Orientation of cavities

[Anders Pedersen]

Hello

 

Is it possible to change the orientation of the cavities as the convection of the air is not the same in alle directions?

 

Best regards

 

 

[fostertom]

You're talking about Material>CEN frame cavities? You know that up/down/left/right don't mean 'on screen', but in reality? Up/down mean the heat flux is moving vertically in the real building/window; and left/right mean it is moving horizontally.

What I want to know is, why left and right - what difference does it make, and why not fore and aft as well? Previous versions just had 'horizontal' - what's wrong with that?

Another thing - what's the Side1/Side2 temperatures about? These temps are auto-filled (but can be edited), always something different, so not a fixed default figure, must be computed somehow, but no resemblance to actual, often e.g. -4C when the lowest temp anywhere is 0C. And changing the temps seems to make no difference to anything. So what's it about?

Is the cavity's lamda calculated depending on its width? It should be, as lamda varies greatly with cavity extent, both width and height.

Can these cavities be very wide e.g. to model an enclosed unheated porch/comservatory? What is the max width a cavity can be?

[fostertom]

Ah maybe you're asking ... You have to select a polygon then Libraries>Select Material, before you're presented with the dialog where you can set the orientation (and other things).

[Marles McDonald]

Anders and Tom,
I'll try to answer some of the questions, first fair warning, I'm a long time user but not one of the programers, nor am I an engineer or physicist so if I make a physics error someone correct me.

As a practical matter the orientation of the cavities is determined by the selection of the correct frame type from Therm File Properties/ Cross section Type. Selecting the correct type sets the orientation of the section relative to gravity so the convective flow is correct for the section. In view/gravity arrow you can check the direction of gravity relative to the section. If you select head or sill for example the gravity direction is down, if you  select a vertical type ( jamb for example) you will note the direction is still down but all you see is the tail of the arrow.

In regards to the screen directions, these are just determined as conventions, for practical reasons the left side of the drawing ( looking at the screen) is assumed to be the exterior surface, the right the interior surface. since Therm is only 2D gross heat flux is then from right to left looking at the screen but since looking at the screen is a mirror image of reality the call out direction is left to right. selecting the correct cross section type makes left to right heat flux work in all cases.

Autofill temperature are a starting point for the calculations, if you'll look at a cavity after calculation the values become reasonable.

For vertical cavities ( jambs) the default cavity height is 1 meter, this can be changed in Therm/ options/preferences.

I'm not sure what maximum cavity widths work in Therm but I do know that large model cavities are not supported. Therm is intended to work for parts of the building structure, not full building sections.

I hoped this helped and I didn't get much wrong,  Under documentation on the Therm web page the NFRC manual and the Therm2 Manual are both  available. I suggest that you download these, both are very good tutorials for the use of Therm and Window. The NFRC manual settings, etc will be wrong for those using EN673/ISO 10077 but the process will be basically correct regardless of the Standards used.

[fostertom]

Thanks Marles - beginning to see who is active here!

Didn't know about orienting via Cill vs Jamb - simply means vertical section vs plan section? - no further implications?

About left/right - in a typical wall corner plan, heat is going not just leftwards on screen, but 'up' as well (and even rightwards too, in an 'interesting' model). Do we really have to specify heat flux direction via left/right, or does either left or right equally serve as indicating horizontal flow? In the v2 manual it says the only options are up down and horizontal.

BTW, any comment on my Flux Vector Line idea (which you 'me too'd'), and the use for it that I tried to describe - make sense, or just a hard way of doing something daft?

[D. Charlie Curcija/LBNL]

THERM recognizes 4 cardinal orientations for heat transfer in the plane of the screen and these directions are not meant as input, but are result of calculation. In-between heat flow directions are reduced to these 4. Besides informing users which way heat flows, this is also used in deciding which correlation to apply in a cavity. Selection of correlations are documented in a technical documentation posted on the web site.

I have posted reply on flux vector line idea.

[fostertom]

To summarise my understanding - is this correct?

Gravity direction can be set in Properties>Sill (= vertical section) or Jamb (= plan section), or by View>Gravity Arrow, but not necessary (makes no difference) if the model is entirely solid, no cavities. 'Sill', 'Jamb' etc have no further implication, other than setting the orientation.

It is not necessary to draw the model such that heat flow is from right to left on-screen, tho this is the convention. If heat flow is both leftward and upward on-screen (as in a typical building fabric, walls meeting at corner, model), that's OK.

When it comes to cavities, gravity direction must be defined, and in setting the cavity 'Material', heat flow direction must be set.

In v2 manual, the only options are Up Down or Horizontal, presumably relative to gravity direction.
But in v5.2 on, there's Up Down Left and Right, and in v7.2 the dialog box seems to make clear that these are now relative to the screen, no longer relative to gravity direction - is that correct?

However, in a plan section of a corner where 2 walls meet, heat flow goes leftward (on-screen) and upward (on-screen) as well. Both 'legs' include CEN unventilated cavities, narrow measuring parallel to heat flow direction, and extensive measuring crosswise to flow.
Whatever combinations of setting gravity direction, as in para 1 above, but setting heat flow direction separately for the two 'legs', one Up, the other Left (as correct on-screen), I get nonsensical result:

To the left vertical (on-screen) 'leg', I set Heat Flow Direction [Left] (relative to screen); dialog reports "Jamb: left to right (relative to gravity)" - that is correct.
To the top horizontal (on-screen) 'leg', I set Heat Flow Direction [Up] (relative to screen); dialog reports "Jamb: bottom to top (relative to gravity)" - that is wrong.
But if I set the latter to [Left] (relative to screen) - which is not true - I get the satisfactory report "Jamb: left to right (relative to gravity)" - which is correct.

Am I missing something, or is this a glitch? (temporarily unable to upload either a thm or a pic, for some reason)

[fostertom]

Here's the file

[D. Charlie Curcija/LBNL]

Correct about gravity vector.

Correct that it is not necessary to draw model where heat flow is from right to left. This is default orientation of the model for NFRC simulations, which you can see when you import glazing from WINDOW, but you can orient the model anyway you want (including with heat flow going top to bottom).

Correct that for cavities gravity needs to be correctly defined. Most frequently it is done through the definition of cross-sections, but can be done by setting gravity arrow manually.

V2 manual is outdated. We are going to issue the updated version of manual for v7.  Yes, the heat flow direction indoicated in frame cavity dialog box is relative to the screen.

The rest of your discussion about setting heat flow directions: Heat flow direction is NOT set by user. This is very important because this is only for information purposes and is calculated by program based on the model setup. THERM calculates heat flow direction for each frame cavity based on the heat flux vectors and heat flow intensities and it has only 4 cardinal directions. While heat flow in reality will be somewhere in between those 4 directions, THERM calculates which cardinal direction is closest and displays that one. Direction of heat flow is relevant ONLY to frame cavities (convection and radiation). It is NOT relevant to solid, other than when you display heat flux vectors and isoflux lines, which gives you information about the intensities and directions of heat flow. In the case of frame cavities, this direction is used for selection of appropriate correlation for calculations, while for solids it has no bearing on the calculation.

Charlie

[fostertom]

All clear now (I think) - thanks v much for time and attention - on Sunday too!

[Christian Kohler/LBNL]

Your questions about setting the heat flow direction for cavities made me realize that it is somewhat confusing that those fields are changeable. As Charlie explained, the values in these fields are updated after the simulation. The values you put in are used as an initial guess, but the defaults should be fine. We will probably grey-out this fields in the next version, and add an option in Preferences to allow changing them.

[fostertom]

Great. I hope Help will explain when and why I might want to make that change in Preferences.

I'm a bit puzzled - in a 2D model, what can it mean for the field to report that within the cavity heat is flowing from top to bottom of the cavity i.e. flowing down into the screen?