[TRNSYS-users] Strange results from Type52
Avatar C. K. Lee
a8304506 at graduate.hku.hk
Wed Jul 17 04:59:51 PDT 2013
Thanks for your reply. I have made a slight modification of the model
code in order to trace the problem. Then I find that when the control
signal drops below 0.4, the water flow inside the tube lies in the
transition region. The resulting heat transfer coefficient on the tube
inside surface decreases by more than 80% from the "just turbulent"
situation to the "just laminar" condition. As a linear interpolation is
used in the model to calculate the heat transfer coefficient within the
transition region, the reduction in the coil capacity becomes much
larger than the reduction in the water flow. Hence, the outlet water
temperature starts to decrease when the control signal drops below 0.4.
The simplest way is to rectify is to choose a smaller tube size. It
helps increase the Reynolds number inside the tubes. Another problem is
the adoption of a constant Nusselt number when the flow becomes laminar
inside the tubes in the model code. I have made some modifications of
the original model code. Here, the calculation of the Nusselt number in
the laminar region is based on the formulations from "Nellis G, Klein S.
Heat Transfer. New York: Cambridge University Press, 2009". For the
Nusselt number in the turbulent region, the more accurate equations from
"Gnielinski V. New equations for heat and mass transfer in turbulent
pipe and channel flow. International Chemical Engineering
1976;16:359-68" is followed. These formulations are used in EES in
calculating the heat transfer coefficient for internal pipe flow.
Attached is my revised source code for your reference. The original
equations are converted to comment statements so that you can easily
revert to the original model. Try and see if it is better.
Chun Kwong LEE
City University of Hong Kong
David BRADLEY wrote:
> Chun Kwong,
> I suspect that when the water flowrate drops below a certain point,
> the heat transfer rate between the air and the water begins to
> decrease. I modified your simulation so that there is an equation that
> sets the pump control signal starting at a value of 1 and ending at 0
> after 12 hours. That way, you can see the progression of the outlet
> water temperature (I also used a shorter timestep). As you say, the
> water outlet temperature starts out increasing but when the flow rate
> through your coils gets to about 4000 kg/h (1.1 L/s) the outlet
> temperature starts to go back down. At that point, the face velocity
> of water through a single tube in the coil is only about 0.2 m/s,
> which seems a bit slow. I don't think that you have reached a
> laminar/turbulent flow point because there is not a sharp break in the
> water outlet temperatures (which you usually see because the Reynolds
> number drops very suddenly).
> In any case, I am not sure that the model is wrong and I certainly
> don't see anything incorrect in your implementation.
> On 7/10/2013 02:30, a8304506 at graduate.hku.hk wrote:
>>I try to investigate the performance of a cooling coil at different chilled
>>water flow rates (see attached .tpf file). When I adjust the control signal of
>>the water pump from 1 down to around 0.4, the chilled water leaving
>>temperatures increase as expected. However, when I further reduce the control
>>signal, it appears that the leaving chilled water temperatures begin to
>>decrease. Is it the problem from the cooling coil model or any other issue?
>>Please advise. I am using TRNSYS16.1. Thank you!
>>Chun Kwong LEE
>>City University of Hong Kong
>>TRNSYS-users mailing list
>>TRNSYS-users at cae.wisc.edu
>Thermal Energy Systems Specialists, LLC
>22 North Carroll Street - suite 370
>Madison, WI 53703 USA
>d.bradley at tess-inc.com
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
More information about the TRNSYS-users