INTELLIGENT WORK FORUMS FOR ENGINEERING PROFESSIONALS
Come Join Us!
Are you an Engineering professional? Join EngTips now!
 Talk With Other Members
 Be Notified Of Responses
To Your Posts
 Keyword Search
 OneClick Access To Your
Favorite Forums
 Automated Signatures
On Your Posts
 Best Of All, It's Free!
*EngTips's functionality depends on members receiving email. By joining you are opting in to receive email.
Donate Today!
Do you enjoy these technical forums?
Posting Guidelines
Promoting, selling, recruiting, coursework and thesis posting is forbidden.

heat transfer through pipe

To my helpful collegues, I am trying to determine what size and length of 316 SS pipe i would need in order to cool steam at 180°C @ 692Kpa with a flow rate of 260 kg/h to 135°C, at which point it will be water. My ambient condition are 15°C min and 38°C max. My goal is to use as small and short of pipe as possible due to space constraints.
I will also need to calculate the pressure drop through this length of pipe which i will most likely be coiling to save space.
I have a basic understanding of heat transfer and Q=m*cp*(TinTout)= U*A*LMTD, but i am not sure on how to apply it for my exact situation.
I would appreciate it if someone could walk me through the calculation that i need to do. Thank You! 

Hi ddelia19
Is this related to work as it reads like a student homework question which is probably why no one as responded. If you can tell us the application you might get responses
desertfox 

Sure I can explain, I am a project engineer at Lawrence pumps Inc. located in Lawrence, Massachusettes. I have been working here since summer of '07. This is my first engineering job since i graduated college. I desing and run projects for industiral sized pumps for applications such as the petrochemical industy and oil refineries. I am currently working on a fully lined slurry pump which is going up to the tar sands in Canada. Their utilities there are rather limited and all they have avaliable for a flush source is steam from their steam traps. In order for the pump to not vapor bind, it is required that water be used as the flush. The customer does not want to add any heat exchangers to their scope due to cost and how far they are in their design process. So i thought if we could use a simple coil of pipe to condense the steam to water, it would be a relatively cheap alternative to a heat exchanger. I hope this helps explain my situation. The reason i asked some one to walk me through it is because my heat transfer is a little rusty seeing as how i have been working moslty with hydraullics since i graduated. Thank you, Danette PS. Please feel free to visit www.lawrencepumps.com to view our products and applications. 

Hi ddelia
Thanks for the response people will realise now your not a student and hopefully you will get the response's and help you require.
regards
desertfox 

If you have access to simulation software, such as Hysys, ProII, or ASPEN, you may set up a simulation that will simultaneously solve the heat transfer and hydraulic losses portions of this problem. Other approaches will probably be quite a bit more work but can still be successful. 

Unfortunately I do not have access to any analysis software. I am stuck with doing it by hand. Thanks for the input. 

Danette, Don't be discouraged. The hydraulics & heat transfer are pretty much independent of one another (I think). I'm having some trouble visualizing your total situation. You say your steam comes from a steam trap(s). That confuses me since you get condensate and not steam from a properly functioning trap. If you have saturated steam at 180C, then it will lose the latent heat first. Your temperature differential will be 180  38 = 142C (worst case). Refer to a good heat transfer book to estimate a reasonable U for a steamair condenser. Combined this will get you U and LMTD. Get Q by taking the enthalpy of the steam less the enthalpy of your water. Then solve for A and use that to get the length of pipe or tubing you'll need. For hydraulic calcs, get yourself a copy of "the standard", Crane's Technical Paper 410 and plan for some fun, late night reading. Good luck. 

Bear in mind that human accessible surfaces need to be kept below about 55°C for safety, which means that you need to enclose your exchanger, which means that you'll probably need some serious airflow, which you needed anyway. Bear in mind that if your air exhaust is hotter than 55°C, there will also be need to be safety measures at the exhaust port of your exchanger. There is a free, downloadable heat transfer text: http://web.mit.edu/lienhard/www/ahtt.htmlThere are also a number of online calculators that might help out. Your basic procedure is to determine the amount of heat you have to remove and then determine the amount of heat your exchanger can dump into the air, which determines the length of the coil. Note that airflow across the coil, and possibly adding fins to the coil can drastically affect the coil's efficiency in heat removal. TTFN
FAQ731376: EngTips.com Forum Policies 

1) First find out how much heat you need to "remove". Use a steam table  one that has both superheat and suturated or a Moullier diagram
2) Use your "basic heat transfer" knowledge. But its a little tricky since the outside is the limiting. Forget the inside heat transfer resistance (its steam) and forget the pipelwall  its all about the outside. Here you can use some textbook stuff. Most basic textbook will have direction on estimating h values for various geometries incl. horz. and vertcal pipes.
3) Now use you basic heat transfer knowledge and use the outside heat transfer and dT=Q/time/A*1/h
(hopefully they will use the same symbols  but most tend to at least those influence by German engineering/heat transfer). Remember a dimensioncheck for the units.
This is not a fee consultantcy so i wont go into more details here.
Best regards
Morten 

Thank you all for your input. I believe i have gotten the advice i needed in order to get at least a prliminary idea if this design will even be feasable for my application. Thanks again, Danette 

rmw (Mechanical) 
8 Apr 09 21:37 
You have been given good advice. The irony is that most people are trying to keep the steam in just such a pipe from becoming water via the use of insulation, heat trace, etc.
rmw 

Common problem, it doesn't take much to reduce the steam to saturation, the big issue is condensing the saturated steam.
you know the amount of heat you have to remove from the steam based on enthalpy difference to cool to saturation, and you know the heat released on condensation, believe you have a rough estimate of the heat transfer coefficient or bare pipehand book data, and the flow rate, bingo with the mean steam temperature
hardly a homework problem, good question


pastyl (Mechanical) 
12 Apr 09 14:17 
Hi ddelia19
The reguired lenght can be find from the equation (as you described)
L=((m*Cp)/(U*P))*ln((T2Tambient)/(T1Tambient))
where
L=pipe lenght(m) m=steam mass flow (kg/sec) P=external perimeter of type= pi*Dexternal (m) T2=steam final temperature (K) T1=steam entering temperature (K) Tambient=higher ambient temperature (K) U= heat transfer coefficient (W/m2K) which given appr. from the equation
1/U= 1/hi +1/ho + x/k
where
hi= steam heat thermal conductivity (W/m2K) ho=ambient heat thermal conductivity ho=816(W/m2K)for ambient free air k=pipe thermal conductivity (W/mk) (appr. 32 W/mk) x=pipe thickness (m)
hi is given by equation
hi*di/k=Nu=0.023*Re^0.8*Pr^0.3 (see for example Lienhard...)
But If ambient conditions is free air then
As hi and k/x are very high comparative ho you can accept that
U=ho
Then you have
L=((m*Cp)/(ho*P))*ln((T2Tambient)/(T1Tambient))
for example, for Dex=50mm you have L=50m
As you understand in case you have not much space you must find a forced external air circulation system so as ho=O(3) as hi and x/k is.
Best Regards


zekeman (Mechanical) 
14 Apr 09 11:21 
Condensing 250Kg/hr steam comes to a whopping 500,000 BTU/hr Air cooling? How about water cooling with water towers. 

boo1 (Mechanical) 
14 Apr 09 16:08 
Feed pressure estimation first requires determination as to whether the feed line flow is laminar or turbulent. For this purpose, dimensionless Reynolds number Re should be calculated using:
Re = D*V/v where D (in.) is the pipe diameter, V (in./s) is the fluid velocity, and v (in.2 /s) is the kinematic viscosity.
If the Reynolds number is above 2,300, flow is turbulent and the pressure drop in the feed line is given by
Pl = 0.069*Q^2*L/(Re^0.2*D^5) where Pl is pressure drop (psi), Q is flow in gpm, L is equivalent line length (ft), and D is inside pipe diameter (in.).
If Re is below 2,300, flow pressure drop Pl (psi) is given by Pl = Q*p*v /(132*D^4)


brlewis (Electrical) 
15 Apr 09 3:15 
What you are looking for is a steam condenser. Look up some designs. Typically you need a cooling media like air or better water. The steam should come into a vessel and pass over coils with water flowing through them. Ambient air with no circulation will not be effective in cooling. According to my gatecycle program you would need 236 m3 of surface area with an air flow of 138,027 kg/hr @15 degC 60% humidity. That is if I did the model correctly, sometimes it is hard to tell. Ill try to run a model with water cooling. I would need to have some idea of the water temperature that you would use. 



