Panel rebuild guide?

[Jazzman53]

Hi all,

Occasionally someone posts a thread about an older speaker with low volume and expired warranty, and they end up purchasing new panels ($$ ouch).

Panels can be rebuilt if the stators aren't damaged, and with all the DIY'ers on this forum, I'm a little surprised that someone hasn't posted a step-by-step panel rebuild guide (I searched and didn't find one).

I would not advise anyone who isn't mechanically inclined to attempt replacing a diaphragm. But for those who are so inclined, it's certainly do-able if one understands the panel's construction and function, has a procedure that satisfies that function, and has the correct materials.

Although I don't own ML speakers and I've never repaired one, I do have extensive experience building ESLs and I would be happy to assist in creating a panel rebuild guide-- I just need a collaborator with a panel for rebuild who can provide photos and fill in some gaps in my knowledge of ML panel construction.

I know that ML uses 12-micron Mylar for the diaphragms and I've deduced most of the panel construction from the videos on their website. I'm unsure about these (7) items:

1) Thickness of the clear adhesive tape in the spar/stator bond line and the spar/diaphragm bond line. (I'm guessing it's 0.010")

2) Adhesive tape definition in the stator/diaphragm bond line at side edges.
Also; is the bond line 1 layer of a thick tape only, or is it a plastic spacer shim with a thin tape on both sides? (I'm guessing it's 1/16" VHB urethane foam only)

3) Same as question 2, except for bond lines at upper/lower panel ends

4) Copper foil bias supply strip connection:
Is the foil applied [adhesive side down) directly onto the diaphragm, or is it applied on the opposite stator, to contact copper side to diaphragm when the two stators are mated together? (either way works.... I'm just curious)

5) How is the bias supply wire attached to the copper foil strip?
(I would solder it in any case, but I'm curious how ML does it)

6) On the assembled panel; is there any kind of tape wrapped over the edges of the mated stators?

7) How are the transformer leads attached to the stators (solder or mechanical)?

Notes / Observations:
ER Audio (Australia) offers rebuild kits for ML panels, starting at around $70, which would be significantly less cost than buying materials in bulk, separately.

I gleaned from a thread on another forum that ER Audio recommends using a spring type weight scale to tension the diaphragm but [not knowing how ER derived the pull weight] I would take a deflection measurement from an intact panel to gauge the tension (or at least verify ER's pull weight).

I would build a simple wooden jig to stretch diaphragms for two panels concurrently (you would need to do both speakers, even if only one is bad, and tension them identically). And I would provide a drawing for this jig.

A very important ESL design consideration is the diaphragm's drum-head resonance, which is solely determined by it's tension. ML (or anyone else) would have to tune their passive crossover and EQ filers to accommodate a specific diaphragm resonance frequency, and then build every panel with that same resonance, consistently.

Accordingly; a replacement diaphragm must be tensioned to produce ML's design resonance. Otherwise, the panel would not have correct frequency balance / tuning.

This is less complicated than it sounds: Before dissembling the panel, the diaphragm's tension could be measured (in the center of the largest span between spars) using a simple deflection gauge (i.e. a match stick, a small weight, and a scale). The replacement diaphragm would then be tensioned to the same deflection (or equivalent spring-scale pull weight).

After assembly ML sandwiches the completed panels between two caul plates and vacuum bags them to consolidate the bond lines. This would save time and effort if one were equipped to do it (I could explain how) but it's not a necessity. Manual pressure applied long all bond lines is would be sufficient.

I'm retired with nothing better to do, so If I get positive feedback on this proposal and someone steps up with an ML panel to collaborate on a rebuild, I will be happy to assist in creating a step-by-step panel rebuild guide.

Jazzman
https://jazzman-esl-page.blogspot.com

 

[Brandon Hartwick]

I'd be very excited to see this happen, unfortunately I have no bad panels to give away. I don't think it would be hard to find someone willing to donate a dead panel or 2 seeing how often we see the old ones go bad.

 

[Jazzman53]

I'd be very excited to see this happen, unfortunately I have no bad panels to give away. I don't think it would be hard to find someone willing to donate a dead panel or 2 seeing how often we see the old ones go bad.

I would do it of course, but I prefer empowering others.

I think it would be better if the panel owner (hopefully mechanically inclined) does the actual panel rebuild, with me merely advising and documenting.

This would demonstrate that you don't have to be an experienced ESL builder to rebuild a panel, and hopefully inspire others.

 

[Robert D]

One other option is to have Russ Knotts try a rebuild for you. He might be a good source for information on this online help guide. He is usually very willing to share his knowledge.

http://www.eslrepair.com/Contact_us/index.html

 

[Jazzman53]

One other option is to have Russ Knotts try a rebuild for you. He might be a good source for information on this online help guide. He is usually very willing to share his knowledge.

http://www.eslrepair.com/Contact_us/index.html

Yes, I purchased materials for my very first ESL build from Russ many years ago. He's a stand-up generous guy and a great resource for panel repairs.

 

[Lightloopy]

I think there are probably a lot of old panels around mainly because it seems to somewhat involved depending on one's skills. When I was researching it , was beyond me and would require a significant time investment.

I think it's a great idea and I have a pair of old CLS II panels if anyone wants to have a go.

 

[Jazzman53]

I think there are probably a lot of old panels around mainly because it seems to somewhat involved depending on one's skills. When I was researching it , was beyond me and would require a significant time investment.

I think it's a great idea and I have a pair of old CLS II panels if anyone wants to have a go.

I would prefer a hybrid panel for the proposed rebuild guide, because that's what most owners have.

But if you want to rebuild those CLS panels someday, I'll walk you thru it (it would be easier than you think).

 

[JonFo]

2) Adhesive tape definition in the stator/diaphragm bond line at side edges.
Also; is the bond line 1 layer of a thick tape only, or is it a plastic spacer shim with a thin tape on both sides? (I'm guessing it's 1/16" VHB urethane foam only)

I've taken apart Monolith (2' x 4' panels) and SL3 panels (1' x 4' panels) for re-use as elements in my sidewall acoustic treatments, so it was full destructo-mode. And good thing, as the Monoliths were next to impossible to separate the stators, I wound up tearing the mylar and using solvents to weaken the glues.

The SL3 was much easier; I pretty much could have rebuilt them, as they came apart cleanly.

IIIRC, the side-edge is a single double-sided glue foam on the SL3. The Monolith was possibly similar, but had a rubbery feel, so not a foam. And they used super-glue on that.

 

[JonFo]

3) Same as question 2, except for bond lines at upper/lower panel ends

On the SL3, seems it is the same material as the sides. But again, these memories are from 2008, so I might be off.

 

[Doug033]

I think there are probably a lot of old panels around mainly because it seems to somewhat involved depending on one's skills. When I was researching it , was beyond me and would require a significant time investment.

I think it's a great idea and I have a pair of old CLS II panels if anyone wants to have a go.

I’d be interested in your old panels; are they in good condition? Are you willing to ship?

 

[Lightloopy]

They were original panels with weak sound level in one side. Sorry- these would be local pick up only.

 

[Jazzman53]

I want to contribute but I may come to regret posting this thread.

Silly me...
I had imagined walking an owner through a panel rebuild, and then assisting with a rebuild guide to be posted by the owner. This would be easier for me, empower the owner, and save shipping the panels to me for the rebuild.

This remains my preference but I get it that most owners have never replaced a diaphragm and may lack the confidence to attempt it without a step-by-step procedure for their specific speaker, beforehand. 


Having made this offer (putting my entire foot in my mouth);
I would agree to rebuild (1) pair of hybrid panels if necessary, at no charge for my labor, and owner covers materials costs (about $300) and shipping both ways (I'm guessing $200).

Successful rebuild is assured provided there are no burn thru's on the stators. Burn thru's can be repaired but it's iffy so--no guarantees there.

BTW; I checked with ER Audio Australia—they have suitable materials but their rebuild kits are not for ML’s. So I priced from McMaster-Carr, Mouser and Ebay.

My $300 materials estimate includes a 35-yd roll of 1/16 urethane tape at $99, about $65 for the clear spar tape, $53 for the Licron Crystal, and $35 for the 12um Mylar, plus cleaning solvents and shipping. BTW; the last time I ordered Mylar from Ebay it took 7 weeks to arrive.

I stopped building perf-metal stators about 10 years ago but I still have a perf-panel back page on my website here: Building Perf-Metal Stat Panels


FYI:
Physics (inverse square power law) pretty much dictates that hybrid ESLs must have about 1/16” diaphragm-to-stator spacing (d/s). Anything less risks driving the diaphragm into a stator and anything more lowers output so much that a minimum d/s is practically mandatory. Fortunately, 1/16” is a standard tape size, so there you have it.

Full range ESLs typically need twice the d/s (and/or huge offsetting surface area), double the bias voltage, and up to double the transformer step up ratio (more than 150:1 limits bandwidth and smokes amps), and even this doesn't fully offset the efficiency loss imposed by doubling the d/s-- a full range ESL is still significantly less efficient than a hybrid (inverse square law is a killer).

Typically; 2-sided 1/16” urethane foam tape is used as the spacer that sets the d/s, and also the adhesive that bonds the diaphragm to the rear stator.

For aesthetics, the horizontal diaphragm support spacers (spars) are clear plastic strips bonded to the stators with an optically clear 2-sided adhesive tape.

For disassembly, the stators can be pried apart using thin plastic wedges to avoid damaging the insulating powder coatings, and possibly leave the diaphragm intact. This could be difficult, and if so the periphery foam tape can be cut thru with a box knife (destroying the diaphragm).

Another method uses a string saw made from 60+lb braided fishing line, which cuts thru foam tape like butter, without damaging the stator's powder coating. This works great sawing horizontally thru flat panels but probably not as easy sawing vertically thru curved panels.

The onerous part of rebuilding a panel is removing the adhesive tape residue. It can be softened with acetone or a citrus based solvent, but nothing actually dissolves it for easy wipe off. it basically has to be solvent-softened and then rolled off by hand. 3M sells and recommends a rotary eraser that mounts in a drill motor, but I haven't tried it.

If the diaphragm coating loses conductivity and volume over time, these can be restored by cleaning and re-coating. If the diaphragm is damaged during panel disassembly, it must be replaced and then conductive-coated.

Tensioning the diaphragm can be done by hand using duct tape to grab its edges and a spring-scale to gauge the tension. Or in a tensioning jig which holds the film in a curved state during tensioning.

ML uses a diaphragm material that’s pre-coated with optically clear, vapor-deposited indium tin oxide. This coating option is not available for DIY, but a popular alternative coating is Licron Crystal aerosol spray, which dries almost clear with a slight blue-gray tint, and about 2 microns thickness.

ML conducts the biasing voltage onto the diaphragm using an adhesive backed copper foil strip applied onto the diaphragm along its periphery. The copper foil's adhesive backing is itself conductive, to transfer the bias voltage onto the diaphragm.

Another option is placing the copper foil charge ring on the opposite [front] stator, which then contacts the diaphragm when the stators are mated together. DIY'ers typically opt for this arrangement.

The best way to connect the wire lead from the HV bias supply to the copper foil is by soldering it. I'm not sure how ML makes this connection but I would solder it.

Once the diaphragm is installed and coated and the copper foil charge ring and bias supply connection are in place, the two stators can be mated together to complete the panel.

Contrary to popular belief; the front and rear stators DO NOT have to be bonded together—they need only to be held together in intimate contact. This would hold true for all the curved ML hybrid panels. Bonding the stators together does add significant rigidity to the panel, and that may be another reason why most manufacturers do it.

The CLS's are much larger, with complex curvature, so I would completely bond their stators together to ensure stability.

Some ML panels are bonded along the upper & lower edge bands only, with the long edges mechanically held in contact by the speaker frame. Some models, (CLS’s I believe) appear to be bonded around the entire periphery, and spars.

I’m not aware of any commercial ESLs where the front & rear stators are completely un-bonded. If the speaker frame is rigid enough by itself, and can hold un-bonded stators in contact with no lateral slippage, doing so would allow easy disassembly for cleaning and re-coating the original [undamaged] diaphragm.

But then you wouldn't have to buy new panels $$$

 

[ih82lose]

I’ve got a pair of Request panels that could use refreshing, and I do love this kind of effort to enable documentation for DIY.

@Jazzman53 where are you located? My main concern would be shipping the large panels without them getting damaged in handling.

 

[Jazzman53]

I’ve got a pair of Request panels that could use refreshing, and I do love this kind of effort to enable documentation for DIY.

@Jazzman53 where are you located? My main concern would be shipping the large panels without them getting damaged in handling.

I live in Savannah, GA

Yes, shipping is always a concern with ESL panels if you don't have specifically-engineered shipping boxes.

Eight years ago when I transitioned from perf-metal panels to wire panels in my DIY speaker builds, I gave away (2) pairs of my older (but still-working) perf metal flat panels, for the cost of shipping.

I thought I had packaged them pretty well--- I even had them sandwiched between two sheets of hardboard, but one pair arrived damaged with a bent corner. Since I packed them, I took responsibility and paid the shipping bill ($90) for the damaged panel. I then decided not to give away any more panels.

I will PM you my address.

 

[Mirolab]

For anyone interested, I have old SL3 panels, free (except forshipping), or pickup in Long Beach, CA.
And I have an M-L box that my new SL3 panels were shipped in.
The box may be worth more than the panels!?

 

[birdbox]

What about doing a "Panel Rebuild Guide" using a ML Theater center channel panel? I just purchased that speaker last week and I'm very unimpressed with the current sound. I'm assuming it's due to age and dirtiness. I emailed Martin Logan and they no longer support providing replacement panels for Theater & Theater I due to "vendor issues", and they have no plans to provide any in the future, So anyone with this speaker is "out of luck" on replacing the panel.

With that said, a good rebuild guide could help all the current owners out there bring their older panels back to life (it's going on 20 years since the Theater was discontinued). For now, I'm going to give my panel the shower treatment and hope for the best. I'm also waiting to hear if ML would take my current panel and put a new diaphragm on it, and if so, what that cost would be.

If the shower method doesn't fix the issue and ML won't rebuild my panel, I might be willing to be the "test subject" naive and motivated "DIYer" for a well documented panel rebuild.

 

[Jazzman53]

If you're up for a DIY panel rebuild, I would be happy to assist you remotely. It's probably easier than you imagine.

I should reiterate that I've have never rebuilt an ML panel but I've built many ESLs, the same physics apply, and it's not rocket science.

The prime directive for an ESL panel is efficiency, and efficiency is determined by the drive voltages and [predominantly] the diaphragm-to-stator gap (D/S). All physical panel features follow from the D/S.

For a hybrid panel, the physics dictates that the D/S should be approximately 1/16" -- any greater sacrifices output and any less would not provide enough room for the diaphragm to move and make sound without slapping a stator.

We can therefore deduce that the panel's edge spacers will be approximately 1/16" (which is also a standard thickness for shim and tape materials). ML spacers may be combination of a hard plastic shim with a thin double-sided adhesive tape on both sides (adding to 1/16"), or just a single layer of 1/16" thick double-sided VHB polyurethane foam tape. Either would suffice.

ML has historically used 12-micron Mylar diaphragms, but I read somewhere that the newer micro-perf panels may use thinner 6-micron Mylar. I prefer and use 6-micron film for all of my DIY panels (lower mass = better high frequency extension).

The worst part of replacing a diaphragm will be removing the old adhesive tape/residue from the stators. Nothing dissolves it or works well for this task-- you must peel off away what you can and then soften the residue with acetone and roll it off with finger pressure (it's a pain).

For your small center channel speaker panel, I would tension the diaphragm to 1% to 1.25% elongation (it's easy to measure using reference marks and a ruler-- see the panel assembly video on my website).

Because the ML panel is curved, the diaphragm must be tensioned predominantly in the direction perpendicular to the curvature, as any tension in the direction of the curvature tends to pull the diaphragm into the rear (convex) stator. In the curved direction, apply only the minimal amount of tension needed to pull out any wrinkles in the film.

Regardless of how ML configured the bias supply-to-diaphragm connection, I would use a full-periphery copper foil charge ring and solder the bias supply wire to it, as shown in the panel assembly build video on my website.

ML applies a vapor deposited indium tin oxide conductive coating on their diaphragms. This coating is not feasible for the DIY'er but two excellent coating options are Statclear A50 (not sold in the US) or Licron Crystal (sold in the US by Mouser, Allied Electronics and others).

In conclusion;
One layer of 1/16 thick 3M double-sided VHB polyurethane foam mounting tape applied to the periphery of each stator will suffice as the spacer material. Also the foam tape will bond the diaphragm to the rear stator instantly with minimal fuss.

After that; spray on one "just wet" coat of Licron Crystal, let it dry overnight.
Then apply the bias supply wire / copper foil charge ring to the front stator.
Assemble stators together (the tape will bond them together).

Enjoy the music!

If you decide to pull the trigger, PM me and I will give you my phone number.

You can also pick up some info from the videos on my website:
Jazzman's DIY Electrostatic Loudspeaker Page

 

[ih82lose]

I started doing some testing on one of my damaged request panel this weekend. As advised, before attempting disassembly, I attempted to measure the Mylar tension. I can get flat edge toothpicks through the perforated stator to contact the Mylar, but was not satisfied with different approaches to get a quantitative measure of the film tension.

I kept imagining a dial gauge with some constant load spring tension and a fixed deflection distance. Then I thought something like that must exist for dialing in tension on drums. And it looks like the “Drum Dial” could be that device.

https://www.drumdial.com/videos/drum-tuning-two/
I ordered one and will report back with results, and pictures.

 

[Jazzman53]

I started doing some testing on one of my damaged request panel this weekend. As advised, before attempting disassembly, I attempted to measure the Mylar tension. I can get flat edge toothpicks through the perforated stator to contact the Mylar, but was not satisfied with different approaches to get a quantitative measure of the film tension.

I kept imagining a dial gauge with some constant load spring tension and a fixed deflection distance. Then I thought something like that must exist for dialing in tension on drums. And it looks like the “Drum Dial” could be that device.

https://www.drumdial.com/videos/drum-tuning-two/
I ordered one and will report back with results, and pictures.

That sounds like a winner idea!

Once you know the deflection you can tension the diaphragm to the same deflection to yield the same tension that ML applied. It's better to over-tension rather than under-tension the diaphragm so you want to err on the high side. Also; before tensioning the diaphragm, add some reference marks on the film using a fine tip felt pen.

Then, after the film is tensioned to the determined deflection, you can measure the distance between reference marks (post stretch) to determine the percentage of elongation, and the elongation is the what you would specify in your rebuild guide.

Others could then use your guide to tension their diaphragm using the same thickness of film and stretching it to the specified elongation, which is easy to measure with reference marks and a ruler.
For example; if the elongation is 1%, you would place reference marks 12" apart and then stretch the film until reference marks were 1% farther apart (12 1/8" in this case).

 

[birdbox]

If you're up for a DIY panel rebuild, I would be happy to assist you remotely. It's probably easier than you imagine.

I should reiterate that I've have never rebuilt an ML panel but I've built many ESLs, the same physics apply, and it's not rocket science.

The prime directive for an ESL panel is efficiency, and efficiency is determined by the drive voltages and [predominantly] the diaphragm-to-stator gap (D/S). All physical panel features follow from the D/S.

For a hybrid panel, the physics dictates that the D/S should be approximately 1/16" -- any greater sacrifices output and any less would not provide enough room for the diaphragm to move and make sound without slapping a stator.

We can therefore deduce that the panel's edge spacers will be approximately 1/16" (which is also a standard thickness for shim and tape materials). ML spacers may be combination of a hard plastic shim with a thin double-sided adhesive tape on both sides (adding to 1/16"), or just a single layer of 1/16" thick double-sided VHB polyurethane foam tape. Either would suffice.

ML has historically used 12-micron Mylar diaphragms, but I read somewhere that the newer micro-perf panels may use thinner 6-micron Mylar. I prefer and use 6-micron film for all of my DIY panels (lower mass = better high frequency extension).

The worst part of replacing a diaphragm will be removing the old adhesive tape/residue from the stators. Nothing dissolves it or works well for this task-- you must peel off away what you can and then soften the residue with acetone and roll it off with finger pressure (it's a pain).

For your small center channel speaker panel, I would tension the diaphragm to 1% to 1.25% elongation (it's easy to measure using reference marks and a ruler-- see the panel assembly video on my website).

Because the ML panel is curved, the diaphragm must be tensioned predominantly in the direction perpendicular to the curvature, as any tension in the direction of the curvature tends to pull the diaphragm into the rear (convex) stator. In the curved direction, apply only the minimal amount of tension needed to pull out any wrinkles in the film.

Regardless of how ML configured the bias supply-to-diaphragm connection, I would use a full-periphery copper foil charge ring and solder the bias supply wire to it, as shown in the panel assembly build video on my website.

ML applies a vapor deposited indium tin oxide conductive coating on their diaphragms. This coating is not feasible for the DIY'er but two excellent coating options are Statclear A50 (not sold in the US) or Licron Crystal (sold in the US by Mouser, Allied Electronics and others).

In conclusion;
One layer of 1/16 thick 3M double-sided VHB polyurethane foam mounting tape applied to the periphery of each stator will suffice as the spacer material. Also the foam tape will bond the diaphragm to the rear stator instantly with minimal fuss.

After that; spray on one "just wet" coat of Licron Crystal, let it dry overnight.
Then apply the bias supply wire / copper foil charge ring to the front stator.
Assemble stators together (the tape will bond them together).

Enjoy the music!

If you decide to pull the trigger, PM me and I will give you my phone number.

You can also pick up some info from the videos on my website:
Jazzman's DIY Electrostatic Loudspeaker Page

I will admit all of that sounds a little daunting at first read.

I did wash the speaker and although it's improved, it is still feeling muffled until I crank up the volume to "11", at which point it sound pretty good, but the level is not something I'd be comfortable watching a movie at.

ML responded to my inquiry about using the existing panel for the rebuild and they said that's a No-Go as the disassembly destroys the stators. Probably just more work and risk than it's worth for them. So it's sounding like all Theater speaker owners are out of luck on replacing the panel via the factory. There is the option to send it off to Russ as well as possible other rebuild companies, however, I'd lean towards DIY due to the cost savings and learning/knowledge gained.

The easy thing to do is sell it to someone else who's ok with a used speaker at a good price that has lost some of it's "sing".

If I decide to go the DIY route, I'll reach out. For now, I'm going to think about it. I did take a lot of pictures of the shower results as "before" and "after" pics, including pics using a stereo microscope, which is pretty impressive what you can see on the mylar and spacer tape.

Thank you for such a detailed description, that does really help me understand the effort required to rebuild myself!