Unforeseen fatigue issue

[trainguy]

Hello all.

I have a situation here. Imagine a railcar truck (the steel casting that connects the wheelsets to the locomotive) having been in service for 40 years.

The truck has been subjected to fatigue loading all this time. It's analysis data is unknown, so the stress range is not known, neither are any of the fatigue category parameters.

Then, someone comes along and adds a welded bracket, for which our office is doing the design. This bracket will support a damper which is again a fatigue structure.

What is the common approach in this kind of retrofit? Clearly we are taking care of all the bracket fatigue issues, but what is a practical approach to making sure there are no surprises (cracks in supporting structure)down the line?

Is it common practice to redo the full structural fatigue analysis including the effects of the added bracket?

Thanks in advance.

tg

 

[boo1]

Your safety program should specify that if it changes the oprerating parameters, analysis is required.

 

[rb1957]

i'd say that you have to consider your added structure in some way. i'd assume that redoing the complete fatgiue stress analysis is impractical, and probably not warranted. if you've got the original analysis then you've got an ideal starting point. presumably you know the loads that your new damper structure is going to add to the car structure. you can follow these loads as far into the supporting structure as you want to go; at some stage you'll say the impact of the damper is negligible, maybe this is the interface with the supporting structure (the weld), maybe when you see the numbers you'll want to go a little further.

if you haven't got the original stress analysis, and can't estimate a working stress, then i guess you're left conservatively assuming the maximum stress for an infinite life.

consider too the effect of your weld on the fatigue life of the structure.

good luck !

 

[metengr]

trainguy;
We have dealt with bolster castings for our new and existing fleet of rail coal cars with regards to potential failures in service. As you are aware failures of bolsters can cause coal train derailments.

Why is the bracket being added? Is there a problem with fatigue crack propagation in service or is the bracket being added to allow for extra hauling capacity?

One of the foundries that manufactures rail car bolster castings and side frames in the US has a large fatigue testing machine that can simulate loading of a full bolster casting in service. We had one of the prototype castings tested for a cost of $10k. You might want to consider similar testing regarding your design.

 

[swall]

You should consider the possibility of casting defects such as internal shrinkage in the area of interest. The casting may have been fine for 40 years service with existing casting flaws, but you may have a problem if your bracket ends up becoming a stress riser right at a casting flaw. Some ultrasonic inspection or even X-ray would seem prudent.

 

[trainguy]

rb1957:

Please forgive my fatigue inexperience, but here goes:
Can you clarify what you meant by assuming the maximum stress for an infinite life? If the existing (pre-bracket) working stress is the maximum stress for infinite life, then I can't add any stress, as I understand it, so I couldn't add the bracket. Is this right?

My concern is that even without adding too much stress, I am degrading the fatigue category near the new welds.

metengr:

The bracket is being added to improve the ride quality for loco. operators by dampening yaw motions in the truck. It's a passenger locomotive trying to be used at higher operating speeds.

tg

 

[rb1957]

i was thinking of the endurance limit. maybe you train guys have some other design stress for infinite life. and yes you're correct; if you make this assumption then you are going to end up with a life limited structure (you can still add the brkt) this is just the consequence of not having the original analysis ... maybe the incremenent in stress is only a small fraction of the assumed fatigue stress and could be acceptable.

 

[Jordonlaw]

Trainguy,

Here's the key statement :-

"The bracket is being added to improve the ride quality for loco. operators by dampening yaw motions in the truck. It's a passenger locomotive trying to be used at higher operating speeds."

Higher speeds = higher loads

Need I say more ?

 

[CoryPad]

It seems like the way to go here is using a Damage Tolerance method. You fracture mechanics to calculate a critical flaw size, and you use some inspection method to test for flaws.

Regards,

Cory

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[Philrock]

Welds are not great in fatigue applicatons. It is often assumed that if a weld is not stress relieved, half of the material in the weld is at yield in compression, and half is at yield in tension. In this case some designers limit theoretical stress to 2 ksi, ignoring the residual stresses due to welding.

 

[NickE]

I have to agree with Philrock, The weld is going to be a huge amount of trouble.

What material is the casting made of?