Need Help to design a colling coil 2

[junumon]

Dear Friends
I am a engineer working in a heat exchanger maintanence and manufacture company in middle east Right now I have a job in which design is involved. My company is a small company and being a recent graduate I dont have anybody to guide me.
The problem is to design a ss coil to cool drinking water in a 1000 ltrs SS tank. The rate of consumption of drinking water is at the rate of 15 ltr/ minute. The cooling medium is chilled water at 6 degree celcius tapped from a chilled water pipe line. The flow rate is 36 gallon per minute. The drinking water enters at around 45 degree celcius and to be cooled to 15 celcius.
I considered this problem similar to cross flow with one fluid mixed. I found the convective heat transfer coefficients for both sides and later overall heat transfer coefficient. Then by LMTD methode i tried to substitute the values and tried to find the length of pipe required. I got the length of the pipe around 200 mtrs and I feel this is wrong.
Can anybody advice me?
Regards
junu

 

[25362]

Without entering into any calculations nor confirming your estimates it appears that the coil's length you found would result in about 58 peripheral turns for a 1 m tank height. A 1/2 in. pipe could then do the job. In order to be able to check your estimates better provide more info on the coil and tank dimensions, and state clearly where and how does the warmer water flows. Good luck.

 

[junumon]

Dear 25362,
Thank u very much. The tank is of 1 m x 1m x 1m in dimension. I am planning to make coil with 1 inch ss tube and to place in side the tank. The water enters the tank at tempreature around 45 degree celcius and to be cooled down to 15 degree celcius. The rate of consumption of cooled water from tank is 15 ltrs/ minute
Inside the SS coil, water at 6 degree celcius enters at a rate of 135 liter per minute. Infact this is tapped from a chilled water supply line. What I have to find is the length tube required to cool the water in the tank
Hope that this clarifies the situation

Regards,
Junu

 

[25362]

To junumon, from data I once took from published literature the overall U (HTC) value, for a coil-in-tank water/water exchange, non-agitated, clean service: 85-100 kcal/[(h)(m²)(K)]. Considering the geometry of the system in which the ratio of diameters tank-to-coil (taken twice) is about 20:1, I'd take the smaller limit., i.e., 85 kcal/[(h)(m²)(K)].

Process duty: 27,000 kcal/h
LMTD: 19.3 K (uncorrected).
A: (27,000)/[(85)(19.3)] = 16.4 m²
Lt of tube: 16.4/(pi*0.0254) = 206 m. The same order of magnitude you have found! Would you consider installing two concentric coils ?

If you add an agitator and baffles, the U (HTC) can increase to a value 7 to 14 times higher, proportionally reducing your area and coil length. Of course, the stirring energy should be added to the duty to cover for friction losses.

If the tank isn't insulated, some heat gain may also have to be accounted for, and, as a result, the LMTD would drop a bit. Remember 1 kW = 860 kcal/h.

Don't forget for each turn to assume a (center-to-center) pitch of about 1.5*d_tube. And, please, remember the formulas in use to estimate convective HTC's are just that, only estimates within +/-25% of the actual values.

Far from me to intrude in the scope of your design, but can you tell me why, if the chilled water is also of drinking quality, isn't it possible to mix both water streams to reach the desired final temperature ?

Good luck.

 

[kenvlach]

I would say that your and 25362’s calculations indicate a different style of heat exchanger should be used. Use a coaxial coil (tube inside tube), with countercurrent flows. Run the chilled water in the outer tube and submerge the HXer in the tank, so that it cools the bulk water as well. Insulate the tank and use a temperature controller set for a tank temperature a few degrees below 15^oC to control the chilled water flow. For a reasonably sized CC flow HXer, you probably only need 24 GPM of chilled water.

 

[tickle]

Kenvlach

Your idea sound like a sensible suggestion. My question to your proposal is - is it possible to bend the coaxial coil? I do not know. I would have thought that it would not be possible to bend the tubes separately and then insert them inside one another.

Junumon

Another alternative would be to cool only the outlet of the tank, then the tank could remain at 40deg C.

I would also refrain from adding baffles to the tank, as these would provide microbial growth points.

Saying that, is the water to be cooled to 15deg C to stop microbial growth in the stored water?

 

[kenvlach]

Re countercurrent coil HXers. McMaster-Carr sells such coils in both flat spiral & cylindrical shapes. Some have a finned inner copper tube inside an outer steel tube, e.g., P/N 35245K88. A Stainless Steel Tube-In-Tube Heat Exchanger, ½” diameter inner 1” outer, is P/N 3248K13.

http://www.mcmaster.com/

Have seen similar designs in SS on some mfr. website, maybe Trantor or Paul Mueller.

http://www.muel.com/

http://www.tranterphe.com/phe.htm

Alfa-Laval sells a similar spiral design, more like 2 co-joined tubes.
"the Spiral Heat Exchanger is made from two metal strips that are wound around a centre to form 2 spiral channels. Removable covers are fitted to the top and bottom of the body in order to retain pressure , create a channel seal and offer easy access to the heat transfer surface."
"Benefits of installing a Spiral Heat Exchanger:
Low operating cost as a result of self cleaning effect
Low installation cost due to compact design
High heat transfer efficiency, typically 3 times that of a shell and tube heat exchanger"

http://www.alfalaval.com/scripts/We...ode?siteNodeID=1471&contentID=-1&languageID=1

If you cannot find what you want above, I am sure one of my local fabricators below could put a 1/2" tube inside a 1" tube, coil it and weld end fittings. Use fully annealed 304L or 316L. Cannot be sure about keeping the smaller tube centered w/o tacking on some longitudinal fins. Anyone know of extruded, finned SS tubing?

http://www.titanmf.com/

http://www.astrocosmos.com/

 

[junumon]

Dear 25362
Thank you very much for your detailed explnation. U asked y dont he use the chilled water itself to mix with the water in the tank. Infact I also raised the same doubt. I was told that the chilled water is not of drinking quality.
Thank you very much for your explntion

Take Care
Junu

 

[junumon]

Dear kenvlach & Tickle
Thank u very much. Ur suggestion is interesting. But I have the practical limitation of manufacturing the double pipe coils. Any way I will make a try. I have not tried it so far.
Take care
Bye
junu

 

[chicopee]

What I have not seen considered in the above suggestion is consideration for transient state. Is the drinking water used intermittently during a one hr period? If so then transient consideration could apply and tube length could be reduced.