Absolute pipe roughness and surface finish of pipe 1

[paramarth]

For pressure drop calculations, Crane's Handbook and many software give the absolute roughness of Pipes as 0.05 mm for clean commercial steel pipes

What is the surface finish (RMS or Ra) to be specified to the pipe manufacturer to attain the above absolute roughness.

Is there any correlation between RMS value and absolute roughness of pipes used in friction drop calculations

 

[rconner]

I see you have not yet gotten any response to your inquiry - what is your service that you feel you need to "specify" internal surface finish "to the pipe manufacturer"(I ask in that ordinary "commercial" steel pipes are typically not internally machined)? [Otherwise, there is a good past discussion of similar subject at

http://www.eng-tips.com/viewthread.cfm?qid=322259

].

 

[paramarth]

My service is super heated steam at 540 deg C

 

[LittleInch]

There is no recognized surface finish normally specified.

you could try "polished", but if you want something better than the standard 20 - 45 micron (.001mm) level which has been recognized as the default standard range for new steel pipes, then you need to coat it or line it with something.

epoxy coatings are normally good for about 5 - 8 micron and PE is usually thought of as 3-5 micron.

I'm struggling to think that a steam line is long enough for the absolute roughness to be a significant issue.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[paramarth]

Vendors are giving surface finish of Internal dia ranging from 6.4 Micron (Ra) - 250 RMS to 15 Micron (Ra) - 500 RMS
My doubt is whether the surface finish given by Pipe vendors can be treated equivalent to a hydraulic roughness of 0.05 mm as specified in CRANE Handbook, or should I take a higher roughness value for Pressure loss calculations. Or should I insist on the Pipe vendor for an acceptable surface finish which can be treated equivalent to a hydraulic roughness of 0.05 mm
LITTLEInch has thrown open another question to me for which I dont have an answer whether surface roughness plays a role in a Superheated Steam line of 170 ata, 540 deg C

 

[katmar]

There are two main reasons why someone would be interested in the roughness of the internal surface of a pipe. For us doing hydraulic calculations the most important reason is to enable us to determine the friction factor. The roughness we use for these calculations is an artificial number based on an equivalent sand grain size (see the link given by rconner). It is almost never specified by the pipe supplier. But for new commercial grade steel pipe a value of 0.05 mm will give you good results. It is an instructive exercise to run some alternative calculations with various roughness values. Double it, halve it, mutiply it by 5, assume it is zero, and see what impact it has on your pressure drop or sizing calculation.

The second main reason for being interested in the roughness is for hygiene and cleanliness considerations. This can be important in the pharmaceutical and food industries, but usually does not bother us with steam flow.

While it is true that roughness is more important in gas flow than in liquid flow (because of the high Reynolds numbers preventing hydraulically smooth flow) I have never found a situation where it impacted on a steam line. If you are pumping natural gas thousands of kilometers then a smoother pipe can save you significant amounts of energy, but if your steam line runs at a slightly lower pressure drop it probably only means that your control valve will settle out a bit less open.

Katmar Software - AioFlo Pipe Hydraulics

http://katmarsoftware.com

"An undefined problem has an infinite number of solutions"

 

[pocketengineer]

Some designers prefer to use a higher roughness value to allow provision for pipe corrosion after years of service. To me, engineering calculations are estimates and are done to the best of my knowledge and assumptions with published data/common industry norm.

 

[zdas04]

API 14.7 specifies a minimum roughness in meter tubes (square edged orifice gas measurement). When I was working as a Measurement Engineer we generally required the fabricator to rough up the surface finish on new steel pipe to meet that minimum. Modern steel pipes are at least 3 orders of magnitude smoother than we saw prior to the 1980's. This is the primary reason that the AGA Fully Turbulent Gas Flow equation rarely gives satisfactory results (the pipe is so smooth that the line is always outside of the fully turbulent region on the Moody Diagram). The values in Crane don't match any pipe I've ever measured the surface roughness on.

David Simpson, PE
MuleShoe Engineering

In questions of science, the authority of a thousand is not worth the humble reasoning of a single individual. Galileo Galilei, Italian Physicist

 

[paramarth]

Can somebody please tell me what is the normal surface finish expected of a pipe as per good industrial practice. Some vendors extrude pipes and give without machining ID while others machine the ID and give good finish based on the manufacturing process.I am talking about high wall thickness pipes used in Main Steam Piping of power plants.If I have to specify a surface finish value for ID of pipe what should be the value

 

[katmar]

paramarth, you need to get away from the mind set of "somebody give me a number". It looks like you are simply trying to avoid taking responsibility for your own design. You have received advice that is aimed at increasing your understanding to enable you to make your own decision. I have suggested that you investigate how the roughness affects your calculations and your decisions. Do that, come up with some tentative conclusions, and then come back here to explain your conclusions and ask for comments on what you have done.

Katmar Software - AioFlo Pipe Hydraulics

http://katmarsoftware.com

"An undefined problem has an infinite number of solutions"

 

[paramarth]

Thanks katmar for your frank comments. I have a situation in hand where I have worked out dropfor various roughness factors. My vendor has now come up with an option of a better surface finish at a higher cost. That is now prompting me to raise this question. I am at the border of my pressure drop limit with 0.05 roughness factor.Anything higher will warrant a bigger size

 

[katmar]

There are always uncertainties in calculating pressure drops and if you have a typical line with bends, tees, valves etc you should expect at least a 15% variation between calculations and on-plant measurements. There is also the factor pointed out by pocketengineer that pipe roughness can increase with time due to corrosion. If I were on the border line when using a roughness of 0.05 mm I would choose the larger pipe. Even if you buy the more expensive smoother pipe there is no guarantee (I suspect) that it will stay that way. A decrease in roughness to 0.02 mm will decrease the pressure drop in the straight pipe by about 15%, and it will not decrease the pressure drops through the fittings at all.

Katmar Software - AioFlo Pipe Hydraulics

http://katmarsoftware.com

"An undefined problem has an infinite number of solutions"

 

[georgeverghese]

Surface roughness values used in hydraulic calcs referes to average in service conditions a length of time after the pipe has been placed into operation. And this surface condition could be miles away from the as new roughness condition, depending on several factors. In this steam service with high temp carbon steel pipes, it is very unlikely that the pipe would have retained as new roughness values a few years later.

Perhaps you can explain why an increase in surface roughness should lead to an increase in required pipe size.

 

[pocketengineer]

at the border of pressure drop limit! Is it the dilemma that one pipe size bigger cannot be justified (don't seems is a cost issue here)? What is your design safety factor allowance? Bear in mind that the design/calculation assumptions can never be the same as as-installed condition especially turns and fittings to suit site conditions. To rectify an underestimated design may cost you a bomb.

 

[paramarth]

I am using a 9cr1Mo alloy steel pipe and that is the dilemna of going for the next higher diameter which will have cost implications. We have never bothered about surface roughness till one of our pipe manufacturers came up with the option of a better surface finish, that poses the question what is acceptable as a good industry standard and what is not.

 

[georgeverghese]

As suggested, the roughness you should use should be that representing the pipe after a few years in service. 9Cr 1Mo is not a corrosion resistant alloy in steam service, so roughness after 2-3years say, will be much more than in as new conditions.
In some cases, where there is a control valve downstream or some device which is somewhat tolerant of variations in its inlet pressure, perhaps you may not have to change line size with the higher roughness. In other cases, maybe you should move to a bigger line.