This is a multi-part message in MIME format. ---------------------- multipart/related attachment ------=_NextPart_001_002D_01C48841.6CC09A30 Hi Ric, everyone, Thanks for all the feedback. I look forward to hearing even more! You wrote: > Stanwoods particular curve shape is=20 > somewhat a subjective choice me thinks... tho he may say different. = He=20 > can and from time to time does provide custom curves for given=20 > situations that are not quite this same shape. Steinways own default=20 > curve is somewhat flatter btw. Ah hah! Yes, the hammers I have are duplicates for a Steinway D, so = that would make a bit of sense. The curve isn't just "flatter," but = really quite flat. It's flat as though by intent. I find that = interesting. A picture is worth a thousand words. (This GIF, at 9 kB, = is pretty economical.) The jump between 21 and 20 is of course the bass/tenor break, with the = bass hammers having more molding to them. The "SW target" curve, in = fucsia, isn't *necessarily* my target, depending on what comes from this = discussion. It's merely the best fit, per a linear regression analysis. Now that I look at this curve a bit more -- and look a bit more at the = hammers themselves -- it would appear that there is a natural curve in = the set, that is stepped apart across 20/21 because of molding = differences. However, the treble end (where the piano is gonna be the = most finicky) is straight as an arrow. My original intent was to even out the larger jags in the curve and then = to see how the hammers perform in the piano, before doing any more major = adjusting. This is in part because my keyframe is vacationing in = Florida at the moment, and so this was something I could do in the mean = time! <grin> I'm now thinking it would be a good idea to wait until I = can install these hammers in the piano and assess their initial = performance before doing *anything* to them. Well, it might prove = informative to experiment with the low tenor by wrapping copper wire = around the tips of the shanks to bring those hammers up to the linear = curve. The beauty of a spreadsheet is that it can tell me what length = of copper wire to cut and wrap around each shank. :-) Incidentally, I've thought about how to do the final adjustments. I've = of course generated an errors column that tells me how much needs to be = added or subtracted from the SW. Provided I'm adding or subtracting = weight from the molding itself, this value is identical to the change in = weight of the entire hammer assembly. Therefore, this process can most = easily be performed by weighing the hammer (and shank and flange -- = higher than the SW, of course), and adjusting the *total* weight by the = same amount. This wouldn't involve the more laborious process of = positioning the hammer "just so," with the shank level and the flange on = a stationary support, sticking up exactly 90 deg. Instead, I can simply = put the entire hammer on the tray. Easy. This process is made easier = if I simply tare the balance to the old weight, but I still need to = record the original weight, just in case time runs out, and my balance = shuts down. ;-) > > So is this something that is the way it is just because of tradition = > > -- because the cauls are built that way, and that's what ya' get? >=20 > It seems reasonable to assume that the chosen manufactured shape of a=20 > set of hammers would directly affect the basic shape of the curve. =20 > Cutting of hammer sets into individual hammers no doubt accounts for=20 > some of the spikes... as does variations in wood density... and = probably=20 > a lot of other things I havent really thought about. I guess I'm thinking more of the linear proportions in the dimensions of = the heads. If I'm not mistaken, you could take a slab of uncut hammers = and sight a straight line along all surfaces. Correct? (Ray??) = Interestingly, the entire keyframe is laid out in straight lines as = well, with longer dimensions in the bass and shorter in the treble. Ray = explains that the felt in the bass hammers is less dense than in the = treble, so that would explain the basic form of the SW curve. With = linear dimensions all around, and with progressively decreasing felt = density, combined with increasing proportions of felt with the = larger/lower hammers, there would actually be a fall-off in the bass end = SW. I would think that if hammers were to be matched in mass to the strings = they're hitting, the hammers in the high tenor and low treble would be = much lighter. But then again, consider the entire range of the piano. = It's HUGE. There's a LOT of difference between the mass of the A0 and = C8 strings, and this difference is so large that matching hammer mass = would be an entirely impractical venture. We'd be hitting C8 with = paperclips and A0 with bowling balls (exaggerating, of course). This of = course leads me back to the question of whether the shape of the curve = really means anything. In consideration of the range of the piano, a = concave downwards, Stanwood-type curve isn't really much better or = worse-matched to reality than a linear relationship. So perhaps the = best approach is to simply even out what comes out of the box and work = from there. Hmmmm? I indeed wanted between a Stanwood #8 and #9, and that's pretty much = what I got, throughout the bulk of the piano's range. I'm not going to = get too worked up if I'm in the #10 range in the high treble. After = all, those notes are pretty easy to play anyway! ;-) I think it's all = going to come down to tone. Whatever needs doing, that will dictate the = shape of the curve (e.g. if some felt removal is ultimately needed here = or there, *BUT ONLY WITH THE APPROVAL AND GUIDANCE OF MY MASTER-SAN*). = My job will be to make it smooth. Taking Isaac's advice about touch and = tone being psychoacoustically/perceptually linked (which I highly = suspect is the case), I imagine I'll be doing a bit of back-and-forth = between touch and tone, until I reach the optimum via the process of = successive approximation. However, I'm going to keep this work to a = minimum until AFTER I rescale/restring!=20 Meanwhile, I would still love to know if there's any magic to the shape = of these Stanwood curves (David??) or whether they largely reflect the = natural curves of hammer sets, resulting from the way they are = manufactured (as I suspect). Thanks for the input, y'all! Interesting stuff, to be sure! :-) Peace, Sarah ------=_NextPart_001_002D_01C48841.6CC09A30 An HTML attachment was scrubbed... URL: https://www.moypiano.com/ptg/pianotech.php/attachments/58/46/6e/10/attachment.htm ------=_NextPart_001_002D_01C48841.6CC09A30-- ---------------------- multipart/related attachment A non-text attachment was scrubbed... Name: not available Type: image/gif Size: 11299 bytes Desc: not available Url : https://www.moypiano.com/ptg/pianotech.php/attachments/58/2c/f0/3f/attachment.gif ---------------------- multipart/related attachment--
This PTG archive page provided courtesy of Moy Piano Service, LLC