Stewart Penketh ESL-63 Repair Pics


From Stewart Penketh in Canada I have received some additional information and very nice pictures describing the Quad ESL-63 element repair process.

I recently repaired several ESL-63 panels and since some expressed interest in receiving photos of various steps in panel repair, I took photos as repairs proceeded.

This image shows front and rear stator panels, with the diaphragm (which attaches to the front panel) removed. When the front and rear panels are bolted together, the aluminium foil along the long edges of the rear panel contacts the coated side of the diaphragm and transfers the high-tension charge to it.  Note the nylon damping material which is on the rear panel but not on the front. The damping material is like the sheer nylon used for stockings and is glued to the stator panel. Note also the three centre-panel spacer bolt protrusions.

Quad ESL-63 stators


Quad Panels
This image shows an early Quad panel alongside the later USA Monitor panels. The USA Monitor panel is actually less rigid (but acoustically more open) than the panel it supplanted.

Quad ESL-63 panels - new and old

This image shows an original Quad diaphragm alongside a rebuilt panel. The white spots on the dark coating of the Quad diaphragm are where the coating has apparently turned to ash, due to arcing. The coating on my diaphragms is transparent. Diaphragms are not coated at their outside corners. This is to avoid contact between the charged diaphragm and the screws at each corner of the panel which fasten the panel to the panel assembly (and to ground).

Quad ESL-63 diaphragms - damaged and rebuilt

This image gives a close-up of two damaged panels. In the first, the bond between the stator panel and the diaphragm has failed. In the second, the diaphragm is ripped immediately adjacent to the point of bonding to the panel, apparently due to fatigue.

Damaged ESL-63 panels

This image is of the mylar stretching jig. The white strips on the inner frame are double-sided tape. Tape is placed on the top of the jig (aligned with each bolt), and on the side. The tape's backing is not peeled off until the mylar is about to be attached. The tape on the top of the jig simply pre-stresses the diaphragm. The mylar is laid over the top of the jig, and these "pre-tensioning" tapes are then detached and re-attached until the diaphragm is evenly tensioned.  After this, the mylar is applied to the side tape. The tape's holding ability is extremely strong because the mylar is wrapped around the edge of the frame.  Tensioning the mylar pulls it more tightly onto the tape rather than off it, and the grip of the tape is reinforced by the friction of the mylar on the wooden frame of the jig frame.

Stretching jig - taping

This image shows a three micon mylar diaphragm fully-tensioned. A micron is one millionth of a metre.  Thus the ESL-63 diaphragm is one three-thousandth of a centimetre thick, so thin that it is like gossamer. I tension the mylar very slowly and cautiously, turning each wing-nut a half turn at a time, until the mylar is tensioned almost to the point of failure. It tears very easily, and fails completely.

Fully tensioned diaphragm

In this image, the diaphragm is resting on a stator panel to be masked prior to being coated. The purpose of the acetate cut-out mask taped at each end of the panel is to prevent any diaphragm coating being deposited on the outside corners of the panel, where it is connected by screws to the aluminium
panel assembly. Although the diaphragm is fragile and tears extremely easily, it can nonetheless support the weight of the jig.

Masking the diaphragm prior to coating

In this image, the resistivity of the diaphragm coating is being tested. Unlike the Quad diaphragm coating, my coating is transparent. I place washers about a quarter of an inch apart and touch a probe to each one. I cannot put the probes directly to the mylar as they would pierce it and cause it to fail. A resistivity reading of 200 megohms or higher is sufficient for a panel to reproduce all frequencies without charge migration. The multimeter in the image is set to read resistivity of 2,000 megohms, and the reading on the screen indicates that the resistivity is too high for the meter to read. The resistivity of the original Quad diaphragm (ESL-57, MT's remark) coating is about 12 million megohms.

Resisitivity measurement

This close-up image shows the stator panel spacer bolts projecting through holes pierced in the diaphragm. Spacer bolts are needed down the centre of the panel to preserve an exact stator-to-diaphragm gap over the ESL-63 panel's entire span. The holes are just large enough to ensure that the bolts and the panel protrusions do not impede diaphragm movement. At high volumes, air resists being moved by the diaphragm, and tries to escape through these spacer holes. This makes no sound when the diaphragm is made of three micron mylar. However, if an ESL-63 diaphragm is made out of thicker material (for instance the 12 micron mylar used in ESL-57 treble panels, or the Saran used in ESL-57 bass panels), you will hear unpleasant candy wrapper or bodily function noises as the
force of the air squeezing through the holes causes the diaphragm in the area of the holes to vibrate loudly.

Element bolts and protrusions


Closeup of the bolt protrusions


Extra bolts


A closer look on the bolting

In this image, the front panel is bonded to the diaphragm.

Bonding the panel to the diaphragm

In this image, the front and rear panels have been partially bolted together. Note that holes for reinforcing bolts have been drilled in each alternating cell around the panel perimeter, forming a final assembly that is significantly more rigid than a stock Quad panel. The panel is tested whie it is still in the stretching jig. I drive the single panel directly from the output of the plinth's step-up transformers, using an ESL-63 plinth which is stock except for removal of delay line circuitry. I use a 45 watt 303 amplifier. Driving a single ESL-63 panel with this amplifier corresponds to driving all four panels of an ESL-63 with an 180 watt amplifier. The plinth's protective circuit will not trigger, as it is calibrated for a four-panel load. When testing a treble panel, I use eight-head jumper cables in order to bridge
all eight segments, so that the entire panel runs full-range.

Testing the panel


Testing mid unit


This site uses frames, but if you have found this page via a search engine and do not see a frame on the left side of this page, you can use the home button to get to the main frame of the MT Audio Design site.