Pump nozzle overloaded due to solar gain of piping.

[MatMan83]

Hi all. I have an issue with a pipe getting too hot in the sun. The line in question is a DN1600 CS pipe which is connected to a piece of rotating equipment. The normal operating temp for the line is 36degC and under normal conditions it will be full of flowing water. However, during major shutdowns the pump will be taken out of service and the line drained. At this point the line will heat up beyond the 36degC.

According to the basis of design, "any body in direct sunlight is subject to a maximum temperature of 84.7 degC." Unfortunately when the stress analysis was performed at 84.7 DegC, it was found that that nozzle was overloaded.

ideally we don't want to start redesigning the piping layout etc so I am looking at alternative solutions. One of the options being considered is to shield the pipe from sunlight to limit the thermal gain to the ambient temp of 52degC, at which temp the pump nozzle is not overloaded. I have looked at ISO 12241 but have gone wrong somewhere and would appreciate a few pointers. My questions are as follows:

1) How do I calculate the temperature of the bare pipe due to solar radiation without any shading. My thought was to establish what the equilibrium temperature is based on radiation absorbed vs radiation emitted. However, when I did this the temperature came out really high (273 degC). This seemes a bit high based on 1000w/m2 and an emisitvity of 1 so I think something went wrong?

2) If we were to clad the pipe with a layer of insulation such as rockwool and clad that in aluminium then how thick do I need to make the insulation to limit the temperature of the pipe to 50degC? Assuming a dusty aluminium surface with e=0.44 then I would assume that the radiation absorbed goes down from 1000W to 440W? Also we now have a layer which will conduct heat to the pipe so I have to take conduction into account?

3) Accepting that some thermal gain will take place and the pipe will heat up regardless of insulation thickness, how thick does the insulation have to be to prevent the pipe exceeding 50degC during the time it is exposed to sunlight (assume 12 hrs?). My thought was to consider the heat input through the insulation and calculate the time to heat up based on q=m.cp.deltaT.

Any assistance appreciated. Thanks.

 

[IRstuff]

What do you mean by "nozzle was overloaded"? Can you replace the nozzle?

Can you paint the pipe white, which is what a lot of outside pipes have.

Oh, and bear in mind that the 1000W/m^2 only lasts about 2 hours on the worst case day.

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[MintJulep]

When doing radiation calculations you need to use Baron Kelvin's degrees, not professor Celsius' degrees.

 

[IRstuff]

I got pretty similar results as the OP, except that I included convective loss, which helped to drop the equilibrium temperature into the 175 C range.

The bigger issue is allowing the entire 1000 W/m^2 to be absorbed in the first place. Even a sunshade would help out quite a bit.

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[georgeverghese]

Of the entire 330btu/hr/ft2 (1000w/m2) max solar radiation received, only 50% is infrared and capable of heating. Moreover, reflectance of steel is approx 0.5 at ambient temperatures ( see Perry page 5-26 in the subsection Polished metals, and since reflectance + absorptance for opaque bodies = 1.0), so only 0.5 is actually absorbed (roughly). So net heating flux = 0.5 x 0.5 x 330 = 82.5Btu/hr/ft2 = 250w/m2

Perry 7th edition page 5-14, table 5-2 gives rough values for the combined convective - radiative htc for bare pipe surfaces. Taking a value of 2btu/hr/ft2/degF, we get 2 x dt = 82.5, giving dt = 41.25degF, which, on assuming max air temp is say 100degF, gives a surface temp of 141degF = 60degC. If max air temp = 50degC = 122degF, then max surface temp = 122+41 = 163degF = 73degC.

This value of 84degC is probably for a perfect blackbody where reflectance = 0. Checking, 2 x dt = 82.5 x 2, giving dt = 82.5degF, so surface temp = 100+82.5 = 182.5degF = 83.6degC.

 

[LittleInch]

Or just add a small spool piece before the nozzle and remove it during the shut down periods? Might be a lot cheaper than shading and insulating long sections of line?

Also remember half the pipe is always in the shade(!)

Expansion of empty pipes in hot weather / sun is a real issue and with less weight on the supports it can become your maximum expansion / force case if you have some long runs

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

 

[chicopee]

That question "What do you mean by nozzle was overloaded?" has still not been answered.

 

[MedicineEng]

Is it an option to install a couple of section valves and keep the pipe full of water during shutdown?

 

[EngrPaper]

The nozzle overload comment seems to be just that when the calculations for thermal stress were performed, the stress on the nozzle exceeded what the rotating equipment is allowed to handle.