[ Editors note:
 [
 [ Our new member Frank Himpsl collects pianos and piano rolls, and is a
 [ fine pianist and roll arranger as well.  Since he is also a chemical
 [ engineer I asked if he would write about the physical chemistry which
 [ bonds old rubber to the brass tubes of the tracker bar.
 [
 [ Robbie Rhodes

My guess is that the reaction is between the sulfur contained in the old
rubber and the brass surface of the tracker nipples.  Sulfur atoms in the
rubber bond the polymer strands together; actually this is the mechanism
which gives rubber its elasticity in the first place.  When the rubber
"ages" and loses elasticity, what is actually happening is that the sulfur
bridging bonds are becoming oxidized by air contact, or by contact with a
foreign substance such as dirt. or in this case, metal.  Once the sulfur
reacts in some way the elasticity is lost and the material condenses and
hardens into a polymeric mass.

In your collection you probably have rolls wrapped with a rubber band but
not played in many years.  The rubber band may still be slightly soft and
not completely petrified, but as you pull it away from the metal "D" ring
on the roll tab you'll see it has bonded to the metal.  That's the
reaction between the bridging sulfur atoms and the steel ring.

I'd imagine that there were dozens of types of rubber used for tracker
bar tubing, and the sulfur content of the rubber would vary.  This would
explain why sometimes it's possible to slip the tube off easily: the
rubber originally had low sulfur content.

Brass is a copper/zinc alloy, and both of these metals readily form
sulfides.  No organic solvent is going to remove copper or zinc sulfide
from the metal surface.  What you need is dilute hydrochloric acid
(sometimes called technical grade muriatic acid in hardware stores).
But "Noxon" does a nice job and is a lot safer to use!

Of course, you need to get the petrified rubber off the nipples first, and
for this I think a solvent ester would work the best.  Ethyl acetate,
which is bottled and sold as nail polish remover (at least, it used to
be), would probably work very nicely.   You'd have to soak the tracker bar
for awhile, and then the old rubber should become very soft.

Gasoline and chlorinated hydrocarbon solvents won't work as well as ethyl
acetate on rubber.  They might soften it eventually, but most will be
removed by scraping and not chemical dissolution.   Once the petrified
rubber is off, you can remove the surface sulfides from the brass with
Noxon or HCl.

Frank Himpsl

P.S.  Thanks for your letter and the Midi file.  But how do I "unwrap"
it?  The only experience I have in doing this is via "pkzip," which I
downloaded from the internet once.  Nothing I seem to have on the
computer will let me read the Midi file so I guess it needs decompression.
I never heard of BinHex4.

 [ E-mail transmission does not guarantee 8-bit data such as in Midi
 [ files, so I used BinHex4 encoding on my Macintosh to send the file to
 [ Frank.  Evidently his e-mail program has no built-in features to handle
 [ encoded ("wrapped") attachments, but I don't know what stand-alone
 [ encoding & decoding programs are currently available for the Mac and
 [ PC.
 [
 [ Could someone help us with this problem?  What e-mail programs
 [ will handle attachments automatically?     -- Robbie