fatigue data parameters

[tiguy64]

Hello,
Does anybody have a source for finding fatigue strength and fatigue ductility coefficients and exponents.

Thanks

 

[rb1957]

since you're in the FEA forum, somewhere like ncode might be a place to start.

Quando Omni Flunkus Moritati

 

[loki3000]

by testing the actual batch, used in a real product.

 

[realtime2]

Testing the batch is the best way, but a test set typically costs $12K+
to run. Many of the nCode supplied constants come from the FDE Comm.
where Bruce Boardman from J.Deere compiled sets of constants for
common materials. He published a table in an SAE paper. or J1099(?)
Results were from literature, steel suppliers, etc
of batch tests. Some material sets are also found here:

http://fde.uwaterloo.ca/Fde/Materials/dindex.html

Another source are a set of books by A.Baeumel Jr and T.Seeger e.g.: ISBN 0 444 41685 4
with a corrections book at ISBN 0 444 88603 6 Publisher is Elsevier.
Be warned that the parameters are from a least sq. fit process and do not
always fit the data. Whenever possible you should always plot the test
data points and the "fit" curve to see if it applies in the life region you are
interested in and to find where extrapolation has occurred.
Also be warned that when you have a mixture of small cycles interspersed
with occasional large load cycles that you can get damage by cycles below
the fatigue limit.

 

[bxtsafe]

Thanks realtime2 for contribute with this very usefull fatigue material database. The professionals at Waterloo university did a great job.

I also reccommend to use the fatigue properties database created and managed by Darrell Socie:

http://www.efatigue.com

. There are dozens of material properties registered and also a fatigue calculator (that I think is a valuable feature) for S-N, eps-N or Crack Growth Method. It's very valuable.

Last but not least, there also is the fatigue database of Jan Papuga website:

http://www.pragtic.com/vmat.php

. This is the database that is inside of the PragTic Software.

Chears

 

[jagad5]

Verify that scatter is considered in the properties that are used. One popular method is to predict life with the typical properties and then divide the predicted life by a scatter factor. I've seen values of 4 used. I prefer to compute the one-sided lower tolerance bound, which shifts the fatigue curve to the left by a multiple of the standard error of the estimate. That multiple is a function of sample size, probability of exceedance and condifence. A-Basis in MMPDS is 99% probability of exceedance with 95% confidence. Google Wirsching, Shen and Cashman for some info on this approach.

Doug

 

[realtime2]

One or two more web sites with fatigue parameters information:

AISI has tested about 120 sets of various bar steels for shafts,
gears etc, including differing heat treat and carburizing methods:

http://barsteelfatigue.autosteel.org/login_bar.asp

There seems to be another at TH Darmstadt which is probably related to the
Seeger books but I don't know anything about it yet:

http://www.wm.tu-darmstadt.de/mat-db/search.php