Heat Transfer in Tank Using Hot Water Pipes

[kl4schmitz]

Hi - I am tying to calculate how long it will take to heat a water tank using hot water pipes. The tank is well insulated so I will not calculate any heat loss. I need to heat the effluent of a digester to 160° and maintain that temperature to qualify for Class A bio-solids. Below are the parameters of my design.

Tank start temp. = 100°F
Tank end temp. = 160°F
Tank Dimensions = 19'W x 16'L x 18'D
Tank Volume = 40,930 gallons
Heating System = 800' of 3" sch 40 stainless pipe (3.500" OD - 3.068" ID)
Heating System Water Temp. = 180°F
Heat System Flow Rate = 200 gpm

I will have a mechanical mixer installed to help with heat transfer. I believe the only thing I am missing is the rate of heat transfer for the pipe.

Any help would be appreciated.

Thanks,
kl4schmitz

 

[srfish]

Hi,

There is a course taught by John F. Pietranski. It has the title "Heating and Cooling of Agitated Liquid Batches: Isothermal Medium. You can download the course. The file name is Isothermal Medium.pdf.

You may know that more discussion and equations can be found in "process Heat Transfer" by Kern.

 

[vpl]

There is a free heat transfer book available here:

http://web.mit.edu/lienhard/

http://www/ahtt.html

The course srfish refered to is available here:

http://www.pdhonline.org/courses/k101/k101.htm

A few questions:

Is the 800' feet of piping all inside your tank? If not, how is the tank being heated? (ie, Is the pipe running straight through the tank? Is there a labyrinth pass through the tank? Is there a separate heat exchanger?) A diagram might help.



Patricia Lougheed

******

Please see FAQ731-376: Eng-Tips.com Forum Policies for tips on how to make the best use of the Eng-Tips Forums.

 

[QMax1]

To get the heat required, you use Q = m Cp dT. Assuming your process is water (Cp = 1), you will need over 20MBTU input into the process.

To get the heat given, you use Q = U A dT. U between two water boundries and a layer of pipe is about 6 BTU/hr ft^2 F (limited by your tank water). Given your 800 feet of pipe (A=~1 ft^2/ft) and your dT (180-140(approx log mean), you will only transfer about 200KBTU/hr.

I don't think this is going to be good news for you, but it looks like 1,000 hours to heat the tank contents. Further, I would not underestimate the heat loss through the "well insulated" wall of the tank. This can add 10-15% to that number.

 

[QMax1]

Sorry, that U should be 60 BTU/hr ft^2 F so the time should be more like 100 hours.

 

[25362]

kl4schmitz,

In addition to remarks by others, if it were clean hot water (or condensate) flowing at 8.7 fps vs a reasonably turbulent dilute water solution, and assuming well-distributed (~15?) turns of a pipe coil, U might even be twice as large as suggested by OMax1.

 

[IRstuff]

http://www.engineeringtoolbox.com/overall-heat-transfer-coefficients-d_284.html

gives values similar to QMax1's

Nonetheless, QMax1's calcs still seem to be off by about a factor of 10. Using his revised htc, the heat transfer rate should be ~2MBTU/hr, and with 20MBTU need, that's about 10 hrs.

Using similar equations from srfish's citation, I get about 10.8 hrs. But that assumes perfect mixing, presumably.

TTFN

FAQ731-376
Chinese prisoner wins Nobel Peace Prize

 

[SNORGY]

I think all the theory you need to solve this is in Perry's Chemical Engineer's Handbook under Batch Processes. In the 6th Edition (which it was when I bought mine) it is in Chapter 10 pages 10-37 thru 10-41.

Having gone through something similar in the past year, I agree with IRstuff. Without actually crunching the numbers, 10 hours sounds about right, whereas 100 hours doesn't seem to sound right to me.

Regards,

SNORGY.

 

[SNORGY]

Further, if this tank is rectangular, the following link may be useful to you...

Admittedly they are more concerned with steam, not water...but the methodology to compute the tank heat loss is still valid.

Regards,

SNORGY.