Kurt Chang's 3836 repair
Since the problem exhibited by the bad 3836 driver I got off e-bay a
while back does not seem to be too uncommon, I thought it might be
useful to report what I have learned in trying to fix it so that
others who see (hear) similar symptoms in their Tannoy drivers can
have an idea of what might be wrong.
First, thanks to Joel Tatelman for sharing his experience of
repairing his 3836. Joel's driver had the same problem of losing a
good chuck of the output from the tweeter that my driver exhibited.
After reading his post of how he had his driver repaired, I decided
to fix the problem myself since (1) the cause of the problem seems to
be easy to fix and (2) I didn't like the idea of sinking another
$200+ to get the driver back in shape. Well, the repair job turns
out to be substantially more difficult and tricky than I expected,
and I am glad that the tweeter appears (??) to have survived through
the brutal surgery.
The first thing I did was to detach the tweeter from the woofer.
This involved removing 3 bolts that hold the tweeter to the back of
the woofer magnet assembly. This part was easy. When I tried to
separate the tweeter dome diaphragm from the tweeter magnet assembly,
however, I found trouble. It turned out that the metal plate that
has a hole over the magnet pole piece to form the voice coil gap was
not in its proper position. This plate should have been glued to the
magnet in a position concentric with the pole piece, but for some
reason it had become loose. Either the glue lost its gripping power
(fried by high power PA operation?) or someone had whacked the
tweeter magnet real hard. Due to the magnetic field, the natural
tendency for plate is to move sideway away from the concentric
position until the edge of its center aperture pinches the voice coil
firmly against the pole piece. Because this plate is still tightly
held to the magnet by the magnetic force, and the offset from the
normal position is very small, it is hard to tell what happened
simply by looking at the outside of the magnet assembly.
The problem was that the plate was pinching the voice coil against
pole piece with a huge force, on the order of tens of pounds! At
first I did not realize why the coil was stuck so hard in the gap,
and tried to get it out by brute force. The dome diaphragm is
attached to an integral plastic cylindrical acoustic lens. I tried
to pull the voice coil out of the grip by pulling on the acoustic
lens, and after a long struggle the coil finally come out. In
hindsight, I consider it a small miracle that the voice coil appears
to be intact (it still has the right DC resistance) and the dome
diaphragm still seems to be attached to the acoustic lens.
Once the diaphragm was removed, the uneven gap revealed itself and I
realized the cause of the problem. After scrapping the old glue off,
I tried to glue the plate back onto the magnet. This task sounds a
lot easier than it really is. As I mentioned earlier, the plate does
not really like to stay in the concentric position and will move
sideway unless it is somehow held in that position. This presented a
technical challenge. The issue is how to keep the plate in its
proper position to maintain an even gap for the voice coil while the
epoxy is curing. I finally came up with the solution of inserting
multiple segments of thin copper wire between the pole piece and the
inner edge of the central aperture of the plate as spacers to define
the gap. Although I am not sure the resultant annular gap is even in
width, it does not look too bad.
In hindsight, I think Joel's $200 was well spent. This repair work
is much trickier than I anticipated, and the risk of destroying the
tweeter is so high that it's better to get someone who really knows
what he is doing to do the job.
I put the fixed 3836 driver back in its temporary box (about 8 cuft
vented) last night, and listened to the pair for a while. The stereo
image appeared to be OK, and the soundstage was wide and deep. The
sound was promising, with good slam and dynamic. Perhaps for the
first time I got a glimpse of the kind of sound you Tannerds are
© Copyrights Hans Hilberink & Kurt Chang, 2002.