Other web Sites
Harmonica Blues  Harmonica Amps
Harmonica Links Harmonica Pages
• Archives Home
• Years
 · 1992
 · 1993
 · 1994
 · 1995
 · 1996
 · 1997 		 · 1998
 · 1999
 · 2000
 · 2001
 · 2002
 · 2003
 
Web HarpL
Ebay Searches:
Amps:
Microphones:
Effects:
Harmonicas and Gear:
Harmonica Music and Instruction:

 

 

Harp-L Archives

[Previous Message] [Next Message]
[Next in Thread]
[Start of Thread] [End of Thread]

DATE: Fri, 25 Mar 1994 09:25:36 CST
From: Dick Anderson
Subject: Reed Fatigue

I have long been interested in the forces acting on reeds and damage caused
by the various legendary taboos such as: Not breaking in a reed, bending
on 5 draw, bending too loudly. I even read one report of reeds "healing"
themselves by aging under one guy's bed.

The first mechanical engineer I talked to mentioned the work hardening
theory of reed destruction, but I discounted that theory on the same basis
as others have mentioned that if the reed was getting stiffer, the note
produced would go up. Instead, we notice reeds going "flat" or lower as
they age and wear out.

Also, I think work hardening only happens when a metal is stressed beyond
it's elastic limits causing a permanent deformation. An example of bending a
coat hanger wire was given earlier, but I'm sure the writer was speaking of
permanent deformation, not an elastic cyclical motion of the wire.

If springs work hardened under elastic cycling, then automobile
coil springs which suffer small elastic deformations all their life should
get stiffer and stiffer until finally they crack and break. This is also
not seen in practice. Although I have seen leaf springs crack and break.

I have been reading about metal fatigue and am in quite over my head, but
from the texts I have been able to glean these few gems:


1. Cyclic stresses below the point of permanent deformation do cause shifting
of the atomic structure and eventually lead to crack formation and growth
which eventually leads to complete failure. (Reed broken off)

2. The larger the deformation and the higher the number of cycles, the sooner
that a crack will develop and grow.

3. Cracks always begin at the surface.

4. Crack formation can be reduced by stress relieving the surface of the
metal. (Shot peening is used to improve life of steel springs)

5. Crack formation can be reduced by improving the purity of the metal
avoiding inclusions and defects built into the material.


Bottom line: Reeds fail due to crack formation as the result of cyclical
elastic deformation over time. This is a classic metal fatigue
problem.

My Hunches: Bending notes on the harmonica causes the "opposite" reed to
vibrate outside of the slot in an "opening" manner. This may
cause greater stress on the reed since its vibrations could
be of higher amplitude than a reed vibrating in a "closing"
manner. (Johno, have you tested this theory?)

Harmonica manufacturers could improve the longevity of reeds
by stress relieveing the surface perhaps by micro-peening or
smoothing sharp edges left by grinding the reeds into tune.
(Hohner reeds have lots of small scratches left by the tuning
process while Suzuki "laser tuned" reeds are somewhat smoother)

Old wisdom: I spoke with an organ reed maker who said that the reeds in some
organs have lasted more than 100 years with consistent use.
He did warn me that the key to long life was to control the
air pressure to the reed and NEVER to blow into a reed to
test it since the pressure was not controlled. (Organ reeds
are "beating reeds" and not "free reeds" like a harmonica,
but the movement of the reed is still cyclical and elastic.)


My question to Johno or others:

Has any quantitative work been done to correlate reed life with:

A. Loudness of playing
B. Amount of bending
C. Lack of "break in"

Has any similar work been done to correlate guitar string
life with the factors above?

Love to talk harps!

dick....

--

Dick Anderson CCMO New Product Engineering Telnet 229-3110
Hewlett Packard Direct Dial 1-303-229-3110
3404 E Harmony Road HPDESK dick_anders~p4000
Fort Collins Colorado 80525 mail anders~c.hp.com