Re: coaxing and strain aging HarpL Archives

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

Ebay Searches:

Amps:
Microphones:
Effects:
Harmonicas and Gear:
Harmonica Music and Instruction:

Harp-L Archives

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

From: cryforhelp
Date: Fri, 18 Oct 1996 00:08:46 -0700
Subject: Re: coaxing and strain aging

WVE~ol.com wrote:
>
> In a message dated 96-10-14 16:13:35 EDT, jdd writes:
>
> >Which it still might be. I don't think that we have yet devoted
> >sufficient attention to the issues of riveting and stamping/finishing of
> >the reed plate. I worked for a number of years for a company which made
> >a variety of small oscillating mirrors which mounted on ferrous reeds
> >vibrating in the 200Hz to 2kHz region. Brinnelling, scraping, spalling
> >and possibly even electrochemical effects on the reed at the clamping
> >point were nagging problems. Brinneling is likely restricted only to
> >ferrous alloys, but scraping of the reed on nearly invisible stamping
> >dross could be one possible failure mechanism. Loosening of the rivet
> >compression is another candidate. I'm not suggesting anything
> >definitive, only that as I'm sure you'll agree there are still
> >second/third/fourth order effects to be explored which might account for
> >observed reed break-in.
> >
> You mention problems that might account for failure. Did any of them improve
> with gentle cycling when the part was new?
>
> Vern

That wasn't the object of investigation of these *ferrous* flexures.
Design parameters were delibarately chosen to keep the strain on the
flexure at all times below the "normal" Young's modulus of the steel
material (coaxing presumably raises this modulus above its commonly
accepted "normal" level); in theory the flexures should have had
infinite life, something which I think is not possible in CuTn alloys.
Failure was so dramatic because it was happening during infant burn-in
for something which should have had at least tens of thousands of hours
of life.

The engineers decided in this case that the principal culprit was
brinnelling at the exact point of contact between the flexure and its
clamp (the flexure was held exactly as you would hold a small strip of
metal sticking out from the jaws of a vise). One can argue whether the
mechanism of the brinnelling was dissolution of carbon in the interior
of the steel's grains and re-precipitation at the grain boundaries as
you suggested Vern, or something else, but I believe that testing did
verify an increase in Rockwell "C" at the clamping point.

So the problem of our mirror suspension flexures was not specifically
applicable to harp reeds. Also note that the edges of the "vise jaws"
holding the flexure were highly polished, very different from the casual
die stampling process used to produce harp reed plates. And of course
the vise-jaw attachment technique is itself at least an order of
magnitude more positive than the simple upset rivet used to attach harp
reeds to their plate.

What I was trying to point out was the impact of nth order effects on
system durability. In harps consider that the reed plate is not
polished and micro-dross edges are possible and can appear on the insert
side of the stamping as well as on the outlet side, especially in soft
materials such as brasses as the die pulls material back up through the
hole on its upstroke (got a good 400x toolmaker's microscope handy?)
And I am definitely inclined to wonder about those brass rivets-- how
strong can a brass rivet be? Which could be a positive. I presume
Hohner has modeled the stress on a gram or two of brass read 1.5 cm long
by only 20 - 40 microns in thickness vibrating at, say, 1 KHz. It must
be a significant stress at the rivet point. If the rivet is infinitely
strong then there must be two highly concentrated areas of stress on the
reed stem during vibration-- at the leading edge of the rivet on the
upward oscillation, and against the lip of the square hole in the reed
plate on the downard cycle. But if the rivet were allowed to gradually
loosen its compression (during "break-in?"), allowing the reed to
"float" (very slightly, perhaps by a micron or two) against the rivet
during its cycling I would imagine that the stress concentration could
be more widely spread out. HIGHLY SPECULATIVE, but possible.

Our engineers thinly layered in an RTV silicone between the "jaws of the
vise" and the flexure. This reduced stress concentrations which in the
ferrous flexure may have been the cause of brinnelling. Might it have
some impact on other nth order effects applicable to brass flexures as
well? I don't know, but you seem interested in harp engineering, which
is why I offer this ridiculously long-winded explanation, in hopes that
it might contain one little kernal of interest. My appologies for
making you wade through all the chaff.

- --cryforhelp--

TFTD: There is no force, however great
can draw a string, however fine
into a horizontal line
which is accurately straight.