4 replies to this topic

[HK1234]

Good day,

This is my first post. I have been delighted at the level of insight and helpfulness that these forums provide, and I would hope someone will be able to assist me.

I am a process engineer verifying some gas flow data on a pilot project. We are utilising Swagelok Stainless Steel 316L grade instrumentation tubing on the site. My question pertains as to what the absolute roughness of the material might be.

From the Engineering Toolbox website (http://www.engineeri...ubes-d_459.html), stainless steel has an absolute roughness of 0.015 x 10^-03 m. This is almost exactly a third of the roughness of that of commercial steel that I have seen from other sources. I cannot find any other source that can confirm this figure, and so I would like to get opinions on whether or not it could be used in my instance, based on the tubing that I am using.

Thanks and regards,
Hein Kok

Edited by HK1234, 11 April 2012 - 03:40 AM.

[HK1234]

Excuse the title, I mixed up absolute roughness and relative roughness there.

[kkala]

Look at http://www.cheresour...-starting-point '> http://www.cheresour...-starting-point , where a roughness of 0.046 mm is reported for stainless steel. Norsok standard P-001 (available free in Web) reports 0.05 mm for stainless steel.

I understand that Engineering toolbox reports 0.015 mm (less conservative value).

[katmar]

Hein, the roughness of new piping and tubing is more dependent on the manufacturing method than on the material. I suspect that the value of 0.015 mm listed on the page you referenced is intended for commercial stainless steel piping rather than for tubing. In my own work I take the roughness of new commercial piping as 0.05 mm for both carbon and stainless steel. For drawn tubing (all metals) the most frequently listed value is 0.0015 mm, or 1/10th of the value you have.

Manufacturers like Swagelok provide very smooth walled tubing, more for reasons of cleanliness than pressure drop. A roughness on 0.0015 mm looks like a mountain to a molecule trying to hide itself from being flushed out!

The situation you are facing is one where it is much more enlightening to use the old fashioned hand calculation techniques than to use a modern computer based solution. In small bore piping the Reynolds number is likely to be low and on a Moody diagram the curves will be closely bunched together. Calculate your friction factor manually using one of these charts and you will see that the results are virtually the same for 0.0015 mm and 0.015 mm. A 10-fold increase in the roughness is likley to make <10% difference in the pressure drop. If you are using software then it is very easy to do a sensitivity analysis by varying the roughness and seeing what calculated results you get.

If you are trying to use the measured pressure drop to back calculate the flowrate then this might not be accurate enough for you, but if you are just verifying that the piping is adequately sized then this approach should be more than good enough. I know of one major petrochemical company that will not use a roughness of less than 0.03 mm in any situation, and this is probably a good policy to follow in pipe sizing.

[HK1234]

Kkala, thanks for the link to the Norsok standard. Will definitely stick to being conservative with designs.

Katmar, that was a great reply, thanks. I've played around with the roughness value in an existing spreadsheet of mine. The flow difference is indeed very small between the two value you've provided. I am working from pressure drop, but it the accuracy is suitable. Good to get some more understanding and feeling on what others use in this area. And it's always great to meet other South African engineers.