if....)
Just a thought mate, but nothing to lose any sleep over
.
First time I tried it the L/R channels were tonally imbalanced, lightweight and "hollow" while the centre appeared full bodied and normal ! This had me scratching my head and was one of the things that led to changing the listening position forward - which had the effect of matching all 3.
