Alternate to static hydrotesting of API tanks 2

[vesselguy]

Although I have designed and purchased storage tanks on and off since 1998, I have not heard of the possiblity of not doing a static hydro test as the final leak testing of a NEW tank. In the last 2 years, I have seen odd specification to perform fracture mechanics analysis to determine critical surface flaw size per API579 in lieu of hydrotesting. I can buy into doing this for existing tanks but I cannot accept this as a replacement for hydrotesting.

Can someone enlighten me on this issue? Is this acceptable alternative to the API650? How is this done? Is this done often now?

Thanks.

 

[metengr]

There is no provision in API 650 for using fracture mechanics analysis in lieu of hydrostatic testing. I am in the process of finalizing a bid specification for a storage tank, and after reviewing API 650, I saw no exemptions.

Frankly, I don't see how anyone would use this approach (API 579 FFS) for substituting a hydrostatic test for a new construction tank. After all, the purpose of a hydrostatic test is twofold - check for leaks from workmanship and to assure structural integrity. In other words, the tank will hold the specified fluid without collapse.

Using your alternative approach, what would fracture mechanics provide? It would only provide the maximum flaw that could be tolerated without unstable crack growth (aka brittle fracture). You can still have a gross defect or subsequent leak caused by poor workmanship regardless of using fracture mechanics analysis.

So, my point is this, hydrostatic testing is required for a new vessel fabricated under API 650. One alternative that I could possibly entertain in lieu of hydrostatic testing is comprehensive NDT of all shell and floor plate long seam and girth welds.

 

[vesselguy]

metengr,

I totally agree with what you said. My thinking is exactly the same as yours. I too don't see this as a alternate specified in API650. What troubles me is I've seen this "alternate" in a company spec requesting this "to be considered".

I would like to know if this is something that is being used on recent projects.

 

[JStephen]

API does allow an exception to the hydrotest requirement, see 7.3.5 paragraph 2, but it's not related to fracture mechanics.

 

[unclesyd]

I heard the same thing last summer when an API 650 tank was being constructed. Here is where the exemption idea may have came from. There have been a couple of papers like this one.

http://www.tank-consultants.com/download_documents/OGJ_20000131.pdf

I see no way to exempt a newly constructed tank from a hydrotest The last two large tanks, a SS and a CS, I was involved with both had several leaks. The SS had 2 leaks on the T-joint due lack of fusion on a poorly prepared joint.

 

[vesselguy]

Thanks unclesyd for this info. This is the kind of info I want to know.

 

[vesselguy]

Unclesyd,
I took a quick look at the paper you linked to and it is a discussion of API579. It is plausible to use that as an alternative for hydrotest for repair in an existing tank, but I don't see how someone could possibly think it can be stretched to use it as an alternative for a NEW tank.

Anyhow, I'm convinced that I cannot accept the proposal in the specs that I'm reviewing. Thanks.

 

[unclesyd]

As the politicians always cop out by saying "That information was taken out of context".

I should have stated that the misinformation probably came from an article such as this. In today's environment there is no way I would accept a site built tank without the proper hydro. If I was forced to my objections would be covered all walls at the site and probably a little graffiti on the tank.
As I stated in my previous post the last two tanks I was involved with both leaked.

 

[r6155]

In general (tanks and pressure vessels) the purpose of "hydrostatic test" is equal to "strength test", not for leak testing.
For API 650 tanks, the hydrostatic test if for control of foundation too.
Regards
r6155

 

[JStephen]

It actually serves both purposes. But note that if you have a product significantly heavier than water and/or elevated tank temperatures, the hydrotest doesn't accomplish the strengh-test, but is still required.

 

[unclesyd]

The old fashioned way to test for leaks and strength was to add a standpipe to a top nozzle to raise the water level to compensate for the higher specific gravity product.
In our case we had we had a in process product that had a SG of 1.34. We designed the whole tank for the fill line pressure and to test we added a standpipe to the center vapor nozzle.

 

[LOUDOG]

Depending on the size of the tank, an air test can be done to check for leaks.

I would be confortable with this for large shop built tanks (>1000 BBL). I have also seen it done on tanks as large as (40' dia x 60' high). Sometimes a clean source of water isn't availible for a hydrotest and other options need to be looked at.