Adrain....so I would need a way to test your theory.
If I understand you correctly, 88's right string, the one that sounds nice in the oxX key driven hammer test, as evidenced by the discrepancy in the carbon paper test, is still being hit, but later than the right and center...maybe only grazed by the hammer. I suppose a way to test that would be to file the right side of the hammer such that it was way below the string at strike, to make sure it never hits the string at all...confirm there is no grazing of the right string
I did kind of do this by mistake at first when I mis-diagnosed the strike pattern. I did note that the when really making the hammer lopsided, perhaps taking the right string out of the equation at the hammer, the sound got worse. Though the way I did this also reduced the middle string's contact, as I filed the strike point in a straight, though lopsided line. Perhaps the test is to get the strike so I have the original ox for left and right, and only take the right string out of the picture, by filing it aggressively all by itself. I will try to mess with this on friday when I see it...though its much harder to mess around like this once the piano is on-site.
If this theory were true though, I still think it will leave an essential frustration for me. Filing all to equal as we normally try to do, will leave all three with equal strike, and equally too long contact time. When doing the final prep on this piano, I elected to leave the unequal condition, because it sounded better up there...but I will revisit my impressions now that I have some perspective, not having the piano in my face in the shop for a couple of weeks..
------------------------------
Jim Ialeggio
grandpianosolutions.com
Shirley, MA
978 425-9026
------------------------------
Original Message:
Sent: 12-15-2018 04:14
From: Adrian Carcione
Subject: Interesting and weird high treble effect
Maybe it was actually the leftmost string of #88 (in more firm contact with the hammer, and hence probably lower) which was precisely the culprit responsible for a quicker hammer rebound speed that allowed the rightmost string to speak louder during your striking sound profile test.
Then, it would make very good sense how the carbon paper profile suggests Xxo, while simultaneously the striking sound profile suggested oxX.
I like your notation system a lot!
Essentially after it has been well mated, we're persuading the hammer contact time per string, or rebound response per string, to become more equal among each wire in the unison.
So perhaps this discrepancy between the Xxo and oxX test results has less to do with string coupling phenomena.
------------------------------
Adrian Carcione
Simi Valley CA
805-578-8941
Original Message:
Sent: 12-09-2018 10:49
From: Jim Ialeggio
Subject: Interesting and weird high treble effect
Behavior of the high treble can be downright weird. Take this observation. Chinese grand makeover I just finished and sold. In the rush to get it up and running for a chapter technical I was doing on the piano, the top half octave, which needed mating, got ignored until after the technical...no time.
Later, at my leisure, I went to mate these hammers, no.83-88. I use this notation when mating hammers, per unison: oxX, = "o"left string no contact. "x"middle string contacting but weaker than the right, and "X" right string contacting well.
88's mating pattern was oxX. Proved by striking the note with the key driven hammer, and muting individual strings. Ok...so I thought "right string is sounding the loudest, so I have to take some off the X, middle string is sounding but weaker than the right, so I have to take a little off the x, and left string is not sounding, so I take none off the "o". I do that, and the strike pattern gets worse...still oxX but worse. I repeat the removal of felt logical that seems logical several more times, until the crown is noticeably tilting down to the right side...the side supposedly striking the string. The tone of the entire unison starts to get noticeably worse.
There was one puzzling observation I made while messing about in the proceeding paragraph that I didn't mention. Though the original "un-corrected" no.88 had a striking sound profile of oxX, the carbon paper visual evidence said Xxo...the opposite. I scratched my head, but didn't think about it initially, assuming the carbon paper was giving a bogus result for some reason.
However I soon realized a bizarre thing...the carbon paper was absolutely correct. "X" the loudest sounding struck note, and the note with nice tone, was indeed not being struck at all by the hammer...not at all. "x" was striking but not firmly, and "o" the string producing no sound at all, was the only string that was actually being firmly contacted by the hammer. This was confirmed both by the carbon paper trails, and by raising the hammer carefully to the strings and plucking the notes.
I went on to confirm this weird behavior in the entire top half octave, at which point we seemed to exit the worm hole and return to normal results.
My theory is that that hammer is completely damping the vibration, and the only good tone came from vibrations initiated at the bridge. no. 88 hammer is 3.7g (not SW, but HW) Hammers are light, though this set was heavier than usual for me, as the last piece of wood used in the moulding was very dense. Usually I can get them down to 3.2-3.0 gm. Leverage was slightly elevated with hammer weight adjusted for touch. WNG shanks. Regulation and letoff functioning well. 2mm letoff with drop a touch higher than letoff...but definitely no blocking at drop. I push shank friction in the treble up to 4g measured at 20mm from the center.
David Love's comment on hammer rebound time (COR) made me think about this. I have been thinking about how I could have increased rebound speed up at 88 where there is little felt to serve as a serious spring, and requirements of stiffness in the hammer are greater than lower in the scale. 88 was well juiced, as the factory 49mm SL which I did not get into changing on this instrument, was challenging tension up there to start with. When re-scaling I have been pushing SL's up there to 55mm for a while, and getting good results with the higher tension. I have not noticed this weird behavior before though...but I will certainly pay attention in the future.
But yeah, how to cut rebound time when my specs are already geared to reducing that time?
------------------------------
Jim Ialeggio
grandpianosolutions.com
Shirley, MA
978 425-9026
------------------------------