It amazes me that as many times as it's been mentioned in this thread about not using detergents and brushes, people still do it despite all of the warnings.
Also, has it ever crossed anyone's mind that the improvements you hear after rinsing the panels could simply be the fact that you're creating new, fresh contact points on the wires going into the front connector for the stators once you reassemble the speakers? If those connections were slightly loose or corroded, simply tightening those connections could make a world of difference (depending on the severity on said connections).
My SL3's were built in January of 1998. They have sat 3 years unused. Before firing up again recently, I removed the rear panel just to have a look at everything and "re-seat" all of the connections as well as reconnect the crossovers as I had bypassed them a few years back to run the SL3's fully active with a digital crossover and a couple amps.
Long story short, the SL3's are back to factory stock, are 22 years old, are all original, and sound great. The panels sound to be operating at 100% with zero loss of output or top-end "sparkle".
It might be wise, short of going to the extreme last ditch effort of hosing down the panels, to just go through the speakers and re-seat all of the screw down connections, both inside the woofer enclosure where the crossover/power supply is as well as above the woofer on the front baffle where the ESL connects to the terminal strip there. You might just be surprised at the results and skip the risk of possible panel damage.
Basically, this is the same concept of getting behind your stereo equipment once a year or so and re-seating all of the interconnects on the gear as to not have corrosion/oxidation build-up and have fresh clean connections.
I was thinking the exact same thing, after reading about separation issues I became a bit skeptical. But since I bought my Aerius used and I have owned other electrostats, I can tell there is a slight loss in sound overall and that's what brought me here.
I have cleaned many JansZen panels and I used water on q-tips to clean in the pleats, they are made completely different in every way and my comparison with them may not be relative to the sound I'm getting from the Aerius. I've read quite a bit but there's no real side by side test, I'm going on what I heard a year ago before my beloved electrostats were destroyed in a fire.
Although the film looks clean I felt the smartest thing to do would be to set a q-tips and gently twist it back and forth between my fingers to see exactly how much dirt comes off into the tip.
I did 4 or more holes per swab and the first 2 had very little dirt, a few more does show some build up as you can see it's very black. I'm not sure what the coating is made of on these. They are 30 years old and have been in AZ their entire life and look to be in excellent condition so I doubted they were weighed down by this slight build up. I can't see this tiny bit of film causing sound to degrade much.
Some quick tests showed a roll off around 15KHz but held on in a slow decay. I felt this was probably typical judging by the size of the panel.
The JansZen panels are only 5"x5" and like I said are a completely different design which uses twice the mylar in width due to it being pleated and uses wires running across the panels. The picture is a JansZen designed panel from an old Infinity Electrostat panel made in 1971 (only 4 panels survived the fire, 24 others melted). The other JansZen panels were made in 1959 and 1967 yet they tested flat from 660Hz to 24KHz (before and after cleaning, it made no difference in perceived output.
Has anyone tested their panels using an RTA while bypassing the crossover to find the panels overall response? I would expect these much larger panels to be able to go much lower than the JansZen and wonder why the crossover points is set to 550Hz?
One more thing I noticed is the JansZens have a higher efficiency, with a total area of around 300sq inches per side (using 14 panels) is very close to the area of the Aerius panel, the output of the JansZen group was above the efficiency of a pair of Klipsch Heresy. So that kinda threw me a bit when I connected the MLs to my same system (old amps all survived the fire with only cracked plastic speaker connectors on the positive output of 1 amp).
The more I'm reading the more I'm questioning doing the wash. I do have some of the double sided tape (commonly used to hold phone glass in place) it's very cheap found on AliExpress for $2 a roll and comes in a variety of widths and thickness. The separation seems like it has affect the sound greatly once tension is lost? I envisioned a clamping set up that used 2 long runners on each side with multiple c or vise grips clamps keeping the seam together but then I thought about how foam tape will become crispy over time and smashing it tightly might just destroy the tape completely.
So there's many factors to consider before jumping in and doing a wash like this, not knowing exactly how they should sound when new is what keeps me from acting on them. I'm going to be bypassing the crossover circuit completely and using an active crossover. From my DIY tests I did while looking for a perfect way to integrate a woofer with the panels I found using an overlap of 1 octave with the woofers and natural roll off of the panels worked seamlessly for me. From what I've read, ML uses the sharpest (24db) slope crossover made to separate the woofer from the panel. This makes for a very distinct change in tone as it transitions. I'm not sure if they are still using that same sharp slope in newer stuff or if they came to the same conclusion I did many years ago.
What worked best for me was finding a woofer that had a dip in its response at the same point the panels rolled off, I used a 6db slope which has the least chance of ending up out of phase. Another thing I found out was how a BBE maximizer can perform magic with Electrostats. I leave the controls flat and it corrected the phase after my dual 31 band EQ. It lifts the sound off the floor and makes the mids sound as life like as you are there. Well at least it did with the JansZen set up.
I'm changing out the woofers with some Mirage 8" subs that I tested to be flat from 30Hz to over 2KHz.
If they don't test we'll in the ML cabinets I'm testing some 10" Monitor Audio subs that tested flat from 24Hz to over 2KHz with a dip around 700Hz (of course they have a different timbre and I would not use them above 1KHz because of it).
I have read about decoupling the woofers which makes very good sense since the constant vibration could be a problem for the deteriorating tape connection on the sides. The construction is solid but the design will amplify the vibrations as they reach up along the side panels and the mass becomes less at the top. I made my DIYs without sides and mounted all panels together using stotted wood rails mounted directly to the woofer cabinet face (like a sunflower) with the woofer in the middle and all panels mounted on the same flat plane with the panels mounted near the front of the woofers motor gap (within an inch).
The curved design has made me conclude that the reflected sound will always be inaccurately timed due to reflection of the waves crossing from the backside and will test with a piece of insulation behind the panel to stop the rear waves. It will deaden the output even more but I think there's a real possibility of achieving a more coherent sound allowing the speakers to be moved closer to the back wall as well.
I'm open to any suggestions and if anyone can say that I'd benefit from washing them or not, based on the light layer of dirt on the swab. Maybe others can do the test and see how it compares?
I thank all that have contributed here, there seems to be a lot of conflicting info on this topic and reading about separation is really the main reason I'm going to try and avoid it until it's my only choice or I'll sell them. I'm not going to spend a bunch of $ on speakers that I know I can make for half the price of new panels and have a better performer but will not be as appealing on the eye as the MLs and will require many days of work to complete.
I think I can get better results by simply bypassing the crossovers and working with the active crossover. I have to add that I have tried using a number of DSP units and found them all to be inferior to an analog EQ and active analog crossover. Difficult to use and loss of detail were very noticable.
Ill report back once I figure out how to get them apart. I plan on doing this today and be testing by this evening if things go as planned.
Thanks for reading!
