Evaporator Size 2

[gradkid]

How do I size an evaporator that is used to transfer 26000Btu/h from a 70%/30% MeOH/water solution FR 1.4gpm to R-22?

 

[willard3]

Get coil sizing software from Trane, Carrier and etc.

Your requirements, other than tonnage, are opaque and I don't understand them.

At minimum you are going to need:
load (tonage)
coil face velocity
Entering air dry bulb and wet bulb
Leaving air dry bulb and wet bulb
coil material
system pressure parameters
properties of fluid to be cooled
coil tube velocities
and etc.........

 

[Yorkman]

This MeOH/water solution sounds like a liquid not a gas am I correct?
If it's a liquid, and 1.4 GPM is the flow rate then you will you will need:
the specific heat of the solution
Entering liquid temperature
Reativity of the solution to the heat exchanger materials

You have already pre-determined what your [Δ]T for the liquid is going to be based the 26,000 BTU/Hr and the flow rate. The supplier will probably need to know the operating pressures your working with and the available pump FT/hd capacity you have to work with.

I wonder how you came by the BTH/HR and the flow rate?, only because you didn't provide your entering liquid temperature and your required leaving liquid temperature. It seems you may have gotten the cart before the horse as they say.

I'm not a real engineer, but I play one on T.V.
A.J. Gest, York Int./JCI

 

[Yorkman]

Ooops
One thing I left out they will need to know the density or weight per gal of the solution, sorry about that.

I'm not a real engineer, but I play one on T.V.
A.J. Gest, York Int./JCI

 

[gradkid]

Working with lots of data. The MeOH/H2O is a liquid soltuion. Entering temp is 53F exit temp is 50F. Flow rate was found using Q=m c delta T (cp must be found for mixture at 53F) I believe a plate exchanger would be best transfer solution. To determine required plate area, use Q=U A delta T for fluid. U must be overall heat transfer coef.= 1/(1/h fluid+k exchanger/t exchanger + 1/h R22)
Is this correct?

 

[Yorkman]

I may be in error and admit I am a bit out of my league on this matter, but I think that equation Q=U A [Δ]T is for heat transfer between two fluids. I think with a refrigerant and because of the U in the equation a different method is required. With a refrigerant you are getting into two phase flow due to the change of state that happens in the heat transfer process. I'm sorry but I must sit back and see if someone with more experience can help you through this one, as my knowledge here is limited. This will be a discovery process for me as well.

I'm not a real engineer, but I play one on T.V.
A.J. Gest, York Int./JCI

 

[gradkid]

The phase change has been giving me a headache as well.

 

[Yorkman]

I can tell you how many Lbs/min you will need to circulate, But when you talk about sizing the plate area, I would recommend talking to a manufacturer/supplier they have software that can select the proper heat exchanger for your load and flow requirements, and refrigerant selection. Don't forget R-134a may also be an option.

I'm not a real engineer, but I play one on T.V.
A.J. Gest, York Int./JCI

 

[insult2injury]

for heat exchanger sizing there are two predominant methods
1. LMTD
2. Effectiveness-NTU

LMTD is closest to what you've done, so let's go with that.

Q=U A F deltaTlm

U is the overall heat transfer coefficient
A is the area
F is a correction factor for flow arrangment (usually 0.6 to 0.85) which is read from a graph
deltaTlm is the log-mean temperature difference which is defined different ways for different flow arrangments...essentially it calculated the average difference in temperature between the two working fluids

now to complicate things more, because phase change is occuring, the heat exchanger has to be discretized into three segments (liq-liq, liq-2phase, liq-vapor)
each of these segments will have vastly different U-values which should be obtained through a nussult correlation. In the case of the 2-phase U-value, this evaluation is very complicated and hard to get accurate without some experimental data.

That being said, and since you don't appear to be familliar with HEXs, you should consult a HEX manufacturer which will be able to provide performance characteristics of their particular models. Care should still be taken as most OEMs overstate the performance by as much as 10 percent.

Still another worry is that the performance will degrade over time as fouling occurs. The heat transfer coefficients will decrease requiring the R-22 flow rate to be increased. Be sure to get the performance results in both new and fouled conditions.

 

[Yorkman]

Insult2injury,
Thanks for the post although I'am not familiar with the terms, the concept makes sense and gives me a bone to chew on as it were.
Now where did I put that book on nussult correlations

I'm not a real engineer, but I play one on T.V.
A.J. Gest, York Int./JCI

 

[gradkid]

Thanx. Back to the calculations.