Welcome! Need support, you got it. Or share you ideas and experiences.
In Reply to: So now, how about a write-up... posted by Gannet on February 19, 2001 at 16:40:51:
My MG-I Rewiring Experiences
Rewiring MG-I's is not particulary difficult, but requires some patience and a light touch. I've divided the following into multiple sections, each of which addresses a specific item or issue, in the hope that this will make appreciation of the whole process easier. Your mileage may vary; also bear in mind that many (most?) of the chemicals and adhesives involved are highly flammable and also have an insidious narcotic effect if inhaled. Do this only in a well-ventilated area, away from open flames or unprotected sources of electrical spark.
I picked up a pair of MG-Is cheap, as-is. Although cosmetically good, one was completely nonfunctional, the other had a dead tweeter section and an annoying buzz. Upon pulling the socks, I found that the panel wiring was severely corroded; sections of the tweeter wires were nonexistent and the midrange/bass wires were loose and brittle. Magnepan was unhelpful, they "don't have anything for MG-I's." I searched the PSA archives and found a great deal of useful information, particularly regarding adhesives, and 59mercedes was kind enough to provide an excellent description of how to rewire Tympani panels. At this point I had enough information to start choosing materials.
I managed to salvage small snippets of relatively uncorroded wire from near the terminals. The bass panel wire is enamel coated aluminum magnet wire. I stripped the enamel and measured the diameter of the wire with a vernier micrometer; the wire appears to be approximately .0249" diameter, corresponding to 22 AWG. The tweeter wires appear to be uninsulated, or if they had been insulated, the enamel had fallen off the section I recovered. They measured at .0089", corresponding to 31 AWG.
I discovered that aluminum magnet wire is difficult to find. Just for grins, I decided to investigate what would happen if I used copper wire. According to a wire chart I found online (www.pleo.com) the resistance of 22AWG aluminum magnet wire is about 26 ohms/1000 ft. Matching this up to copper wire, 24 AWG copper has about the same resistance. In the case of the tweeter wire, 33 AWG copper has about the same resistance as 31 AWG aluminum. By multiplying the cross-sectional areas of the respective wires by the densities of aluminum and copper, I found that the mass of wire in the panel would increase by about 2.09X - the much higher density of the copper is partially offset by the smaller diameter. For the tweeter, this means an increase of only about .75g. I deemed this to be insignificant. Over the 180-odd feet of wire in the woofer panel, this corresponds to an increase in mass of about 50g. This is a fairly large change in mass, but I decided to try the experiment anyway.
I obtained 24 AWG copper magnet wire from McMaster (www.mcmaster.com). Their entire multi-thousand-page industrial supply catalog is online, is hopelessly addictive, and will occupy you for hours if you're not careful. McMaster has wire only as small as 28 AWG; luckily I had a spool of 34 AWG magnet wire in my junk box. The difference between 33 and 34 AWG is such that the resistance of the tweeter panel increases by about 1 ohm; I figured I could change the crossover cap later if this unduly upset the crossover frequency.
One of the panels had a fairly sizable tear and a number of small holes. I found adhesive backed Mylar tape, in a number of widths, at McMaster. I bought a roll of 1" wide tape, and cut to size as needed.
I chose a combination of 3M Super 77 (available at crafts houses) and 3M 1357 (McMaster, again). I used Super 77 as the sole adhesive on the tweeter. On the woofer, I tacked the wire in place with Super 77, and overcoated with 1357. The solvent in the 1357 will attack the Super 77 and lift the wires, but this can be managed with application technique (more on this later).
Solvents and Brushes:
Acetone (available at hardware stores) will dissolve the old adhesive. It also evaporates rapidly, is extremely flammable, is an asphixiant and narcotic, and will dry out your skin. Use it with caution. For applying the new adhesive, I used tin-handled "acid brushes" which are cheap enough to simply pitch them rather than cleaning them.
THE REBUILD PROCESS
Make a sketch or take a picture of how the wires are run on each panel. Once the old wires are removed it will be harder to figure out where all of the wires went. It also helps if the speaker is elevated to a comfortable working height. The process is time consuming, and bending over a low work table will cause a stiff neck and back. I elevated my speakers to mid-chest level.
Removing the Old Wire:
I first snipped the old wires off at the solder terminals. I then poured acetone from a small polyethylene beaker onto small areas of the panel and allowed the solvent to soften the adhesive. This allowed the woofer wires to be gradually stripped from the panel. The tweeter wires will come off in a zillion little pieces, which can be gently wiped away with a shop towel. Be careful not to rub too hard, as the wire fragments and abrasive corrosion products can abrade or puncture the Mylar panel. Once all the wires are removed, use more acetone and a soft cloth to gradually and gently remove the old adhesive. The Mylar is transparent, and the magnet structure behind the panel will gradually become visible. This structure will serve as a visual guide to align the new wires.
Patching the Panel:
Any tears or small holes should be patched. I used the Mylar tape mentioned above. The trick is that the adhesive backing of the tape should never face the magnet structure, to prevent the tape from adhering to the magnets. In the case of a small hole, cut a small piece of tape, slightly larger than the hole, and stick it, face to face, to the actual patch. Then apply the patch with the smaller piece over the hole.
Applying the New Wires:
Rather than build a frame, as 59mercedes did, I decided to run the wires freehand. I started with the woofer wires. I sprayed a stripe of Super 77 along the length of the panel, corresponding to about two out-and-back wire runs, using a handheld cardboard shield to keep the adhesive only on the panel. The adhesive becomes exceptionally tacky after a few minutes, and remains so for a few minutes. Leaving a long pigtail of wire, I started gently pressing the wire down to the panel, according to the photos and sketches I had made earlier. A small amount of tension in the wire will help keep the runs straight. The magnets, visible behind the panel, serve as a guide (the wires must run between the magnets). When I reached the edge of the tacky area, I applied another stripe of adhesive, let it become tacky, and then continued. The MG-I has a two-turn woofer panel, so it's necessary to come back around for the second turn. This is easier, since the wires already in place help to align the second set. To form the turns at the ends of the runs, I used the cylindrical aluminum handle of an X-acto knife.
After the wires were all tacked in place, I began coating the panel with 3M 1357. As I mentioned earlier, this will attack the Super 77. I prevented the solvent from lifting all the wires off the panel by applying the 1357 in three or four 1" wide stripes, perpendicular to the wires. After these set up, I then painted on more stripes, adjacent to the previous ones. In this way I gradually covered the whole panel without allowing significant lengths of wire to be wet with adhesive at one time. At the terminals, I scraped the enamel from the wires and soldered the copper wire to the lead solder at the bottom end of the terminals, rather than to the aluminum solder at the top end.
To prevent the woofer wires and tweeter wires from touching each other where they cross, I placed a strip of theatrical gaffer tape (black cloth tape with very tacky adhesive) over the tail ends of the woofer wires where they run to the terminals.
Applying the tweeter wires takes only a few minutes. The tweeter wires are applied in exactly the same fashion, except that the wire is much thinner, and there are fewer runs. I did not apply 1357 to the tweeter.
I have not had time to perform any objective tests, so I cannot yet establish whether the increased wire mass produces any change in the frequency response. I would expect a downward shift in the overall woofer frequency response curve, with an accompanying drop in sensitivity. Because the change in tweeter wire mass is relatively small, I would expect little change in its characteristics. Subjectively, the speakers sound remarkably clean and transparent, with no particularly obvious bumps or holes.
This project was my first firsthand experience with Magneplanars, although I'll admit I've been fascinated with planar loudspeakers since I was a kid. The design is a fascinatingly different implementation of the same physics that govern conventional electromagnetic transducers - and appears to be surprisingly insensitive to assembly variations.
Thanks are due to 59mercedes, Arbelos, and all of the other Planar Speaker Asylum contributors who have laid the groundwork to make this project possible.
This should get linked in with the other repair articles.
tinballoon, thanks for your efforts! Get back to us with listening tests. :)
This post is made possible by the generous support of people like you and our sponsors: