about use of Goodman fatigue diagram

[sachin46]

Hi friends,
I am working on propeller analysis, where I have to calculate safety factor for NiAl bronze material in order to check its strength.
After reading various materials on it, I understand that Haigh diagram (Goodman line) is generally used for it. I have referred same diagram for NiAl bronze material and calculated factor of safety. But for this I have one doubt that, in diagram which I have referred there is line named as “proportional limit” wheather it is same as “Endurance limit line or goodman line”.
Kindly help me to resolve this, I have attached referred diagram
Thanking you
sachin

 

[metengr]

The line you refer to is the Goodman line, which illustrates the combined effects of stress amplitude and mean stress on fatigue life.

 

[tbuelna]

sachin46-

Your non-ferrous NiAl bronze material is a bit different than a ferrous material in terms of fatigue properties, and it doesn't have an endurance limit. The fatigue strength is usually defined at somewhere around 10^8 cycles, as shown by the blue curve in your chart. As metengr noted, the Goodman diagram takes into account factors that the S-N diagram does not.

You mentioned this was a propeller design, so you need to make sure your material fatigue data is appropriate for the application. For example, material like your NiAl bronze can have different fatigue properties when loaded in the longitudinal or transverse direction. The method used to produce the raw material or component pre-form (casting, forging, etc) can affect fatigue properties. Stress ratio and stress concentrations can also have a large influence. So if you don't have access to fatigue data for the correct material form, loading condition, R, Kt, etc, then you'll need to give some thought on what analysis factors to apply to the fatigue data you do have to compensate for these variables.

Fatigue analyses can get quite complicated if an accurate result is required. The alternative is to use very conservative analysis factors, which will usually result in an over-designed part.

Hope that helps.
Terry

 

[sachin46]

hi tbuelna,

Thank you very much for your post.I understood that the for non ferrous metal it doesn't have endurance limit line.
But the the diagram I am refreering has two 10 ^ 8 cycles, and the designer has given limit of proportionality 70 % yield stress.So I am calculating factor of safety considering that line, with mean and amplitude stress 74.6 N/mm2 and 26.7 N/mm2
respectilvely, I have attached refreeed line for your refrence.
Thanking you.
Sachin

 

[sachin46]

Does any one knows fatigue strength of Ni Al bronze metal ?

 

[bxtsafe]

Sachin,

See this topic:

http://www.eng-tips.com/viewthread.cfm?qid=365623

Although, Ni-Al metal is a singular metal. So, I strongly recommend to get the fatigue properties of this metal through testing.

 

[sachin46]

Dear Bxtsafe ,
Thank you very much for your reply.
Actually metal testing facility is not availble currently, So it is difficult to get it.
mean while I got one reference for material properties though some literature, it is saying that NiAl bronze material corresponds with standards "ASTM B 148" with Fatigue strength at 10⁸ is 214 MPa, wheather it is appropriate to use same?
Also designer has used fatigue diagram (attachment for thread 1 & 4) with mean and amplitude stress 74.6 N/mm2 and 26.7 N/mm2 respectively and based on this he has calculated "factor of saftey =1.8" ,but if we use Goodman diagram and try to find out FS then it is coming arround 3.97, with (Se=214 Mpa and S_uts =585 Mpa.)
Kindly give me feedback on calculation.
Thinking you in advance.

 

[bxtsafe]

I recommend to continue using the curve called "10⁸ cycles FS=1.8" and don't use the curve called "proportional limit".

However, I guess the curve "10⁸ cycles FS=1.8" was obtained by testing and therefore is a fitting of the experimental data. In other words, this mean that this is the mean curve (50% of failure). So, it's recommended to correct this curve by the reliability factor (see =>

http://www.roymech.co.uk/Useful_Tables/Fatigue/FAT_Mod_factors.html)

 

[EdStainless]

The proportional limit is fine for non-fatigue applications.
The 10^8 curve is the very maximum that should be considered for fatigue service.
But you need to de-rate it (such as the 1.8 S.F.) to allow for variations in material and scatter in fatigue results.
I honestly don't think that the 1.8 S.F. line is nearly conservative enough.

What design life are you looking for?
I would suggest that you read up on the Soderberg approach, it is better suited to this application.
You might try: Marine Propellers and Propulsion By John Carlton


= = = = = = = = = = = = = = = = = = = =
Plymouth Tube

 

[tbuelna]

I believe ASTM B148 is sand cast material. The mechanical properties in a sand cast material can vary greatly depending on the quality controls used to produce the casting. I don't recall seeing any published fatigue data for sand cast materials, likely due to this reason. Fatigue properties are based on testing a large number of representative samples for a huge number of cycles. The specific material properties resulting in any given sand cast component can vary greatly. So establishing accurate fatigue properties for a sand cast material is not practical.

With regards to analyzing basic tensile or torsional strength of sand cast components, there is typically a casting knock-down factor applied which varies depending on the application. For critical applications the casting knock-down factor might be 2.0.

 

[bxtsafe]

The FKM-Guideline (german guideline) recomend to use S_E=0.3*S_UTS for cast aluminum alloys. This value correspond to N_E=10⁶ cycles.

Please see attached figure (from pg 131 of book "Metal Fatigue Analysis Handbook"), from

http://books.google.com.br/books?id...BR&source=gbs_toc_r&cad=4#v=onepage&q&f=false

Unfortunately, FKM doens't mention properties of Ni alloys nor Cooper alloys.