Stringers in 722M24 En40B Nitriding Steel

[RustyTrombone]

Hello all,

I'm a newbie to the forum and am in need of some help.
We have a customers crankshaft which we've tested on the Mag bench using the W/C method and we've been finding 'stringers'. Some defects are surface breaking whilst others are not. Some are grouped together but all follow the length of the shaft. How detrimental to the component are these subsurface defects? I've read various articles which would lead me to the conclusion that any of these defects are rejectable even though the customer would like to have a limit on defect size?
Any help would be appreciated.

 

[TVP]

Are the stringer defects being detected in the original bar, or in the final crankshaft? If it is the former, then typical practice is to limit the defect to some detectable size, say 0.13 mm maximum, that will be removed during subsequent machining/grinding. If it is the latter, then it is typical to reject anything that has a visual appearance during magnetic particle inspection.

 

[RustyTrombone]

Thanks for the reply. It's the final crankshaft so this would mean their rejection rate could be high. If they still insist on using this material is there any safe way of coming up with a limit for this type of defect in the final product?
Thanks.

 

[redpicker]

The effect of these imperfections on the product depends on their orientation with respect to the principle stresses. If their lay is parallel to the direction of principle stress, they will have negligible effect, but they are perpendicular to the principle, they can be very detrimental. In addition, if they are present in a case-hardened layer, they can cause problems regardless of the principle stresses from applied loads because residual stresses can significantly affect the direction of the principle stress in the case hardened layer.

Most of the acceptance criteria I've seen for such stringer imperfections usually address both a size (length) limit and a density (number found within a specified area) limit. For example, a maximum length of 0.500 inch for any single indication and no more than 3 indications over 0.063 inch in any 1.0 square inch area.

The actual values for the limits have to come from experience and/or trial and error. In the absence of experience, initial limits could be based on the consequence of failure combined with limits that would reject the worst cases while accepting minor groupings, and adjusting depending on performance.

rp

 

[RustyTrombone]

Thanks for the advice. I think it may be wise to section one and 'see' what is going on. The crank will be going into a very powerful engine so I'll use a bit of caution before setting a limit with the customer.