316 Blind Flange on a Super Duplex Line 1

[MechEngJM]

Hi All,

I have a 1" super duplex (ASTM A790) pipe that has had a 316 blind flange (150lb) installed. Piping spec calls for A182 Gr F55 to ASME B16.5, but for various reasons, this temporary solution was used. There has now been a request to extend this temporary solution to circa 1 year (mod made 2 weeks ago).

Additional materials:

[ul]
[li]Gaskets are SPWD 6 Mo, bolting is A276.[/li]

[li]Process medium is produced water at atmospheric pressure and ambient temperature (0 degC - 15 degC)[/li]
[/ul]


Obviously, I am concerned about the potential for galvanic corrosion but I am fairly new in my role and lack the experience to have an informed answer off the top of my head -- hoping for some insights here?

Thanks,

J.

 

[EdStainless]

What is the service fluid?
If this is high Cl water then the 316 may not last a year.
I wouldn't let them do it.


= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[marty007]

If this is an issue of sourcing the correct flange material, one option could be to use a full diameter gasket. This would isolate the flange material from the process fluid...

 

[MechEngJM]

The fluid is produced water from a sour reservoir that has been through the standard water plant of multistage separation, hydrocyolones and skimmers and is now in-spec for overboarding. So yes -- high Cl content.

A full diameter gasket may be an option, but currently I'm looking to establish the risks of running with the current SDSS-316SS setup. I've been hunting for something quantitative on the compatibility of A790 and 316 but seeing conflicting information.

Appreciate the responses!

 

[SJones]

At worst, the failure mode will be leakage as stresses and temperatures probably aren't high enough to drive a crack. What are the consequences of a small leak in a DN 25, class 150, treated water system at the location of the flange? Not high, I would guess. The actual risk will come from the end user's risk assessment protocols. You will be unlikely to find anything 'quantitative' if you are seeking to compute a wastage rate. It's purely qualitative and subjective.

Steve Jones
Corrosion Management Consultant

http://www.linkedin.com/in/drstevejones

All answers are personal opinions only and are in no way connected with any employer.

 

[MechEngJM]

Thanks for the feedback. Perhaps 'quantitative' is too strong a word. What I was looking for was something like a galvanic series with a given p.d. range relative to SCE for both SDSS and 316SS, allowing me to state something like 'p.d. < ~0.15v hence risk of corrosion low'... if that happened to be the case.

 

[SJones]

You will readily find a galvanic series in seawater, but not facility-specific produced water. The risk won't be low; it's the likelihood that will be low. You need to know the consequences to assess risk. As it stands, you won't have a negligible likelihood, but it won't be a certainty either. The expectation of leak size would be pinhole around the gasket. So what would be the consequences of a pinhole leak of produced water at the location? Take a stab at lowish likelihood, say "heard of in the industry," evaluate the consequences with the process safety guys, pin it on the risk matrix and see whether it's a tolerable risk. If it isn't, then clearly change the material immediately. If it is, just make sure that it's an anomaly flagged in the CMMS so that it doesn't drop off the radar, and get it changed out at the first opportunity.



Steve Jones
Corrosion Management Consultant

http://www.linkedin.com/in/drstevejones

All answers are personal opinions only and are in no way connected with any employer.

 

[moltenmetal]

It's a blind flange. Install a corrosion resistant, nonconductive gasket which covers the entire flange face so that none of the 316 material is wetted. Unless the flange is immersed externally, your problem would then be solved- assuming that the presumably higher flange rating of the 150# duplex lines relative to 316SS (I didn't check the B16.5 table to see if this is the case) is not meaningfully being relied on in this design..

 

[MechEngJM]

Thanks all - replies appreciated.