Apologies in advance for the long email. I am placing the most significant information first, so you can stop reading when you've had enough!
It took me this long to get back to the list because I contacted the technician who did the work, who in turn contacted the seller, Pacific Piano Supply. Randy Morton replied. The important info he had was that Tokiwa redesigned the wippens about 10 years or so ago, after my customer's parts were installed on the piano in August 2003. He generously offered to send me a few of the newest rep levers to pin on the "high traffic" areas of the action, should there be any more trouble. Here is his response:
Since this time, (10 -12 years ago) I have had Tokiwa make several changes in this area to improve the quality and function of grand wippens.
1. Changed the curve on the tip of the springs (copied Hamburg Steinway) – This changed the tip to more of a "half-circle" instead of the earlier "half-oval" shape.
2. Changed the spring slot (also to copy Hamburg Steinway wippen)- made this slot wider, with less angle on the sides
3. Changed the slot lubricant from graphite to emralon (this was to stop some "squeaking" of the springs – this is Yamaha's method and material)
The problem as described sounds like the groove is too narrow, and there is some kind of excessive friction on the tip of the spring.
Margie says she repined the rep levers (5-6 grams). She also states that the springs are "strong as I can make them without causing a bump" She also tried "round nose pliers to broaden the radius of the tip" (our changing the shape was to make the contact point less, thereby lowering the drag an propensity to "squeak"- broadening the radius would increase the surface contact and friction)
These are all admirable and logical solutions, but may have increased the spring tension and surface friction at the slide point, thus the rapid recurrence of the "digging" of the spring into the wood.
Although quite popular with many Techs, the repining the rep lever at much higher friction at the center, to make spring adjustment less "touchy", is not recommended by Tokiwa. Factory friction at the rep lever center is 1-2 grams
Finally, the use of VJ lube makes a temporary improvement, but then gets "gummy" and makes everything worse in my experience.
What I didn't mention in the original post (because it didn't help) was that one of the fixes I had tried was installing Renner springs of a smaller diameter in the low tenor. (I had experienced this action failure on brand new Steinways from the late 80's/early 90's. More info at the end of the email on the reasons for trying that fix.*) So when the problem recurred, I thought that perhaps I should stick to the original profile of the tip of the spring that contacts the grub. So I used a round-nosed plier to straighten the tips of the Renner springs to resemble the original Tokiwa parts. For comparison, attached is a photo of an old Steinway wippen spring (complete with a glob of whatever the lubricant-du-jour was) which shows a very flattened out spring. It could be that the shape of the spring doesn't matter that much; we have ample evidence that both shapes can work.
Randy sent me photos of old and new wippen grubs, and I can see that the new one is definitely wider and more U shaped as opposed to V shaped. Newer Steinway parts (I have a newish Steinway wippen of unknown age) seem to have the wider grub, but old Steinway wippens, of which I have samples from several different eras, have a tight fitting grub. Again, we have ample evidence that both grub shapes can work.
What did fix the one note (and thankfully, unlike past visits, I could not make any other notes misbehave) was an extremely slight rep lever height adjustment and more burnishing. (On all my visits to this piano, I have always used a shaped piece of maple, such as from a hammer shank, and not just a pencil tip.) I also applied Dag lubricant, and lightly polished/burnished it. Action regulation seems to be unusually touchy for this piano.
As for the rest of the suggestions from the list, I faithfully checked out all of the rest of them except for trying to harden the slot with CA glue or similar. (That seemed a bridge too far; however, I may be hardening and compressing the wood by repeatedly burnishing the grub.) I appreciated all the suggestions, but alas, none of the conditions that were suggested were present for me to correct. How I wish one of them had pertained to the problem.
An interesting addition to the question of how to get the jack to successfully seat itself under the knuckle was to pay attention to the profile of the back edge of the jack. Two local technicians who have had more experience with the Tokiwa wippen commented that the back of the jack on Tokiwa parts is a sharp 90 degrees, whereas the Renner jack is slightly beveled at the back. This bevel is a good idea, to help give the jack every advantage to slip back under the knuckle. The first thing I tried was adding a bevel using very fine sandpaper and re-graphiting, but only a very modest improvement resulted--not enough to call it a fix. I asked Randy if the jack profile had been changed as well, but he didn't answer that question. Again, checking my inventory of old parts, my old Steinway jacks didn't seem to have an eased edge. I'm sure I've seen some old Steinway wippens where the back of the jack has a slight bevel. Again, both types of parts can work, as we have had lots of years of playing the piano in real life to prove it.
I don't know how to organize or think about all this disparate information and seemingly contradictory dimensions and production techniques. There does not seem to be one overall cause or fix of the problem I described. The closest I came was burnishing and re-burnishing the grub--perhaps the cumulative effect is making the wood denser and harder after all, so that it resists digging a hole. I don't recall having the problem of catastrophic action failure (hammer sinking below hammer line) on any old Steinway parts, no matter how dilapidated, worn out and decrepit the parts were or how out of regulation the action was. Even verdigris, although making the jack extremely sluggish, didn't seem to permit the hammer to drop below the hammer line. A sluggish jack due to verdigris is an action failure, but can be addressed fairly easily, if not permanently. Similarly, I don't think that pinning had all that much effect either. Not to downplay the importance of the correct friction (or approximation thereof, since humidity and wear/tear wreak havoc with our most meticulous work) but I'm sure we've all seen actions with very tight centers or very loose centers, neither of which produces this kind of action failure.
After-market manufacturers of action parts have a tough job to fulfill. Since their parts have to work on a variety of pianos and with a variety of other parts, namely hammers whose weights can vary from maker to maker, they have to choose a middle ground. This is unlike the old days where factories generally had control over the whole process: exact dimensions of parts, weights of hammers, choice of lubricants etc. and the proper pairing of all of the above.
*Steinway installed Renner action parts with NY hammers on their models B and D starting in the mid 80's and continuing for a few years. When my husband, Mark Anderson, (who has since become another version of RPT--Recovering Piano Technician) was working for the local Sherman Clay dealership, he encountered 3 or 4 pianos with this problem of the hammer sinking below the hammer line, aka catastrophic action failure. It occurred exclusively in the low tenor. He checked out all of the possibilities we discussed here but could find nothing that made any difference whatsoever. He then observed the jack as it was returning on a slow release of the key. Of course it didn't make it back under the knuckle, allowing the jack to jam on the front side of the knuckle. Mark somehow got the bright idea to think about repetition spring size. To test out his theory, he swapped out a wippen located farther up in the treble that had a smaller repetition spring diameter. When he placed that wippen with the smaller spring size on a low tenor note, which naturally had a heavier hammer, he had to increase the strength of the spring to get adequate rise-on-release. Et voila. The problem was fixed permanently by installing a smaller size spring on all the low tenor notes. It seems counterintuitive to select a smaller size spring, but not when you consider the functions of the spring (jack return but also support of the hammer at rest) and how we determine the correct regulation. I think watching the rise on release is the only indirect method we use during action regulation. Everything else of importance (let-off, drop, jack position, key height, dip, backchecks, rep lever height, etc.) is a direct observation of that particular function. Rise on release stands in as a proxy for the real observation which is difficult to see: getting the jack to return under the knuckle, especially on a slow release of the key. A heavier hammer paired with a smaller spring requires greater spring tension, giving the hammer more support while at rest. (If you push down on the hammer while it's at rest, you should not be able to jam the hammer below the hammer line and make it stick.) Although the original spring size could produce the desired rise on release, it turned out to be too flabby at rest. Although we don't know for sure, one theory is that Renner hammers in the low tenor were heavier compared to the NY hammers in the low tenor, resulting in a hammer weight/rep spring mismatch.