Crack Propagation if no load 1

[thermmech]

Hello there,

I apologize in advance if I am asking a question that has been answered elsewhere on the forum or is really easy to get the answer to - I am not a materials engineer and need some guidance.

We had a failure of a CS strainer bucket
- strainer nom dia 32"
- strainer body = forged CS
- service: mostly C1 @ 65 bar; C1 is clean, no H2o, no H2S, no HG
- op. temperature = 15-30°C, no fluctuations of the temperature or flow other than those caused by upstream mol sieve bed changever (dT = 8°C every 2.5h)
- there is a stopper ring inside the strainer, material SA-516-70 plate rolled the hard way and continuously fillet welded to the inside of the strainer body, purpose is prevent the strainer bucket from moving.

The strainer bucket was improperly installed (not tight against the stopper ring) which led to bucket rattling inside the strainer body and bucket failure.

This rattling also created 2 cracks on the stopper ring itself,
- location: on the edge of the fillet weld HAZ
- orientation: circumferential along the perimeter of the ring
- crack L = 30mm
- stopper ring t = 10mm, W = 75mm

We have already decided not to repair the strainer bucket, and NOT to return the bucket inside the strainer tee. We also have the following options:
1. Drill holes at the tips of the cracks to stop propagation of the cracks (difficult access for welded repair, need to isolate for hot work, needs time), or
2. Box up the strainer and start up the unit to avoid more downtime.

The questions are:
- Do we have to drill those holes regardless of whether the ring/cracks will be exposed to the load from the bucket or not? - What is the likelihood of those cracks propagating further through the ring if the strainer bucket (mechanical load) is removed? Ring failure would cause no concerns as its function is strictly to restrain the strainer bucket.
- In case of ring failure what is the likelihood of further crack propagation into the body (piping tee) of the strainer? Is this a credible scenario?

Thanks!

 

[EdStainless]

The cracks are probably fatigue from the vibration. Finding the leading tip to drill an arrest hole can be tricky, you will need to make sure that the hole in large enough also.

There is no such thing as 'no stress'. either from flow past them or residual stress from manufacture you have to believe that there will be some stress. If you put them back in the need some crack control.
And a plan for re-inspection soon. You need to know that you got the crack stopped.


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

 

[metengr]

The questions are:
- Do we have to drill those holes regardless of whether the ring/cracks will be exposed to the load from the bucket or not? -

No. Using stop holes would be ineffective in this application.

What is the likelihood of those cracks propagating further through the ring if the strainer bucket (mechanical load) is removed?

Low. If you remove the main source or load stimilus, the rate of cracking should be significantly reduced given the orientation and operating conditions. It will not be zero.

Ring failure would cause no concerns as its function is strictly to restrain the strainer bucket.
- In case of ring failure what is the likelihood of further crack propagation into the body (piping tee) of the strainer? Is this a credible scenario?

I would guess you would loose pieces of the retainer ring before you would propagate cracks into the main body.

 

[thermmech]

Gents,
Thanks on your replies. Just a request for a clarification to Metengr:

Quote:
Ring failure would cause no concerns as its function is strictly to restrain the strainer bucket.
- In case of ring failure what is the likelihood of further crack propagation into the body (piping tee) of the strainer? Is this a credible scenario?

Response:
I would guess you would loose pieces of the retainer ring before you would propagate cracks into the main body.

Am I understanding your comment properly that the propagation of the circumferential crack from the stopper ring (structural part) to the pressure bearing part is possible, but very unlikely?

 

[metengr]

Yes, that is what I mentioned.

 

[thermmech]

Thanks again.