PIPE ROUGHNESS API 5L 2

[GabrieleB]

I have to make some calculation on pressure drops along a pipeline.
Reference standard is API 5L. What roughness I have to use? Is it correct to use 45micron. I' haven't seen any maximum roughness specification in the API standard. Can anyone help me on finding a reasonable roughness value to put on pressure drops calculation? Thanks

 

[zdas04]

For new, commercial steel pipe virtually every reference you find will have a number equivalent to 150E-6 ft (1.8E-3 in, 45.7E-6 m, 45.7E-3 mm, I'm not going to look up "micron" since it seems to mean different things to different fields). I strongly recommend people pick one unit and always use it for ε values in the ε/ID calcs. People screw this one up more often than any other calculation that I review. In classes I teach, every surface roughness number I present is xxE-6 ft. If I come across an equation that uses absolute surface roughness instead of surface roughness relative to pipe ID then I check carefully to make sure I'm using the units that the equation expects.

When I was doing a measurement engineering job in the mid 1990's I had occasion to verify the surface roughness on dozens of lengths of API 5L pipe. The minimum surface roughness allowed in ASME 14.3 for gas measurement is 150μinch (12.6E-6 ft) and about 1/3 of the time we had to use a hone to rough up the surface to meet that. We were never anywhere close to half 150E-6 ft (1800 μinches).

This is too small a sample to give me standing to fly in the face of hundreds of well respected researchers (and tens of thousands of idiots who use numbers they don't understand), but since most of the time a three-order-of-magnitude change in surface roughness will result in a 2-3% change in calculated pressure drop, I've never gotten too excited about using a number that is clearly out of date with manufacturing practices. The only reason that I stress it to my classes is that if you use the number that people expect, then you don't have to explain "why?" so often, and if it doesn't make a material change to your final decision it makes more sense to give people what they expect to see.

David Simpson, PE
MuleShoe Engineering

"Belief" is the acceptance of an hypotheses in the absence of data.
"Prejudice" is having an opinion not supported by the preponderance of the data.
"Knowledge" is only found through the accumulation and analysis of data.

 

[BigInch]

You should use a roughness that will give you acceptable answers to the question (what is the pressure loss, or what is the flowrate) you need to answer.

Roughness is not defined in API 5L, although as zdas said, new commercial pipe is probably going to be 0.0018 in x 25.4 = 0.04572 mm. There is no guarantee it will be better, or worse.

If you run corrosive fluids, natural gas with water and CO2, H2S, water with oxygen content, etc., it will soon be worse. And perhaps you are planning on running this pipeline for 10, 20, 50 years or more, you might want to use a roughness that you predict the pipe will have 50? years or more from now.

Roughness is up to you and your material specialist's opinion. Roughness value is always dependent on pipe condition. That depends on the pipe, what type of pipe you buy and what you put in it, how and how long you intend to operate and maintain it. What does your design specification say. Use new pipe roughness, or we want to be able to flow at least X m3/h for the life of the pipeline. Predict and use a roughness that is appropriate for the ultimate design conditions.

What would you be doing, if you knew that you could not fail?

 

[zdas04]

I've checked some gas pipes in CBM (90% CH4/10% CO2) service that had been in the ground in wet gas service (and never pigged) for 10 years and the surface roughness was around 180 μinches. Ten years after that I checked it again (pigged twice a week for 8 years) and the roughness was 100 μinches. I haven't seen any evidence anywhere in the world that gas flow hurts steel pipe.

Now lines that are not pigged generally have places where water stands and those places can get noticeably rougher, but I've never checked the surface roughness quantitatively in those places.

David Simpson, PE
MuleShoe Engineering

"Belief" is the acceptance of an hypotheses in the absence of data.
"Prejudice" is having an opinion not supported by the preponderance of the data.
"Knowledge" is only found through the accumulation and analysis of data.

 

[katmar]

The short answer is that 150E-06 ft (or 0.045 mm in katmar-speak) will give you good answers for normal process piping design.

The long answer, which you didn't ask for but which zdas04 has introduced, is that there are several different types of roughness and it is good to have some appreciation of the differences - even if you (like me) don't really understand it. We just need to know enough to be able to do reliable calculations.

The roughness that we use for the e/D calcs is based on experiments done back in the 1930's by the German J. Nikuradse. We all owe this man a huge debt of gratitude. He took carefully sieved samples of sand with very narrow size distributions and glued them to the inside of pipes to determine the effect of roughness on pressure drop. The roughnesses he reported were the average grain sizes of his sand. Now, you can visualise that these particles were much larger than the apparent roughness you would see under a microscope - you would see only half the diameter as a projection. And this is confirmed by zdas04's experimental investigation that found the real projections inside the API 5L pipe were much smaller than the "roughnesses" that gave good answers in his pressure drop calculations.

Also, the size range of surface projections inside a pipe is much more widely distributed than Nikuradse's carefully sieved sand samples. We use what Idelchik calls the "hydraulically equivalent uniform-grain roughness" (i.e. Nikuradse's sand particles) to represent the far more complex real-world situation. For most practical purposes this is plenty good enough, but there are ways of addressing the distribution of the roughness in mission-critical calcs (but you didn't ask about that so I will stop now!).

Katmar Software - Engineering & Risk Analysis Software

http://katmarsoftware.com

"An undefined problem has an infinite number of solutions"