Replaced the cable in my SR-009 yesterday. They are playing great now.
When soldering the wires to the stator and diaphragm tabs, I came across some details that I thought should be passed on.
Desoldering all the wires and removing the old cable was not a problem. See the "Ants in My Stax" thread for 009 disassembly. Remove all the plastic parts mounting the cable to the ear pod, don't try to desolder the wires/cable with these in place -no room.
On reassembly, first note orientation of the cable for each ear pod, The arrow on one side of the molded strain relief is to be assembled on the inside only (ear side) and will help ensure signal polarity between L&R channels.
Soldering the +/- signal wires to the stator tabs was straightforward. Where it got interesting was when I went to solder the pro bias wires to the diaphragm tabs. The diaphragm solder tabs on my SR-009 are a small additional part that had been riveted to the actual diaphagm. The right side rivet was well attached and the wire/tab easily soldered. But the left side rivet was loose and the tab rotated wildly when I touched it with my solder tip. Took some doing, but got it soldered. Then to double check when done, I measured the continuity of each wire from solder tab to cable pin. All measured 0.01 ohm, except the left bias wire. That measured ~25 ohms and varied wildly as the tab was minimally rotated. I then attempted to solder the bias wire to the diaphragm directly, but did not want to heat too much and possible melt it. Also, the material did not take to solder (probably aluminum?) and so the reason for the riveted solder tab. I finally resorted to adding a blob of solder (with the bias wire) on the tab, and rotating until the tab wedged tight under the diaphragm. That ended up also measuring 0.01 ohm and was stable with the cable assembled.
Powering up the headphones, I noticed that both L and R channels started playing at nearly the same time. Previously, the L channel took several seconds longer to start playing, and usually had a sllght and wandering imbalance with the R channel for the first 1/2 hour or so. I took that as amp warm up (KGST), as things settled in nicely after that time. Now, the central image is rock stable from the start up, and the soundstage is balanced and huge. Got me thinking that loose rivets may be a cause of imbalance in some SR-009s, and takes longer to fully charge?
My 009s are around 3 years old. I did not take photos while I was working on them, but attached detailed photos from the "Ants in My Stax" thread from 2012. In those earlier headphones, Stax is seen to have used a screw and nut to assemble the solder tab to the central diaphragm. Later versions like mine were "cost reduced" to rivet that connection. When that changed over in production, I do not know.
I could image the intermittent coming and going of stereo balance problems written about, may be due to a loose rivet and random cable handling. That might be something others with SR-009 channel imbalance issues may want to check out.
Earside view (note arrow), circa 2012:
Opposite view, circa 2012: