13 replies to this topic

[Marco78]

Dear all,

I'm trying to perform a simulation using HTRI for a condenser. The condenser is an existing equipment (therefore its geometry is known) and it is a TEMA S&T exchanger (AEM).
The inputs, among others, are as follows:
1. Shell, condensation
- Fluid composition (pure component)
- Flowrate
- Tin

2. Tubes, cold fluid
- Fluid (pure component)
- Flowrate
- Tin

performing the simulation I get both Tout but for the condensing fluid Tout is very close to Tin --> subcooling seems not to have happened. I don't have the data here but the surface is definitly overdesigned.
Is there any option that I left out? Does anybody of you use this simulator?
Thanks in advance,

Marco

[Zauberberg]

What do you mean by "you don't have the data" and "the surface is oversized"? Based upon which facts you have made such conclusion?

What's the performance of actual condenser in the field, when compared against simulation results? Why do you expect significant subcooling? Does it happen in reality?

You need to provide much more data in order to get some meaningful answers.

[Marco78]

Dear Zauberberg,

your are right, my explanation wasn't so clear.
I don't have the simulation results in this pc but I remember well that the simulator returned to me something like "1000% overdesign"; I also tried to put different Touts (forcing subcooling) for the condensing fluid and I've always obtained a positive overdesign.
Unfortunately the equipment is not still installed (even if it is real) and I can't check its behaviour.
Thanks

EDIT: I've checked in the HTRI online help and from version 5.0 which I'm using "an option has been added that allows estimation of the exit temperature of the condensate(Condensate subcooling)"
Does anyone know how to perform it during a simulation?

Edited by Marco78, 29 September 2010 - 02:34 AM.

[Propacket]

Without knowing the data, it is difficult to tell you what is the exact problem?

We can just speculate - which is not going to help you.

[Marco78]

Here the data:
1. Shell, condensation
- Toluene
- Flowrate=2500 kg/h
- Tin=110°C (normal boiling point since I've put Pin=0 barg)

2. Tubes, cold fluid
- Water
- 3 attempts: 20/30/40 m3/h
- Tin= 20 °C

The condenser is a TEMA AEM:
Horizontal, ID 496 mm, Baffle geometry: single parallel segmental, Tubepasses=4, n.tubes 146 equal to 32 m2. Cocurrent flow in the 1st tubepass and countercurrent in the 4th tubepass

Thanks

[Zauberberg]

You certainly haven't provided to us all the input data which you must feed to the HTRI software in order to be possible to run a simulation mode case. Please re-post the full set of data, an Excel sheet would be very much appreciated.

[Propacket]

It would be helpful if you can upload HTRI Rating Data Sheet or HTRI TEMA Spec Sheet.

[Marco78]

Please find the attached images captured from HTRI.
Many thanks

Attached Files

•  htri.zip   358.75KB   422 downloads

[Propacket]

Please find the attached images captured from HTRI.
Many thanks

I will check it and revert to you if there is anything comment able .

[Propacket]

Macro,

You haven’t gone wrong anywhere.
Your HTRI simulation seems good. I have checked the results with Hysys and they are fairly in agreement with HTRI output. I have attached the Hysys findings herewith.

You can also check your results manually as follows:
Q=U x A x LMTD
Where Q=0.2577 MW=257700 W
& LMTD=86.2°C=86.2K
Therefore, calculated UA =Q/ LMTD=2989.55 W/K
Heat Transfer Area=n x surface area of one tube=146 x (3.1415 x 0.019050 x 3.657) =31.952 m2
Now, HTRI has determined actual U=1190.94 W/m2.K
As a result, HTRI UA=1190.94 x 31.952=38052.91 W/K
You can see that HTRI UA is almost 1172.86% higher than the calculated UA. Overdesign calculated by HTRI (1137.9 %) is pretty good.
Therefore, you have done a good job by accurately simulating the existing condenser with HTRI.

Now problem is why so much overdesign has been provided to the condenser. Just look at the pressure drop in shell side. Its only 1.164 kPa (0.1689 psi). This is what has governed the size of the condenser. Condensers are always designed for as low pressure drop as possible. We normally employ a cross flow shell (AXM, BXM) for condensing services requiring low pressure drop. E Type shell is not suited to low pressure drop services. What will happen if you use AEM type exchanger? You will have to increase the diameter of shell to reduce the pressure drop below the allowable value (normally 0.2 psi) and increasing diameter of the exchanger increases the surface area by leaps and bounds. You can try cross flow shell in HTRI and see the difference.

Thanks

Attached Files

•  Toluene Condenser.pdf   28.01KB   205 downloads

Edited by P.Engr, 30 September 2010 - 07:50 AM.

[Marco78]

Thank you for your check! Your are very helpfull!

I'm also glad that the simulation has been set up well. One aspect remains not very clear to me: for a such overdesigned condenser (and actually the flowrate used in the simulation is a maximum value, normally we expect a lower flowrate...close to 1000 kg/h) is it not reasonable to suppose a subcooling of the condensate? As I've already reported, forcing the output temperature of the condensate to i.e. 30, 40, 50 °C, the overdesign remains quite high (percentage).

I keep not understanding if my thought is wrong and if a simulator is able to provide such a value

[Zauberberg]

One of the errors is that you have specified Toluene vapor fraction = 0 at the condenser outlet, forcing HTRI to avoid any subcooling during iterations. Normally if you want to run software in the Simulation mode, you should specify inlet conditions and exchanger geometry only, and then see what comes out as the result.

See attached snapshots.

P.S. I had to edit the post since I had originally attached the simulation case identical to yours, i.e. when outlet vapor fraction was specified. Completely different results are obtained if you leave this as blank, which is normally done in the simulation mode.

Attached Files

•  Design Rating Simulation.zip   71.08KB   185 downloads
•  Condenser.jpg   107.83KB   98 downloads

Edited by Zauberberg, 30 September 2010 - 03:34 PM.

[Zauberberg]

In simple words, when you force HTRI to evaluate this exchanger as a non-subcooling service, the software (in your results) says that there is so much excess heat transfer area in the existing exchanger - if you want to achieve condensation only, without any subcooling.

Greetings to Marco & Danijela

[Marco78]

In simple words, when you force HTRI to evaluate this exchanger as a non-subcooling service, the software (in your results) says that there is so much excess heat transfer area in the existing exchanger - if you want to achieve condensation only, without any subcooling.

Greetings to Marco & Danijela

Many thanks for your response, that's the point I didn't know/understand. Now it is clear!
Thank you for your gretings, we return them to you!