Many transistor amps will double power as you cut the load impedance in half. The Quad has high impedances in the bass and low impedances in the treble/HF region. Most transistor amps will play the speaker with weak bass and too much highs.
Bass is hard to get with transistors and Quads. The problem is that the speaker does not run according to the rules that a lot of box speakers use- it seems to work best with amplifiers that can make constant power regardless of load- which is usually a tube amp. Many transistor amps will double power as you cut the load impedance in half. The Quad has high impedances in the bass and low impedances in the treble/HF region. Most transistor amps will play the speaker with weak bass and too much highs. |
Given Atmasphere's explanation, maybe this is why some find that the old Quad 303 works reasonably well with Quads. It is solid state, but my understanding is that its power output actually drops as impedance drops. There are a few firms that do updates, like Net Audio or the new one that advertises on EBAY. |
Post removed |
The transistor amps will not give you what you want , in my opinion. I have been a confirmed user of transistors since I built a Heathkit in 1963 and was a Quad dealer for 15 years. Transistors are undoubtedly faster on cone speakers , because of higher dampening factor among other considerations. But these factors do not apply to the Quad, which is already faster than any cone. When I was a dealer I drove my 63s with Krell, Musical Fidelity and Stax class A solid state amps but could not detect a difference in speed or bass between them and good tubes. A friend drives his 57s, which I sold him in 1975, with a rebuilt Heath tube amp , it sounds as good as anything I have ever heard drive them. Foster Blair rebuilds them, they are available through Mapleshade. This is one case where having a transformer in the line helps the sound. After all, they were designed with a lower powered tube amp in mind. If I wasn't happy with the sound of the 57s driven by a good tube I would look at another speaker, they are what they are and the amp will not change them that much. |
The Quads will only take a limited amount of power and can be damaged by over driving. I never had any trouble as I was careful with the volume but one of my customers bought 63s and a Musical Fidelity 270 [100 Watts?] and had to have them repaired 3 times because he constantly over drove them. A big amp is fine used with care. |
Never used them with the speakers, but based on Atma-sphere's comments, it is possible that the older Pass Aleph amps might work in terms of the impedance issues mentioned - they are of course very good sounding SS amps, but the matching issue may be handled by the Aleph's output into changing loads, if I rememebr correctly, they worked much like tubes in that sense. |
Sarcher30, Innersounds are built with lower impedances in an effort to make the speaker more compatible with transistors. The Quad ESL 57 has a peak of more than 50 ohms in the bass; there simply are not a lot of transistor amplifiers that can drive a load like that with any power. To give you an example, let's try a 200 watt SS amp, which does 400 watts into 4 ohms. Into the Quad, it will be limited to about 30 watts in the bass. OTOH, it will be able to make 500 watts at high frequencies! Its easy to see why people perceive such amplifiers as very bright and no bass when used on the ESL 57. |
Atmasphere, Thanks for setting me straight. I have no experience with quads and Atmasphere's knowledge is much more complete than mine. So it looks like a tube amp is the best course. I read somewhere that for instance if 100 wpc into 4 ohms produces 100db spl then 100 wpc into 8 ohms should produce 103 db. I think it was a car stereo book. It seemed to say that every time you half the impedance you loose 3 db of output. This seems to contradict what you are saying. Does what they are saying only apply to the voltage paradigm and not the current paradigm. I've read many of your posts about this and I still don't quite understand it all. Thanks Sean |
I found this exerpt from The Complete Guild To High End Audio. "Some amplifiers barely increase their output power when driving 4 ohms; others can double it. This means that not all 100wpc amps are created equal. One 100wpc amplifier might put out 150w into 2 ohms, while another might deliver 350wpc into 2 ohms. This ability to drive low-impedance loads (specifically, to deliver lots of current) has a large influence on an amplifier's sound and subjective power capability. This is where our dBW rating again comes into play. When calulating dBW, subtract 3dBW from the 8 ohm dBW rating when the amp is driving 4 ohms, and subtract 6dBW when the amp is driving 2 ohms. In the example of of the two 100wpc amplifiers in the preceding paragraph, each is rated at 100wpc into 8 ohms, or 20dBW. But the first amplifier puts out only 150wpc into 2 ohms, while the second puts out 350wpc into 2 ohms. Converting the 2 ohm power ratings into dBW, we see that the first has a dBW rating of 15.76 at 2ohms, while the second has a dBW rating of 19.44dBW into 2 ohms. Quite a difference-nearly 4dB-between two 100wpc amplifiers." He doesn't explain why you have to subtract 3dBW when driving 4 ohms as opposed to 8 ohms but I have read this elsewhere also. Maybe some transitor amps run out of voltage when trying to drive a high impedance load and thats why there is a tonal imbalance? Or am I wrong about this? |
Hi Sarcher30, in the case of transistor amps, the ideal is to be able to deliver the same output voltage regardless of the load impedance. That being the case, then a 100 watt amplifier driving 8 ohms will make 200 watts into 4 ohms and 400 watts into 2 ohms until the current limits of the power supply or output section are reached. They thing they don't tell you is what happens when driving higher impedances, like you see in the Quad ESL 57. Into 16 ohms you get 50 watts, into 32 ohms you get 25 and the ESL 57 has impedances in the bass well in excess of 45 ohms. So transistor amplifiers cannot make power in the bass, while at the same time they make too much in the highs, where the impedance of the speaker is down to 4 ohms. This is what I was trying to explain earlier. The Quad's impedance curve has nothing to do with box resonance in fact it has nothing to do with resonance at all. So it does not use the rules where the constant voltage characteristic is useful. It expects constant power out of the amplifier, or at least the attempt at it, for best results. It is what I call a Power Paradigm device, which is why transistors for the most part are tricky at best to get even mediocre results. IOW its an equipment mismatch. see http://www.atma-sphere.com/papers/paradigm_paper2.html |
hi ralph: i own a pair of pk quad 57. you can get bass extension with a solid state amp. wayne has used several. in my own experience, beefy power supply and more power=greater extension and fullness. personally, i wouldn't use a ss amp on the quads. i do not like the upper mids, lower treble, where ss amps are notoriously , unbalanced, i.e., somewhat peaky. the problem with ss amps is not bass response, it is at 1000 hz and higher. I experienced this on my magnepan as well. i got reat bass with a 350 watt class a, but female voices and violins were a big problem. |
I just noticed one that should do a good job if you want to use solid state, the Audiophile Boutique is closing out the Van den Hul Array 1 for under $1700 US, about 40% of original cost. They are very overbuilt for their output and should be able to drive Quads, I would check with them to make sure. If you are not familiar with Van den Huls work look him up, he knows what he is doing. My first post was not intended to say you could not use transistors on Quads, obviously I did. I just did not experience a greatly increased bass. I use transistors for the same reason Mr. T uses tubes, better vocals and top end. Many like the euphonic coloration tubes add, I prefer accuracy. On the power question, a transistor amp SHOULD double its power when you halve the impedance. Those that do are called voltage source amps, the ones that level off or decline are current source amps. The voltage amps require much bigger supplies and heat sinks etc. and are usually much more expensive. The output of a speaker is governed by the efficiency of the speaker as well as the output of the amp. The efficiency is given in terms of what its output in dbs are for one watt input. Doubling the amps power output will raise the output of the speaker 3db. So a speaker of 89db efficiency will require 1/8 the power of an 80db speaker if my math is on. [ 1 watt times 2 =2 watts, times 2 =4 watts, times 2=8 watts. ] You can see how very efficient speakers require little power. |
Hi Ralph, So the box resonance is what makes the voltage paradigm work. Without the box the current paradigm is the way to go. So I assume open baffle designs and horn designs would need the current paradigm also if their impedance curve is not flat. That makes sense and would explain why alot of people perfer tube amps for these type of speakers. Thanks for taking the time to explain. |
I know everything Ralph has said in this thread is dead-on accurate (as always). It really makes me wonder how some people (some people who DEFINITELY know what music sounds like such as Chesky) power their Quad 57s with the ASR amp. The Emitter is, indeed, a class 'voltage-paradigm' solid-state amp, with power output nearly inversely linear to impedance. Based on the Quad's impedance plot, it should sound terrible, but apparently it sounds excellent and perfectly balanced. (6Moons also attests to this in their report on Robin Wyatt's system.) |
Hi there I've run ESL57's both single, stacked and with sequerra ribbons with Quicksilver 8417's. I got much better bottom end when I ran them as a single pair by putting them on stands about 18-20" off the floor and tilted up so the panels were more vertical. Better bottom end and much larger soundstage. You might like to try this. |
Post removed |
Yes, the Quad 57’s impedance curve clearly plays havoc with a lot of amplifiers, but I don’t understand the reasoning behind some of explanations for this. And at least one very basic factor about the amplifier/speaker interface is missing from this discussion-and quite ironically at that, as I will later show. First, just today, a friend of a friend-both owning 57s-bought a used Atmasphere S30, being impressed with the overall design. But he found that it was a “terrible match” and gave nonexistent bass with the 57s. Second, Atmasphere says that Bass is hard to get with transistors and Quads because many transistor amps will double power as you cut the load impedance in half. That’s quite true with amps like the Pass Labs XA series. And he says that the Quad has high impedances in the bass region and low impedances in the treble/HF region. "Most transistor amps would thus likely play the Quad 57s with weak bass and too much highs." Okay, that’s certainly believable criticism. But then he says that the 57’s impedance in the bass is “…well in excess of 45 ohms. So transistor amplifiers cannot make power in the bass, while at the same time they make too much in the highs, where the impedance of the speaker is down to 4 ohms.” Well in excess of 45 ohms in the bass? According to this obviously credible source, the 57’s impedance doesn’t rise any higher than about 33 ohms, and that’s at about 80Hz. http://www.quadesl.com/quad_main.html Last but not least: Damping factor and bass response. In my discussions with Ralph Karsten he imparted some very valuable info on this subject. He said that high efficiency speakers (like my 97db 7 to 8 ohm two-way Radian 745 Neo/Be horn drivers/ GPA Altec 416-8B midwoofers) want a low damping factor amp-no more than about 20-like the S30 amp, otherwise midbass and certainly low bass will sound lean. I don’t know what the S30’s DF is but low feedback amps tend to have low DF. Exceptions are amps like the Pass Labs XA30.5 and the new XA30.8.The Audio Research VSi75 has 4db feedback and ~ 4 DF. http://www.arcdb.ws/VSI75/VSI75.html But the S30 has only 2db feedback, so either amp would likely be a good match for my efficiency and all but flat impedance speakers. Conversely, ESLs in general are at least 10db less sensitive than my hybrid horns.Soundlab recommends a 100 wpc amp for their smallest full range ESL model. More powerful amps tend to have lower output impedance and thus higher damping factors, again like the Pass Labs XA series, with typical DFs of 150. https://en.wikipedia.org/wiki/Damping_factor OTOH, the 57s, though probably even less sensitive that the Soundlabs (except perhaps for the Piquet-rebuilt 57s, says http://www.audiocircle.com/index.php?topic=96271.0 ) can’t take lots of power without risk of arcing and damage. But, like the Soundlabs, they apparently need an amp with a much higher DF-that is, with a relatively high impedance, like many tube amps have.Damping factors of probably at least 20, which is probably 10 times higher than the Atmasphere S30. No wonder my friend’s friend got no bass from the S30 driving his 57s. But again, the amp must have BOTH ample damping factor and a high impedance output; not so easy to find in most solid-state amps. But guess what? The First Watt J2, though a solid-state, dc coupled, all JFET, Class A amp, with relatively low feedback indeed has a damping factor of 20. http://www.firstwatt.com/pdf/prod_j2_man.pdf And John Atkinson’s measurements http://www.stereophile.com/content/first-watt-j2-power-amplifier-measurements#Uh4sMghuhX5g34Rh.97 Accordingly, the J2 not only prefers high impedance loads, but when the 57’s impedance drops to about 3 ohms at ~ 18kHz, the J2’s output power drops to about half what it is at 8 ohms. Thus, wouldn’t that keep the 57s from sounding bright? All told, the J2 looks to be one of the very few solid-state amps that could actually play nice with the Quad 57s. If not, please explain.
|
They thing they don't tell you is what happens when driving higher impedances, like you see in the Quad ESL 57. |
A "properly" designed tube amp is really the way to go with the ESL 57. Roger Modjeski designed his Music Reference RM-10 amp specifically for the 57, using the speaker as the amp's load during development. If one can't use such a tube amp, perhaps another speaker would be a better choice. The speaker/amplifier symbiotic relationship is, along with the cartridge/tonearm one, the most influential in the chain. |
with a much higher DF-that is, with a relatively high impedance, like many tube amps have.Damping factor has little to do with how much bass you get! - but it has a lot to do with how much distortion you get with that bass. IOW low damping factor does not translate to lack of bass! Nelson Pass has an interesting article on one of his sites regarding this fact and demonstrated it at the RMAF using a set of open baffle loudspeakers and his amps a few years ago. In the case of a Quad 57, the S-30 has a much higher damping factor in the bass simply due to the impedance of the speaker. Its far more likely that your friend simply hooked the S-30 up out of phase. The problem you run into with tube amps on ESLs in general is not that they don't make bass, its whether or not the amp can make highs on the speaker due to the low impedance at high frequencies. The J2 does look like an amp that might do the job. When using a solid state amp on the Quad 57s, due to the impedance curve an amplifier that does not use loop negative feedback is likely able to do a better job than one that does use feedback. |
I've run ESL57's both single, stacked and with sequerra ribbons Dover, what are the Sequerra ribbons? Did you get these tweeters from here? http://www.sequerra.com/common/data/ribbon.html http://www.sequerra.com/ Are they still available? How did you integrate them with your Quad 57s? Did you disconnect the Quads’ own tweeters? Is entire system still passively crossed-or are stock Quad 57s actually crossoverless? Or did you now have to bi or tri-amp it all? Sequerra ribbons $$$? If beyond my budget, are there one or more tweeters that you might recommend?
|
I had Quad 57s and there was precious little bass until I put them on Arcici stands, removed the metal grills and upgraded the power cords. Then the Quads had excellent solid bass and were much more open and dynamic. I drove the Quads with modded Dynaco tube mono amps with regulated everything and aircraft grade tubes. At that point in time I had an isolated air bearing Maplenoll turntable with special 50 lb platter and 500 feet of air tubing and two air flow buffers. |
Amplifier Requirements This speaker is designed for operation with a standard Quad II amplifier and it is essential for both proper performance and reliability that it be used only with this amplifier or one designed specifically for the Quad loudspeaker. Speaker Specs https://mniec.files.wordpress.com/2010/06/quad-esl57-user-manual-page0010.jpg The above information has been taken from the Quad owners manual. Also it is my opinion, based on current personal experience, that anyone needing to add external tweeters to this speaker is doing so because of one or more of the following factors, which could be affecting the setup and causing problems. This assumes no hearing issues. - Physical condition of the speakers themselves. (are they functioning at 70% ) - Room size and positioning - Room Dampening - Humidity levels - very dry or humid. Especially if speakers are not kept "on" plugged in. - Amplifier compatibility ***************************************** For information here is the Quad II amp http://www.quad-hifi.co.uk/product-detail.php?pid=20 ***************************************** Hi Ajant - how did your audition go ? I never heard back. Cheers |
I, like Geoff, took the metal grilles off my Quads. Big improvement! I also removed the rear grilles with their attached absorptive pads, about which Quad owners have mixed opinions. If done, the speakers must be further from the wall behind them. One fanatic Quad owner I new went so far as to removed the plastic dust covers from his 57's! That leaves the ESL drivers free to attract dust, leading to their early demise it is said. |
bdp24 I, like Geoff, took the metal grilles off my Quads. Big improvement! I also removed the rear grilles with their attached absorptive pads, about which Quad owners have mixed opinions. If done, the speakers must be further from the wall behind them. One fanatic Quad owner I new went so far as to removed the plastic dust covers from his 57’s! That leaves the ESL drivers free to attract dust, leading to their early demise it is said. I also removed the plastic dust covers. Another good idea. And guess what? Never had a problem with dust and there was no early demise. Or later demise. When operating correctly sans grills and plastic dust covers the Quads can be placed so there is no toe in, which is really best, and the distance between the speakers can get determined precisely using the speaker set up track on the XLO Test CD. |
ct0517 Sadly, not yet. We were on for a Friday early last month until Jim got pneumonia. He said he was still very sick almost two weeks later. I’ll email him this week, see how he’s feeling and maybe we can set up for something this month, I hope. Btw, I emailed Nelson Pass attaching the Quad’s impedance curve http://www.quadesl.com/quad_main.html and asked about using the First Watt J2 amp. His reply seemed a bit vague, though he did say “the Quad wouldn’t hurt it” the J2. Later when I, more wisely, asked that of all his designs which First Watt or Pass Labs amp would make the Quads perform at their best in every way, he said probably the F6, F7 and J2. But he said what he really wanted to know for himself was how the SIT-1and SIT-2 would work with the Quads. Of course, that would be an intriguing session, but those two amps are somewhat more than I’d want to spend and run a lot hotter than the J2-which is hot enough for me. |
His reply seemed a bit vague, though he did say “the Quad wouldn’t hurt it” the J2. Hi Ajant that comment implies to me that he hasn't actually heard the J2 with the Quads? Let me know how it goes. I stand by my thoughts that one can insert whatever amp/s you want in and try them - but this speaker is unique and therefore you should either buy the intended amp from Quad, or one that was built specifically for them, like Roger Modjeski's RM10. The only one I am familiar with. What reinforces this for me is that I have had amps that have worked well with my Acoustat and B&W Electrostatic Speakers - but did not work with the 57. If you have to add an external tweeter to the 57, the amp in my books, isn't doing the job, or there is a problem with the speaker. The only amp I have personally owned that worked well with the 57, and also my Acoustat and B&W Electrostats, was my Music Reference RM9. And this is what put me into research mode, and on the path where I ended up with the RM10 years ago. Also - The heat from the RM10 amp is not that bad and imo any SS Amp Class A needed will also get hot. Cheers |
ct0517 (Chris) uses a Music Reference RM10 with his Quad 57's, and he will tell you how well that particular amp works with the speaker. Roger Modjeski designed the amp with the Quad as it's load! Though a tube amp, Roger designs his amps with reliability as a high priority, and for low maintenance requirements and considerations. His amps provide very long tube life, unlike many of the more well-known and owned contemporary tube amps. |
The Quad 57's are an anomaly. Both physically and electronically. Electronically and on the subject of this thread - I am not familiar with any SS amplifier design, that is able to concentrate its energies for bass, which sucks up the most power - at 30 ohms+, which the 57's need. And why should they be designed this way if the Quad 57's are an anomaly? There are no business reasons that I can see for a manufacturer to build an amp this way. @Atmasphere, (other amp manufacturers) if you are still following this thread - does such a SS Amp beast exist, and or can a SS amp be built that is like the Quad II, ......I mean if someone really wanted to ? I believe.... and its theory only, because I am not an EE, but, I think that SS amps get smoked and dizzy, start oscillating (what's the technical word I am looking for?) by the 30+ ohms bass requirement. Many Class A seem to be designed from 8 ohms, to double down in watts and half up in watts. If designed for the Quad 57, from 8 ohms they should be doubling up to allow the amp to cruise and not be taxed. |
had to find this on my hard drive. imo - the reason Acoustat will work with good Solid State. very different requirements from the 57 graph that Ajant posted. https://photos.google.com/search/_tra_/photo/AF1QipPLvcWgX83WAzSfmba9hKxw7vtl2Jx0HOEOzS4G |
@Atmasphere, (other amp manufacturers)Actually a solid state amp isn’t ’taxed’ at least not in the way that it would be working hard. The problem is that the impedance is high so the amp can’t make the required power. This is not a matter of debate, its simple physics. Put another way- if a solid state amp makes 100 watts into 8 ohms its likely to only make about 25 watts in the bass region. Contrast that with the possibility of the same amp making over 200 watts in the upper treble and you start to see what the problem is! The problem here is that generally speaking, most solid state amps are intended to be a ’voltage source’, which means that regardless of the speaker load they can make the same voltage. This works great on a lot of box speakers where the impedance often represents the speaker efficiency as well- for example if there is a bass resonance (which will have a higher impedance) in a box speaker the amp won’t make much power and so does not exacerbate the resonance. But the Quad does not have a high impedance in the bass due to resonance- it is there as a function of it basically being a capacitor (that being the ’electrostatic’ part of the way the speaker works). IOW, its impedance curve is not an efficiency curve like it is with many box speakers!! I can’t emphasize that enough- ESLs are fundamentally different in this regard. What is needed is an amp that can make about the same power regardless of impedance, rather than the same voltage. This is so simply because the efficiency of the speaker is fairly constant over its range. Obviously this is not served by most transistor amps (although many owners that use solid state tend to back the speaker up against the rear wall in order to get it to play bass; that works in a way but you get a ’one-note’ bass because what you are really doing is reinforcing the bass at one frequency only). Quads need to be out in the room a bit in order to play properly and not be disturbed by the wall behind them- 5-6 feet is about the minimum (this is true of any panel speaker). The only amps I know of that can produce anything like a constant power characteristic are all tube amps. However, if you **do** use solid state, its quite evident that loop feedback should not be used in the amp’s design!! The lack of feedback will steer the amplifier’s voltage response to be more like that of a ’power source’ and although it will fall short of that, it will be much better than an amp that does use feedback. Obviously this severely limits the field! This is why tubes and ESLs have really had a liking for each other going back to the day ESLs were invented. If you want to know more about this see http://www.atma-sphere.com/Resources/Paradigms_in_Amplifier_Design.php The article at the link defines the differences between what is called the "Voltage Paradigm’ (the rules required if the amp is to be a ’voltage source’) as opposed to the ’Power Paradigm’ (where the investment in the speaker is best served by an amplifier that behaves as a ’power source’). The Quad, and any ESL for that matter, are Power Paradigm devices; its the rather simple physics of the matter which is why they don’t fit into the voltage model (which was first proposed and championed by MacIntosh and ElectroVoice back in the late 1950s). Most of us have grown up in the age where the Voltage Paradigm is the dominant model of amplifier and speaker interface (so we may have trouble understanding how there could be any other way to drive a speaker); that is why threads like this exist where people are trying to sort out how to work with the outliers like a Quad. The fact is that if you attempt to make solid state (Voltage Paradigm) work with a Quad (Power Paradigm) you will wind up with a tonal anomaly (which is what the Voltage Paradigm was supposed to eliminate FWIW...): brightness and a lack of bass. BTW this is easily measured as well as heard. I’ve had it suggested to me in the past that the extra bass tubes have on Quads is not correct but the measurements don’t confirm that. What they **do** confirm is that Quads can play bass quite nicely if you have the right amp on them! |
An extremely educational thread, even for non-Quad ESL owners like me. So based on the facts presented earlier, it appears that the ESL speakers do not follow the same amplifier requirements when compared to "dynamic" or cone design speakers. In other words, even though almost ALL speakers can have significant fluctuations in impedance over the frequency response curve, the demands on the partnering amplifiers are significantly different based on the speaker design. As another example, I've posted a picture of the impedance curve for my (ESL) speakers - Martin Logan Aerious i, and according to the preceeding discussion, my "voltage source" amplifier is not handling the the peaks - midbass and midrange, properly. So by extension, this begs the question that every ESL speaker manufacturer should either develop and sell a matching amplifier or almost all ESL speaker owners out there will not be able to take full potential of their investments. The latter would be a sad compromise. http://www.stereophile.com/images/archivesart/MLAFIG1.jpg |
Thanks Atmasphere points noted and helpful IOW, its impedance curve is not an efficiency curve like it is with many box speakers!! I can’t emphasize that enough- ESLs are fundamentally different in this regard. For the bass they play its very good. If the music one listens to is satisfied by this ....great. However most I know have left the 57’s due to the bottom octave missing and lack of HF’s. The lack of HF’s being from what I mentioned earlier. But this thread is 8 years old :^) Jan 2017 - For speakers which are not full range, implying bottom octave missing or bottom couple octaves much lower in db on full range music; subs today take care of it very nicely and integrate well. But they do need to be setup well. Kalali Kalali Well Quad do and Acoustat used to as well. IMO, anyone looking to buy really "any" expensive speaker should be calling the company direct and asking them what amps are in their test labs. Likewise if buying amps - find out what speakers they are using. I have been making these calls in past. |