Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations IDS on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Refrigerated Ethylene transfer pump

Status
Not open for further replies.

Aquasition

Materials
Mar 2, 2016
2
This is a 13 stage pump sucking from about 250 psi and final discharge at around 1200 psi.
Is there a way to inspect the flat bottom plate weld without pulling the pump out of the casing?
Is there a way to detect a leak out of the suction can, into the insulation and annular space?
Thanks folks
 
Replies continue below

Recommended for you

We have a number of pumps of this style that are in light liquid service. Based on an industry event that involved a release, explosion and fire, we created an inspection program for the cans. We require that all of these cans be pulled out, blasted bare and fully inspected to establish their base-line condition. We have stopped short of requiring replacement of all of our flat bottom cans. But, we preferentially purchase ellipsoidal heads for any new cans purchased.

When we pull a can for inspection, we recoat it with our preferred pine-tar epoxy that is used for underground piping. We install a port between the can and the vault to check for VOC levels. We have our operators sniffing this space with a 4 gas analyzer monthly to check for leaks. We require that the can is pulled and re-inspected every 10 years.

All our cans of this type are in steel-lined concrete vaults with an ample vapor space between the can and the vault liner. I have heard of other plants where the can is buried in dirt. That would be a much greater concern that would require a more rigorous inspection plan.

Johnny Pellin
 
Thanks Johnny
It seems to me the baseline evaluation is a must, but 10 years inspection interval may be a little too long, so we are looking into using a full time 24/7 leak monitoring system. It won't detect crack initiation or propagation in a weld, but once a leak is established we think the 250 psig in the suction side will push fluid out of the can and through the thermal insulation fast enough to vaporize and cause turbulence, which is what the trsnducers would detect. it is supposed to be fast and in real time, but obviously this system would need to be calibrated for the high rpm impellers and other things. will keep you posted if you want. thanks again.
 
Our management has decided to change our plan. The cost to pull, blast, fully inspect and recoat the cans is nearly as much as the cost of a new can. So, we are replacing all of our flat-bottom cans with elipsoidal heads over the next two years. We will still pull them for inspection every ten years.

Johnny Pellin
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor