[Harp-L] Re: Steel Reeds



Hi,

I'm a long-time subscriber to harp-l (since 1998) and enjoy very much the discussions here, although I don't post very often. However, this issue with stainless-steel reeds interests me very much.

Here is my experience: I bought a Seydel 1847 in G a year ago and after half a year of (much) usage the 4-draw reed (A4) became badly out of tune - almost down to G4. I tried to tune it without success. In fact, close examination revealed a tiny crack ca 4 mm from the rivet. I contacted Seydel and they kindly offered to repair the harmonica. So two weeks after sending it, I had it back with the reed replaced. That's great service!

Now to my thoughts: Obviously, SS-reeds are prone to metal fatigue just as brass reeds. As I understand it, this can happen to any metal component that is subjected to a high number of deflections. This is very well explained in the article "Beyond the Basics - Performance Over Time: Fatigue Strength" that can be found here (I think someone posted this some time ago on the list but I don't remember who and when):

http://www.copper.org/applications/industrial/DesignGuide/homepage.html

Here is a quote from the article:

"Fatigue occurs because micro cracks develop on the metals surface when it is cyclically stressed. With repeated bending these cracks propagate through the metal thickness to a point where the remaining sound structure fails by ordinary rupture (because the load can no longer be supported)."

There is also a diagram which shows the relation between stress and number of cycles until fatigue sets in. The stress in turn is related to the deflection, which in the case of a harmonica reed depends on the blowing pressure. From the diagram one can conclude that a reed can last for a very large number of cycles if the stress (blowing pressure) is small. But if you increase the stress, the lifetime (number of cycles before metal fatigue sets in) can become drastically lower.

The diagram in the article is specifically for a certain brass alloy. But I guess that similar behavior can be expected for any metal. It is only a matter of *how many cycles* the component can last for a certain stress and geometry. Is there any metallurgist on the list that can verify this? In fact it would be very interesting to see a comparison between brass and stainless steel (with compositions similar to harmonica reeds).

Maybe someone with appropriate equipment can do an AB test? It would require blowing on reeds with a constant pressure until fatigue sets in, that is, when the pitch drops drastically. This might take several days. (100 million cycles of an A4 reed will take approx 63 hours!-)

By the way (as a comment to smo-joe), the thickness of the SS-reeds in the Seydel 1847 seems to be comparable to ordinary brass reeds (judging by eye - I haven't measured exactly).

Regards, Orjan
http://ohw.se

On Apr 15, 2008, at 11:11 PM, lil Buddha wrote:

I am not a metallurgist, but from what I have read, steel does not share the
same fatigue issues brass does.
Many metals will eventually fail even with the smallest loads; the effects
of the stress applied are cumulative. No so with steel. As long as the
stress levels are kept below a certain level, the steel will never break.
Theoretically at least. This would indicate that those who have caused
failure in the stainless steel reeds are blowing harder than necessary,
attaining a failure level instantly rather than cumulatively. Anything can
be broken. Failure of an object may be more the fault of the user in these
cases.


Forgive me if this was addressed and I missed it.
_

To which smo-joe replied:


A reed must be made from anything spring. Steel is better in this
regard, However, given the immense strength (by comparison) of steel,
it stands to reason that the reeds would have to be MUCH thinner.
Since there is so little material involved, ANY flaw will contribute
to a failure.

In an accordion, the reeds are huge (by comparison), and thus there
is less chance of failure. That, AND they have a pressure relief valve.
I agree with your synopsis

smo-joe






This archive was generated by a fusion of Pipermail 0.09 (Mailman edition) and MHonArc 2.6.8.