Heat Exchange Question 2

[sampsonr]

Hi There,

This is not a student question, it is a heat exchange that I have been designing in which I already know the physical size, material, etc of the heat exchanger but I would like to get an idea of what kind of temperature change I’m going to get by calculation rather than physical testing. I have been looking through my old text books from university but they only show examples of how to calculate efficiencies etc when all 4 temperatures are known. Not to mention my thermodynamics are a bit rusty.

I have simplified down the details as much as possible but I'm not sure exactly what I'm doing so I may be missing a few factors but that's why I posted it here. Please let me know if you need more information to answer this.

Basically I am trying to find out what would be my Q or t2 for the hot and cold side.

Hot Side Details
T1 = 40 degrees C
T2 = ?
Flow Rate = 0.142kg/s
C = 4180J/kgK
k = copper = 401 W/MK
x = 2mm
Area = 0.02M2

Cold Side Details
T1 = 15 degrees C
T2 = ?
Flow Rate = 0.0625kg/s
C = 4180J/kgK
k = copper = 401 W/MK
x = 2mm
Area = 0.02M2

I guess I could always work backwards and put in some hypothetical temperatures but I figured this must be a fairly common exercise.

If anyone could give any help on how to find T2 or direct me to a similarly answered question that would be greatly appreciated.

Cheers,
Roy.

 

[IRstuff]

I still don't get how you can be designing a heat exchanger without the required heat load

TTFN
faq731-376

Need help writing a question or understanding a reply? forum1529

 

[marty007]

sampsonr,

Do solve this problem, you would have to first determine your heat transfer coefficients for both fluid streams. Is this a shell/tube? Plate/frame? etc? You will need to find the proper correlations, which will require much more information about your two fluid streams (density, viscosity, etc..)

Next, you will need to build to break up your heat exchanger into bite-sized pieces and calculate the heat transferred in each step (easier to do in excel or mathcad...).

If your exchanger is co-current, you can run through the model directly and determine your outlet temperatures.

If your exchagner is counter-current you'll have to guess T2 for one of the streams and iterate until that stream matches its inlet T1 at the far end.

This isn't a small problem, and is why much the heat exchanger industry uses commercial software, good luck.
.
Cheers,

 

[sampsonr]

My limiting factors are area, initial temperatures and flow, I want to know how much energy can be transferred in this system, thus I'm not designing to a specific heat load.

Thanks for the reply Marty. Both fluids are water and it's a plate heat exchange. For simplicity we can call it a co-current system. Would it be possible to get an example of how you would break down the calculation? Even if they are just hypothetical values.

Cheers,
Roy.

 

[marty007]

Roy,

Here's how I would take a first crack at this for a co-current model:
[ol 1]
[li]First off, find yourself an appropriate correlation for flow between two plates. I know that commercial suppliers of plate/frame heat exchangers use all sorts of channeling on their plates, in that case I have no idea what correlations they use, good luck...[/li].
[li]Let's say the plates are 1m long, try breaking the model up into say 10 zones at 10cm each.[/li]
[li]For zone 1, you have your inlet conditions and flow conditions. Use your heat transfer correlation to determine the heat flow in the first 10cm. Using specific heat, you now know the temperature change in each stream. These are now your inputs into zone 2[/li]
[li]For zone 2, you now have your new inlet conditions, again use your correlation to determine heat flow to determine outlet conditions from zone2[/li]
[li]Rinse/repeat for the rest of your zones.[/li]
[li]Your outlet temperatures for both streams will be your outlet condition results from zone 10[/li]
[li]Once you have your formulas, it's easy to add more resolution by breaking the exchanger into smaller/smaller bits. Maybe do a sensitivity check to see if 20 or 30 zones makes any difference to the end result[/li]
[/ol]

Some things to remember:
[ul]
[li]Your total flow rate is divided by the number of plates when using your heat transfer correlation(or +/- 1... I'm in a hurry here).[/li]
[li]If pressure drop is a concern, you can add this functionality as well[/li]
[li]For each flow path, heat is transferred across both bounding plates[/li]
[/ul]

Good luck,
Marty

 

[RayA1004]

Roy,
Your statement of calling it co-current makes it easy to calculate the maximum dT's you can achieve, since a true co-current HX can't have the temperatures cross. (But take the following with a grain of salt as a plate-and-frame is probably not co-current)

Since both sides are water, the temperature loss of the hot side is proportional to the temp loss of the cold side by the ratio of the flows.
If you had an extremely long exchanger, the hot side would drop about 7.6 deg and the cold side warm about 17.4 deg. They both end up approaching 32.4 deg. Is this the range of accuracy ("I would like to get an idea of what kind of temperature change") you were looking for?

If these simplifying assumptions seem too basic, you'll have to refer to Mr 007's comments.

Good Luck
Ray

 

[sampsonr]

Hi there,

Thanks for the reply Ray, I have previously tried this and essentially balanced out the 2 streams till they meet however I need something a bit more accurate as its only a short Heat exchanger. The temperature change on one side I'm expecting approximately (T1 - T2) 40 - 35 degrees and on the other side more the (T1 - T2) 15 - 25 degrees.

Awesome Marty, that makes sense. I'll give it ago this weekend and let you know how I get on.

Cheers,
Roy.