Rubbings and patterns

[John Delacour]

At 18:54 05/10/01 -0500, Ron Nossaman wrote:


>Incidentally, James Arledge's recommendation for string patterns strikes
>me as a real nice workable process. Take a rubbing of the agraffes, and a
>separate one of the hitches and bridge pins. Measure speaking lengths of
>the ends of the section(s), along with your core and wrap data, and send
>them to your string winder. It doesn't make half a hoot's difference what
>you use for the pattern, or what the dimensional stability of the stuff is,
>because you supply the string winder with the speaking length measurements
>so he can orient the top and bottom, or fore and aft, patterns on the bench
>to your end of section measurements, and exactly reproduce the scale
>lengths that appear in the piano.

Fair enough if postage is $10 a gram.  Let me tell you what I like to receive.

Take a length of brown paper, lining paper or thinnish wallpaper and cur or 
tear it so that it lies over the bass section nice and flat.  In the case 
of an upright you can tack the top end to the top of the piano to keep it 
still.

Having made sure all the pins are covered with the paper, first rub over 
the bridge pins using a coarsish abrasive paper such as triMite P180 so as 
to make holes on the paper over the pins.

Push the paper down onto the bridge so that the pins are sticking up 
through the neat holes they have made and the paper cannot now slip

Next carefully work over the hitchpins: while keeping the paper taut, smack 
the paper with the sandpaper until you get a nice small hole for each 
hitchpin; you don't want to make a rough hole big enough for the pin to 
pass through, just a tiny hole to show where the top of the pin is.

Now move to the top of the piano and, again keeping the paper taut, mark 
the top bridge line.  If you have top bridge pins, rub these through as you 
did the bridge.  If agraffes, smack the paper with the sandpaper to get the 
front corner of the agraffe to make a cut in the paper.  If you have a 
bridge and pressure bar, show where the bridge runs.  If you have a cast-in 
top bridge, go to the next stage.

Finally rub over the copper endings at each end to show where the strings 
are and where the copper terminates.  This is most important in the case of 
pressure bar or cast-in top bridge, since otherwise it is impossible to 
know exactly where the strings run.

If you don't remove grand dampers when taking the rubbing, there will be an 
error of a few mm.  This does not matter because you can simply write 
"rubbing taken over dampers..note 1 = 1379 mm.  Note 26 = 984 mm".  The 
stringmaker can just make a tuck in the rubbing to remove the error.

Send the first and last string of each section as patterns.  This allows 
the stringmaker to check the rubbing and make sure nothing has 
slipped.  Everyone has a bad day and can make a mistake.  The patterns show 
how much the strings stretch when pulled to pitch and they also show how 
they were originally made.  Whether or not you want strings like these 
patterns is not the point -- they are there only to provide the original 
measurements and to prove the rubbing.

In most cases it is not necessary to provide length measurements.  A 
well-taken rubbing is far more accurate than any measurements you can take 
with a measure and it doesn't make mistakes.  In the case above you can 
give two lengths and you can do this in other cases, but it should not be 
unnecessary, though paper can shrink a lot if left in the sun.

 From the stringmaker's point of view, the rubbing is the piano.  He can 
see it, he can lay the strings on it to test for errors etc. etc.  There 
are different ways of working, I know, and a lot of the German firms work 
to measurements, but I've heard these people tell me how long it takes them 
and it's incredible.  The risk of error is also high because there is no 
visual checking.

If I receive measurements only, I charge $40 extra, because I have to 
transfer those measurements to a pseudo-rubbing to check that you have got 
them right, which usually you haven't, and because I don't want the hassle 
of making the strings wrong, no matter whose fault it is.  I never need to 
blame a customer for my making a wrong string.  If I make a wrong string, 
it's my fault for not checking the data.

The rubbing, as I've said, is sure and efficient and enabled a set of 
strings to be made in about three hours from start to finish -- taking 
measurements, inputting to spreadsheet, checking patterns, modifying scale 
as necessary, cutting out and eye-making, flattening or swaging, putting on 
the copper, packing and invoicing.  Most Steinways and Bechsteins take even 
less time and only horrible baby grands sometimes take up to four hours.

If you do take measurements, it's the speaking length that matters.  Unless 
you know all the tensions and the properties of the wire, you can't know 
how much the strings will elongate.  Any good stringmaker nowadays ought to 
have his program calculate the stretch and from this be able to give you a 
perfectly straight copper line.  If your note 13 single is 4 mm closer to 
the agraffe than your note 14 bichords, it's because it has stretched more 
owing to the jump in tension, not compensated for by the fatter wire.

As I've said, there are different schools of string-making, so I've told 
you what happens chez moi, which is English traditional with added 
safeguards, certain secret tools and tricks, and, of course, computer 
aid.  It's a mixture I haven't seen bettered.

JD