Typical corrosion allowances. 2

[0707]

Piping designer, based on service corrosivity, adds corrosion allowances to the calculated pipe thickness. In a new oil refinery construction project, which would be typical corrosion allowances?

Is there any rule to be followed?

Thanks

Luis

 

[BigInch]

I'm going to wait to see who tries this one.

http://virtualpipeline.spaces.msn.com

 

[25362]

From Chapter 20 of Norman P. Lieberman's Troubleshooting Refinery Processes (Penwell), which I recommend reading:

A corrosion rate which reduces a pipe wall to half its original thickness after 10 to 15 years is a reasonable target for common processes such as crude running, coking, fluid cat. cracking, gas plants, alkylation units, etc.

According to this criterion, for a CS pipe carrying high-sulfur diesel at 260^oC with an original wall thickness of [¼] in. or 250 mils (mil = 1/1000 inch), a uniform loss of 10 mil/year would be OK. This rate can be known from previous (published) experience or measured with coupons, probes or suitable instruments. It means the corrosion allowance would be 12.5 years [×] 10 mil/year = 125 mils = 0.125 in.

There are many different causes and types of corrosion in a petroleum refinery. Each involves a different material wastage rate (not always wall thinning) which depends on the corroding environment (chemistry, hydraulics, temperature, pressure, etc.), the selected metallurgy, and even welding and other preliminary procedures.

 

[BigInch]

Star for you 25362.

high-sulfur diesel at 260ºC is a good place to start.

http://virtualpipeline.spaces.msn.com

 

[0707]

For Carbon Steel piping in refinery service, a typical C.A. assumed and given will be = 1.5 mm (1/16"). In some extreme cases, a C.A. of 1/8" or 3.2 mm is given (but not as a general rule). For high temperature services on bottom cuts of corrosive hydrocarbon services should we use 6.35mm corrosion allowances?
From my search on this subject I suppose that there is no rule or recommended practice and the things are a little bit established ad hoc empirical based on licensors experience.
I think that actually in a new refinery construction a RBI approach should be made to define expectable corrosion rates then based on those expectable corrosion rates designer should add corrosion allowances with the following criteria: high corrosion rates high corrosion allowance, medium corrosion rates medium corrosion allowance, low corrosion rates low corrosion allowances.
Thanks for you sharing
Luis

 

[25362]

BigInch, thanks. A big hand on your being elected TMW by the community.

Luis, creo que sus conclusiones son muy razonables.