- Thoroughly clean the diaphragms and stators using a vacuum cleaner.
- Visually checked all surfaces to ensure that was no dust or insect causing a leak between the diaphrams and stators.
- Rechecked and tightened all connections.
Still the problem persisted. I then proceeded to ensure that none of the other equipment that was connected to the speakers caused the problem. To do this I did the following:
- Swapped the speaker cables
- Used another amplifier
- Used another CD player
Again, no change, the problem persisted. With all the possible culprits eliminated, the finger pointed at the electronics, and since the problem affected the treble frequencies, it must be one of the 1.2ohm resistors connected in series with the speakers. Rob had replacement resistors sent to me, free. I promptly replaced the offending resistors and the problem is gone. The ESLIII are now 100% again.
Thanks goes out to Rob of ER Audio for the great support.Return to Home.
The cabinet is constructed out of 2 main modules. In this picture you will see a box with 4 aluminium legs. This box measures h320xw600xd250mm. This will serve as a base for the ESL panels. It will also house the EHT and transformers. Behind the base and leaning against the wall, you will see the removable panel. This panel forms the rear of the base. The large round hole will house the speaker binding posts and the small rectangle will hold the power cables connections.
In the next picture you will see a large rectangular frame that will house ESL panels. The frame measures h1300xw600xd160mm. I did a mockup, and I was surprised as it seemed a lot larger than I expected. Its going to dwarf everything else in my listening room. It stands at about 1.6m. The frame holding the panels will screw onto the base as shown in the mockup. I originally wanted to make the base and the top using one piece of wood, but we did not have a piece large enough. Well, can't always get what you want...
And here are the braces that will go into the top module. These braces serve 2 basic purposes. 1. To 'brace' the top module so that it is more rigid and resists sideways motion. 2. To serve as a backing to support the panels. The braces are also radiused so as to reduce reflection.
Before I can install the panels, my first task was to install the channels where the panel will sit in. This was a simple affair, just nailing the appropriate length of square cross section wooden rod into the top module.
Here is another picture of the channels being nailed into place.
With the channels installed, I test fit the panels. Perfect fit! I love the look of the ESLs, really majestic. I am wondering if I need some kind of cover or should I leave them naked. Opinions...
Rear view of the panels during the test fitting.
A closer view of the Panels sitting snugly in their channels.
I also test fit the braces. While the fit is good, but the support points are too narrow don't quite line up with the edges of the panels that it is supposed to support. I will have to fashion a new support that will extend to support the all the panels properly.
Satisfied that everything fits, I remove the panel to clean them and the cabinets up. Here is a picture of the completed cabinets without the panels.
Here is another view of the empty cabinets. Notice that the channels at the bottom are different from the other channels. They are screwed on to aid easy installation and removal of the panels. They were an afterthought and so the wood is not finished in the same colour as the rest of the cabinet. Notice that I have already predrilled holes in between the channel brackets so that the wires can be passed through to the base that will house the EHT supply and transformer. This will make things look neater.
I clean the cabinets up thoroughly and give the room a good vacuum. I then start installing the panels into the cabinets. I make sure that all screws are tight. Here is a rear view of the final install.
I place the EHT and the audio transformer into the base. I connect all the wiring. The ESLs are now in position. They are right next to the side wall and about 2 feet from the rear.
After making the final connections, I turn on the ESLs, put in some Nat King Cole, Marsalis, Hall and Oates, Eagles..... I am left speechless. The ESLs were used without the cabinets for a good 2-3 months and they were just wonderful. But now in a proper enclosure.... they just got better. The sound emanating from the ESLs has just been elevated one level better. It sounded fuller, mid bass has much more presence and goes the panels go down another half an octave. I'll whip out the sound analyser for tests later. For now I just want to go enjoy the ESLs.
I flip the switch... and no sparks. So I go ahead and adjust the bias voltage for the panels. I let the panel charge up for about an hour before I do anything further with them. Here is a picture of the voltage regulator used to adjust the bias. 
At this stage put some music on and turn up the volume. Well, I hear music coming out through the panels. If only you could hear them. All the superlatives that you have ever heard reviewers and owners sprout about ESLs apply here. These speakers were made for music. It’s an all new listening experience with all my CDs. Instruments sound so real... one thing that struck me was the snare drum from some acoustic pieces. None of my cone speakers, my Fostexs full ranger included can reproduce the sound of a snare drum as accurately as the ESLs. If you have taken notice of the sound of a snare drum you'd know what I mean. From rock to techno, to vocals, to classical... the music is just great. Even my better half who is quite critical of audio equipment (but not fussy) said that they were very "nice". 
Somemore thoughts about bass thru ESLs. I think the ability of the ESL to do bass is really dependent on a host of factors and the most important one is step up transformer. For reasons that I have yet to understand, the number of octaves that a transformer can handle has a practical limit. I read the limit to be 6-7 octaves. One solution is to use a high freq transformer for a treble panel and a low freq transformer for the bass panel. The problem here is working out the crossover and integrating the panels. However I have it on good faith that the ESL's that I have DIYed has a freq response of +or- 6db from 35Hz to 200Hz. However, I do believe the cabinet and placement plays a very important role in achieving this, particularly the use of walls for reinforcement. 
A quick and thorough wipe with the swab followed by a wipe with some dry tissue did the trick. The swabbed, cleaned and dried diaphragms are now ready for the application of the conductive coating. See how clear and reflective they are...
All panels swabbed and dried.
Here is a picture of the bottle of conductive coating and the utensils for the job, a small plastic cup, a foam applicator and a syringe to measure the amount of fluid to be used.
Specific amounts are used to coat each panel. This helps ensures the correct coating density for each panel. No guesswork here.
Using the foam applicator, I carefully apply the conductive coating onto the diaphragm. I apply until all the liquid measured for the diaphragm is completely used up. This takes a little practice to get right.
Here is a view of the coated treble diaphragm once it has dried. At this stage I started getting the feeling that the coating was a bit thick, more on this later.
Here is a view of one of the Full Range Panels. Although it has dried, it would take from a couple of days to a few weeks for the conductive coating to cure. Again looking at the panels, I had a sneaky feeling that the coating was too thick.
Next, I put the front and rear halves of the one complete panel together. The front half of the panel is made up of the support structure, the stator and a tensioned conductive diaphragm. The rear half of the panel is made up of the support structure, the stator and a strip of copper foil that will pass high voltage polarising charge to the diaphragm. This foil will come into contact with the conductive side of the diaphragm when they are sandwiched together. I place the 2 completed halves, front and rear together. I do a final check to ensure everything is in order before I marry the halves together. Notice the foil and the diaphragm will come into contact when they are sandwiched together.
Carefully, the 2 halves are placed together. I ensure that they are correctly oriented. Then I bring out the plastic channels that will be used to clamp the halves together. A channel is required for each side of the panel. So its 4 channels per panel.
Here's a partially clamped panel.
Remember that I had this niggling feeling that the conductive coating applied earlier was a little too thick and did not go on as consistently as I would have expected. A quick check revealed that it was indeed too thick. Rob from ER Audio, said that this will lead to a degradation of the Bass. As such I must reapply the conductive coating again. The conductive coating must be diluted with distilled water. It’s my own fault really for not reading the instructions more thoroughly. But before I can apply the correctly diluted coating, I will have to strip the old conductive coating off. I was worried that I may damage the diaphragm and then have to start from scratch. But as it turned out the process of stripping is really very easy. The conductive coating is a water based acrylic, as such I needed an acrylic solvent to strip it off. Acetone was the best choice. All the local hardware stores do not carry Acetone. But I got lucky and found a can at Ace Hardware. 
I wet some tissue generously with acetone and begin wiping the conductive coating off the diaphragm. The stripping is actually very easy... just be generous with the acetone and tissue. Here is a diaphragm that I have completely stripped. I also clean the diaphragm with the medical grade alcohol swab. Incidentally, after the acetone and isopropyl cleaning, the diaphragm did not look any worse for wear. See for youself...
I then mix up a batch of conductive coating, diluted to the correct coating density. Here is a picture of the correctly coated diaphragm.
When viewed directly from above, the diaphragm is clear, but when viewed at an angle it has a bluish haze. This is exactly the outcome that I was expecting from my conversations with the designer.
The entire process of stripping and re-applying the coating probably took about 2 hours. I am getting really good at this. Well the panel build is finished. I am finally able to clamp all the halves together.
Here is a picture of the 6 panels, all finished.
And another closer look at the 3 panels that make up one side of the speaker.
Next I 
More measurements. 
After measuring the node points according to the designer's specification I mark them off. I then begin injecting the silicone at the indicated markers. The silicone adhesive is thick and spreads evenly between the diaphragm and stator in an oval pattern. Each node point is about 10mm in diameter.
Here is a completed node point, freshly injected.
More node points completed. 
View from of the node points from the diaphragm side of the panel.
I leave the node points to cure overnight before commencing the next step which is to apply 
Next I rollout a length of polyester film over the work surface. I make sure that there is enough film to adequately cover the references that was drawn on the work surface earlier. I also make sure that there is enough overhang on all sides of the film for me to work with I leave an overhang of 5-10cm on all sides. At this stage I just attach sticky tape over the 4 corners with moderate tension just to hold it in place.
Here is another view of the as yet, un-tensioned diaphragm. 
Here is a close up view from one of the corners with the sticky tape attached.
With the diaphragm and surface prepared, I am now ready to tension the diaphragm up to the designer's specification. To achieve this, I use the supplied tensioning gauge which is really a simple spring weighing scale with a piece of machined aluminium attached to the end.
First take a piece of sticky tape, attach about 2cm of it to the aluminium plate of the tensioning gauge. Next attach another 2-3cm of the sticky tape to the polyester film. Now pull the tension gauge until the correct weight is achieved, in ER Audio ESL III's case it is 800 grams. Then immediately press the sticky tape down onto the work surface and release the rest of the sticky tape from the tensioning gauge's aluminium plate. This is a really ingenious way of tensioning the diaphragm for a home builder, and simple too! The only complain is that I have 60 more sticky tape to tension all round the diaphragm. 
The polyester film is only 3.5 microns thick and as such very, very fragile. Do not let anything sharp come in contact with the surface. In my case, the tensioning gauge slipped from my hand and onto the surface (I was sleepy, and it was about 2-3am). It caused a rip and made half of the film unusable. Instead of getting 4 useable panels, I had to settle with just 2 panels with this sheet of film.
After about 30-45 minutes of tensioning, you end up with a properly tensioned diaphragm.
Now that the diaphragm is properly tensioned, I am ready to attach them to the panels. I start by applying the superglue or more aptly cyanoacrylate on to the air gap spacers. I apply two beads/channels along the length of the spacer. The first bead/channel is 2mm from the inner edge of the spacer; the second is just off the center of the spacer. The difficulty that I faced is that the cyanoacrylate dries quickly, so when I am applying the second bead, the first bead is already drying up and so I have to reapply the first bead again. You may not have realised but dry glue is no use to anyone!! This step is important so that the diaphragm is suspended properly from the inner edge of the spacer so as to avoid rattles when the diaphragm is at work.
Once the glue is applied, I quickly place the panels onto the diaphragm and press down firmly to ensure a good bond. In addition I place weights onto the panels to ensure good contact as the glue dries.... I leave it for 2 hours.
.... 2 boxes of electrolytic caps used as weights (20 x 100,000uf @80vdc meant for another project.... watch this space)
After the 2 hours are up, I use a very sharp cutter to free the panels from the rest of the diaphragm. Here is a picture of the completed panel with diaphragm. Because the glue does not spread evenly, I am left with numerous pockets of air between the film and spacer. In addition there are several instances where the film did not bond to the spacer at all. In this case I have to touch it up by dripping more cyanoacrylate on to the affected areas. This gets messy and aggravating to get right!
These pictures just does not show how challenging and aggravating this part of the build is, for me anyway. Here is a picture of one of the treble panels that was just about perfect.
Touching up the not so perfectly bonded area is a matter of applying more glue to the un-bonded areas. Since cyanoacrylate is very runny, I just drip some of it in the area and the capillary effect ensures that the glue fills up all the un-bonded areas. Attached is a picture of the less than perfectly bonded areas. 
According to Rob at ER Audio, there are more than a few ways to attach the polyester film to the air gap spacers.... 2 methods that I could have used was the super fast drying super glue method and the snails pace 24hour curing polyurethane method. On hindsight, I should have used the slower method. Why?