martin logan csl

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi all
I have not posted in a while and with the excellent advice I have gleaned from this forum I have managed to produce a set of functioning esl's, they might even survive my partners disapproving looks. However I have been donated some csl panels that have very low distorted output. has anybody any idea of the membrane bias and any tips as to the replacement of the membrane. it looks straightforward but I am guessing there are some snags
ahead
THANKS IN ADVANCE FOR YOUR GUIDANCE
 
RX for CLS

I had a pair od CLS's with the same problems that you discribe. You need to borrow a high voltage probe to use with your DVM and check the output of the high voltage supplies. My guess is that they will be fine (can't recall for sure but I think they were around 4Kv). Next you need to measure the voltage on the conductive side of the diaphragm. In my case that was where the problem was. I uses some conductive compound to fix the contact point between the diaphragm and the supply lead wire. That solve the trouble. Good luck and I hope that this helps. Best regards Moray James.
 
Thanks Moray
The panels came as is as they were dumped in the back office of a dealer, they were obviously faulty in some way. It is only the stators and wires, but I tested them with my 6-7k supply. I thought that it might be leakage from the membrane to the stators as I have had that problem before with my homebrew jobs but your diagnosis sounds more likely as something would have have to have changed on the edges of the units and the stators and coating are in good condition.

Many thanks
flodge
 
just in case

Flodge: just in case you do have to rebuild these panels and install new diaphragms this may hepl. The acrylic transfer adhesive use to assemble these panels is a bugger to get off the stators and seperating the stators is very difficult. Someone at this forum (think it was I_Forgot) has said that a product called "Goo Be Gone" will wolk like a charm to soften the adhesive. This is a solvent sold to remove sticky residue from lables. I am sure that there is some similar product availabel in the UK. check it out
Something that I did to check out the supply to diaphragm connection was this. I used a small fine guage wire to connect the high voltage supply directly to the conductive side of the diphragm and then measured to see if the voltage showed up at the lead wire of the panel. In this way I was able to confirm that the contact point was the problem. Should the charge migrate out onto the diaphragm and the contact point is also good then you need to look into the supply itself. good luck and have fun. By the way CLS panels work and sound excellent when driven by Acoustat interfaces as both supply voltage and step up transformers are well suited. Best regards Moray James.
 
Thanks moray
I have been out after work and bought some anti-goo liquid and was just about to test it on some of the adhesive. Your tip about lighter diaphram material and the rebuild on the diyhi forum is useful as I have quite a bit of the very light stuff left over. The reports on these speakers seem to be quite confused, some luv em to bits and others say they are harsh in the upper mid-band. A lighter diaphram can accentuate that tendecy I believe, without excellent trannies. But I will of course rebuild them, it is a divine duty, but I am getting a bit tired of wife dodging with them.
regards
flodge
 
I have been smitten by the CLS's for a very very long time, still aim to aquire a nice set one day, maybe two of the anniversary models.
Downsides especially for the CLS's is the high power they require, the bass does not extend that deep, and there is an early end to maximum SPL.
If that is no bother, the Martin Logans are divine, they are divine in my book.
 
whatever next

Well the anti-goo helped and a fine artists palette knife made fairly easy work of getting the stators apart with no damage whatsoever to the membrane. The spacers which are horizontal are eased off with the goo juice and delicate use of a fine but long steel rule. I must admit I was a little surprised by some of the things I found. bias wire wrapped around copper tape not actually soldered (which is fairly easy) in an arrangement just waiting to fail, or perhaps the voltage is so high this makes little difference, but it just seems cheap. The membrane is sound but it is fairly heavy stuff and under good tension. What do you think, count my blessings and check the resistance and perhaps give it a short coat of licron, or, change the membrane. The spacers are designed to suggest nodal arrangements but are obviously intended to make the profile possible and a pattern can be made to replace such things with modern equivalents. As I have just completed some hybrids that I am pretty happy about, this work can be interesting and fun. The problem with the lighter membrane material is getting the tension and it is the jigs which take the time. The foam tape they used on these things seems very dense and stable , much better than the stuff I have used on mine, but they do seem to charge silly money for such mundane stuff as sticky tape.
Anyway back to the wine
see you
 
Go fo r it!

Since you have had such good success so far I would caarry on and reskin them with six micron film. You have the licron so no worries there. The lack of solder at the high voltage connect is not how I would do it but not a big deal with these voltages so don't be too hard on ML.
I found that the Acoustat interfaces sounded much better than the stock ML units. If you don't ask these panels to make bass or play at live levels they are happier all around.
Just some food for thought. With curved panels like these you will have to insure that the new diaphragms are primarily tensioned in the vertical plane to help create the curved in the diaphragm. As with all curved stators the diaphragm will want to colapse toward the back stator and so you will see that ML (and everybody else who does this) have increased the thickness of the stator spacers on the rear side of the panel. Since the diaphragm is always on the verge of colapse toward the rear this trick insures there is extra space on the back side to prevent the diaphragm from getting stuck to the rear stator. While this does do what ML want it to do (prevent colapse) it also makes the diaphragm drive non linear as it is no longer physically balanced. This also eats up some panel efficiency. Further with such a wide diaphragm there is the tendency to go unstable as there is not enough tension in the horizontal plane to keep the diaphragm fully stable and so the use of horizontal spacers is also required (lots of compromize).
Where I am going with this is that if you devide the stators into vertical sections of say 3 inch widths you will find that you can probably reduce the thickness of the rear stator spacers to probably half of the difference between the stock rear spacer and the stock front spacer (almost aproach the thickness of the front spacer). You will be able to apply more horizontal tension to the diphragm to increase efficiency and the narrow panel sections will insure stability. The new panel should be more linear and more efficient.
One other thing is that since the CLS is not all that tall you may want to stop any vertical divisions about an inch short of the top of the panel to leave a horizontal section of diaphragm which is narrow and will naturally provide you with some vertical dispersion. This is an old trick that works to prevent that head out of the water effect when someone tall stands up and finds thier ears off of the direct radiation of the panel and looses all top end response. Would be worth giving this a try as you can always rebuild it to stock configuration if you are not pleased.
Something else you can do at this point in time (before you remove the stock diaphragms) is to have a good look at the diaphragm and see if there are any arc holes anywhere on the mylar. This will tell you where the stator dielectric has broken down. Since you have the panels apart you can solvent clean them (inside surface) and apply a few coats of suitable insulation to the inside of both the front and back stators in the general area of any arc holes that you may have found on the diaphragms. This way you will be rebuilding with good as new stators. The coating on the stators is Nylon so make sure that any spray insulation you choose to use adheres to Nylon.
Hope that some of this can be useful to you or to others like you with an ols set of CLS's that need a little Rx. Best regards Moray James.
 
Linearity

Hi,

@Moray.
I do agree with most of Your above posting, but not everything within.

When we assume a homogenous field between the stators -and this should be the case as long as the stator-stator-distance is constant and the holes in the stators are sufficiently small- then the forces of the electrostatic motor are linear and equal at any point between the stators. The only imbalance is then solely due to different stiffness of the membrane´s suspension compliance. But this plays a minor role as long as diaphragm travel is kept short -which is the case in all curved designs. My own distortion measurements didn´t show any noticeable rise in figures against flat panels as long as diaphragm travel was short and lower transition frequency not too low.

Another point I don´t agree fully, regards the horizontal diahragm tension. In a curved design there should be the least possible horizontal tension.
As long as the relationship of shortest free vibrating distance of the diaphragm falls in the range 70..100 times the d/s it does not matter in which direction the tension is set. So only use vertical tension in a curved design. The requirement of additional spacers applies to all panels that don´t fulfill the above mentioned ratio. So the only minor drawback could be a slightly raised blind-capacitance (the area below the spacers) when the area of the horizontal spacers is greater than the area that vertical spacers would cover for the same stability. Doing it this way -preferably in conjunction with very high mechanical tension- You can indeed have nearly identical spacer thickness. I use a difference of just 0.1mm for gaps of 1.0 resp 1.1mm and very high tension. Up till now I had no problems at all with stability even when driving the panels to earshattering levels --and I mean ear Shattering! ;-)

jauu
Calvin
 
interesting information!

Calvin: thanks for the interesting heads up on your findings with curved panels. I think that the key with curved panels is in the stretching process of the diaphragm. Curved panels do present the diy builder with more complex jigs and I suppose a longer slower learning curve. My thoughts about vertical segmentation were directed toward helping to minimize the expanse of unclamped diaphragm (and so stability)as well as to help reduce the difference between front to back stator thickness. It seems that this issue (assuming we don't ask the panel to make bass(a good idea anyway)) is not the problem that I would have thought given your findings and that is very good to know for those who want to buld curved panels.
I would be interested to know what amount of curve you find is possible and also approx. how much linear travel you think that the diaphragm has with such a curve? Do you think that the diaphragm is stable to as much as one stator thickness peak to peak in a panel that you build? This is probably hard to establish but I would be interested in you estimate. What is the approximate area of one of your panels and how low do you drive them?
I very much appreciate you providing this information and the opportunity that such discussions afford. Being able to learn from others experience is of enormous value to diy builders. Given your personal experience with curved ESL could you comment on how much wider you feel that the sweet spot is with a curver stator versus a flat panel of the same dimension and also do you believe that the sweet spot with a curved panel is as good as that of a flat panel? Thank for your comments and thoughts. Best regards Moray James.
 
curves

Hi,

stretching is indeed the keyword. With curved panels imo only mechanical stretching is senseful. I prefer a tension that gives a Fs well above 100Hz,which is way above whats possible with thermal stretching. You can´t really expect curved panels to produce bass (neither can You with flat panels. Its more of a pronounced resonance and a suck-out in the lower mids than real bass). So diaphragm/stator distance should be kept as small as needed for the supposed frequency range to give maximum efficiency. Design for highest efficiency! It´s much better to match efficiency to a woofer by using a lower transformer ratio than anything else! Don´t expect to get much lower than 200Hz, even with panels of considerable width.
A streching jig is quite easy to build, because You only need the variability in vertical direction. My frame works for flat panels as well as curved ones. Its just an stable outer frame to which variabel inner elements are connected. These Elements can be ´pulled´ outside by threaded rods and nuts. The diaphragm is glued to the inner elements with double side sticky tape. To change between flat and curved style I just change the flat vertical elements against curved ones that have identical curvature as the panels stator. I use just a slight degree of curvature. 20° to 30°. The dispersion broadens up a bit but not very much. Directivity is still very high, but the panels play a bit more pleasant when casual listening. Higher degrees of curvature will make stretching and installation of the diaphragm harder and it won´t really give drastically better dispersion figures. The big advantage of the curvature is a higher mechanical rigidity of the stators, which leads to less resonance, less rattle and a optically very pleasing appearance without the need for additional supporting structures.

With regard to diaphragm stability I can only add my personal experiences. One of my panels is sized 25x125cm with a d/s of 1.0mm to the front and 1.1mm to the back. Horizontal spacers are added app. every 80mm. The diaphragm is stretched very hard and lenghtened thereby by app. 2%. When glued to the spacers this gives a resonance of app 150Hz. This provides for excellent stability. As a worst case test I played tracks from Cincinnatti Pops CDs fullrange(!). My Rotel 980, that can push out more than 300W blew its fuse, but the panel worked flawlessly and stable without the diaphragm touching the stator once. For the intended frequency range well above 200Hz You have plenty of stability headroom left.
Even though You have to correct for the falling frequency response -due to phase cancellation- between app. 200Hz and 600Hz with an equalizing filter.
The German manufacurer PAS, who sell similar panels for PA uses claims for a panel of this size (but with 1.5mm d/s) an frequency range of 180-20.000Hz, crossover point of 200Hz@24dB/oct, an efficiency of 90dB/2.83V/4m and a peak SPL of 113dB/4m@3%THD.

jauu
Calvin
 
similar challenge to me

Hi,

very interesting thread to me, since I decided to replace the foil of my ML Sequel's.
I already removed the old ones. It was not so easy, I didn't use any sover for the glue and of course the original membran got damaged by this action, but anyway I'm going to replace.
I realized ML seemd to use some graphit stuff as conductment agent, like some black powder (applied on an atomar level, haha).
I'll use EC coating and thinner foils.

I would like to renew all, or least the white double-sided sticky tape around the inner side of the panels, but unfortunately I didn't find a supplier. Maybe someone here can provide me informations regarding this.

Eventually I'll reuse the old tape (It's still sticky).

Hello Calvin,
your rig for tensioning the foil sounds very interesting. I tried to do by hands, but i'ts not the right way. It's not doable accurately enough.
I would like to contact you personaly, since I live next to you.
Feel free to contact me by pm, I would be very glad.

Olaf
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.