[Solved] Type 999 – Disposition des couches, profondeur des tubes, et conditions aux limites

@mohammedzaoui

If you have not already done so, I would recommend looking at the "hints and tips" section of the Type999 documentation (located in .\Trnsys18\Tess Models\Documentation\). There are some diagrams and explanations in the documentation that may help. 

Q1: (if there is more than one layer, are they assumed to be in series or parallel?) It is up to the user to define the order in which the flow proceeds through the pipes. The final set of parameters numbers each one of the pipes that have been defined in the heat exchanger and asks which direction the flow goes through that pipe and whether the pipe is downstream of another pipe (and if so, which one) or whether it is an inlet.

Q2: (does the "layer depth" refer to the middle of the pipe or to the bottom of the layer that contains the pipe?) It refers to the middle of the pipe. Figure 6.5.8 might help clarify.

Q3: (is there a way to define insulation beneath the ground heat exchanger?) Not with this model. Type999 and 997 allow you to define insulation above the ground heat exchanger but not on its sides or beneath. There are other models (such as Type1267) that can be used if you have a ground heat exchanger that is enclosed in a ground-coupled insulated box. Type1267 is an individual component and is not associated with one of the prepackaged libraries.

Q4: (there does not seem to be any difference in results when I choose "adiabatic" or "conductive" for the far-field boundary condition). That is pretty normal for ground heat exchangers, especially if they are not buried too deep and/or if the far-field distances that you have chosen are a significant distance from the extents of the heat exchanger. 

kind regards,

 david

@davidbradley

Bonjour, Merci à David pour la réponse rapide à ma précédente question !

J’ai encore quelques doutes et je souhaite juste confirmer certains points pour être sûr.

Q1 : Juste pour être sûr – si je prends la deuxième configuration d’échangeur de chaleur, et que je souhaite que la deuxième couche soit en parallèle avec la première :

La première couche se termine par Inlet 5, Direction 2 Donc, je commence la deuxième couche avec Inlet 6, Direction 2 pour être en parallèle

Et si je voulais que les couches soient en série, je commencerais la deuxième couche avec Inlet 6, Direction 1 Est-ce bien cela ?

Q3 : par rapport à ce qui a été dit : je pense qu'il est en fait possible de modéliser une isolation latérale (périmétrique) dans le Type 999 via la résistance thermique périmétrique .

bonne journée.

cordialement,

M.ZAOUI 

@mohammedzaoui

Mohammed,

I am not sure that I entirely understand the pipe layout that you are trying to achieve. The figure you drew (conf2.png) for layer 1 is entirely correct and you are correct about how to do the series configuration. i.e. if you want layer 2 (below layer 1) to be in series with and to match layer 1 but in reverse then you'll define pipe 7 as having pipe 6 as its inlet and as having direction 1 (just like you describe).

I am a little less certain about the parallel configuration because I am not sure whether you want a second layer of piping exactly the same as the first layer but deeper (with its inlet in the upper left hand corner of your drawing just like in layer 1) or whether layer 2's inlet is not directly below layer 1's.

If you have two layers in parallel whose layout and flow directions are identical and the two inlets are one directly below the other then you'd have

index  inlet  direction

1        0      1

2        1      2

3        2      1

4        3      2

5        4      1

6        5      2

7        0      1

8        7      2

9        8      1

10      9      2

11      10    1

12      11    2

if you have a situation where layer 2's inlet is not directly below layer 1's inlet and/or follows some other flow path then you can still define it. If that is the case and you can send me a picture of how you want layer 2 to be configured then I can suggest the parameters that you'll need to define.

I think you are correct about the perimeter insulation; my apologies for my earlier incorrect answer!

David

Dear Mr. David,

Thank you again for your continued support and valuable assistance.

Subject: Question about T2 (Average Surface Temperature 2) – Unexpected Initial Behavior

This time, I have a question regarding the T2 parameter (Average Surface Temperature 2).

In my simulation, I fixed the following parameters as follows:

• Ambient air temperature = 20°C

• Deep earth temperature = 20°C

• Initial fluid temperature = 20°C

• Amplitude of surface temperature = 0°C 

• Day of minimum surface temperature = start of simulation 

However, at the beginning of the simulation, I observe a strange behavior:
T2 increases from 20°C to about 21.56°C, then remains approximately constant afterward.

I tried many times, with and without insulation, always fixing all temperatures at 20°C — but I still get the same issue.

You will find attached a screenshot showing the simulation results, including  To :the outlet fluid temperature, T1, T2, T3 the average surface temperature .

Thank you in advance for your help!

Best regards,

ZAOUI

@mohammedzaoui

Both Types 997 and 999 can generate temperature profiles. You'll find some parameters near the end of the parameter list that control how often the profile is written. The name of the output file is set in the External Files tab; you'll get one output file per profile.

The profiles are text files. They are probably best viewed using the GroundTempViewer.exe that is in the .\TrnsysXX\Tools\ directory.

~david

@mohammedzaoui

The soil in the vicinity of the pipes is broken up into isothermal nodes that grow in size as they extend away from the pipes in the layers that you have defined. Beyond the farthest away nodes is soil whose temperature is not impacted by the pipes (called the far-filed). The noded soil is called the near-field. The data file contains the temperature of each node in the near field. The first line lists the number of nodes in the x, y, and z directions, lines 2-4 contain the dimensions of each node in x, y, and z respectively, and the remaining lines contain the node temperatures in x and y for each layer (node layer, not pipe layer) in the z direction.

The models contain a few different ways that the surface energy balance can be computed. In some, incident solar radiation and long wave radiation exchange between the surface and the sky are taken into account. It may be that your surface temperature is being impacted by one of those.

~david  

Dear David,

Thank you again for your clarification about the soil temperature output tool — it was very helpful.

I'm writing to you because I have observed unexpected results when testing the two boundary modes in Type 999.

Surprisingly, in the adiabatic mode (Mode 2), the temperature seems to propagate through the sides or into the deep earth, whereas in conductive mode (Mode 1), it appears to be blocked or much more limited. This is the opposite of what I was expecting.

Have you ever observed this kind of behavior before, or would you have any insight on what could be causing it?

Thank you again for your time .

Best regards,

ZAOUI

@mohammedzaoui

Mohammed, I think that what you are seeing in the adiabatic version of the ground temperature profile is that heat has built up inside the near field because it can't get "out" into the far field. As a result, the temperatures near the boundary are elevated. In the conductive version, the temperatures near the near field/far field boundary are colder because the energy coming from the piping layer isn't blocked by the adiabatic boundary and instead gets conducted away to the far field.

In general I think it is probably good practice to make sure that the far field boundary is far enough away that you don't see much difference between those two assumptions. If you are seeing a difference in the temperature field then it means that the location of the boundary is influencing the result.

best,

 David

@mohammedzaoui The "boundary" in the ground temperature viewer only shows temperatures to the edge of the modeled field (what's called the near-field); the far-field temperatures are not actually displayed. So what you are seeing in the adiabatic mode is that the temperature rises to about 26 C at the Z boundary between the near-field and the far-field; if the far-field temperature were also plotted, you would see a sharp drop just beyond the boundary to 21 C or whatever deep earth temperature (Parameter 40) was used. As David said, we expect temperatures to be elevated near the boundary because the energy can't "get out" into the far field, so the temperature profile does make sense. Likewise, in conductive mode the temperatures near the boundary are closer to the far-field temperatures, so this also makes sense.  

If you are using soil surface mode 2 (the default), you can verify the far-field temperatures at different depths by plotting them in an online plotter with Type77 (simple ground temperature model, in the Physical Phenomena standard library). Use the same soil properties, average surface temperature (or deep earth temperature), soil amplitude, and time shift in Type77 as used in Type999.

"Far-field" temperatures are defined as soil temperatures at a distance such that they are unaffected by the ground heat exchanger. I'll second that it's good practice to set the boundaries at a distance where the far-field boundary assumption doesn't much influence or affect the results.