[TRNSYS-users] Types 68 and 34 (shading)

[Fernando Domínguez Muñoz]

Dear TRNSYS users,

I have some doubts about TYPES 34 (overhang and wingwall shading) and 68
(shading by external object) that I would like to discuss with you.
These problems are related to the diffuse radiation calculations.

In order to model a vertical window with an overhang and a wall in front
(of the window), I tried to use types 16, 68 and 34 connected in
cascade. The first parameter of type34 changes the way in which the view
factors are calculated, being offered the following two choices:

(a) If the upstream component is type16, there is no problem:

    Sky diffuse on window = Sky diffuse on horizontal * (0.5 – Fws)

Where Fws = view factor between window and sky (takes account of the
overhang and the windwalls)

(b) If the upstream component is type68, 

  Sky diffuse on window = Shaded sky diffuse on window without overhang
* (1-Fws)

In both cases (a) and (b) I suppose that the sky diffuse radiation input
is always the input number 4 of type34. The label in the proforma is
somewhat error prone (it says “sky diffuse on the horizontal”).

In case (b), locations behind the window were already excluded (by
type16), so the first term is 1. I do not understand the second term,
because the overhang is obstructing part of the sky over the window, so
the involved radiation should be that coming from the vault of heaven,

(b*)  Sky diffuse on window = 1*Shaded sky diffuse on window without
overhang - Sky diffuse on horizontal*Fws

In fact, suppose that type68 is used without obstacles defined (all
obstruction height angles equal to zero). In this case
shaded_view/full_view = 1, so

  Shaded sky diffuse on window without overhang=0.5* Sky diffuse on  
horizontal 

The results of expressions (a) and (b) are inconsistent:

(a) Sky diffuse on window = 0.5* Sky diffuse on horizontal – Fws* Sky
diffuse on horizontal

(b) Sky diffuse on window = 0.5* Sky diffuse on horizontal – 0.5*Fws*
Sky diffuse on horizontal

A different and more difficult situation is when the overhang or a
wingwall (or part of it) sees the external obstacle (that is, there are
two obstacles in series). In this case, part of the shaded radiation is
obstructed. So as to solve this problem, I would try to compose the
geometry of the problem by combining several masks.

Other questions are:

Q1) The geometrical dimensions of the window can be comparable to those
of the obstacle in some cases. In this situation strict view factor
algebra should be used, but if we use type68 it is not clear where to
place the measurement reference for the obstruction angles, because
these angles are different for different points on the window. The best
place seems to be the centre of the window.

Q2) When type68 is used in conjunction with type1 (flat solar
collector), I think that the description given in the proforma is error
prone, because the total radiation given by type68 does not include the
ground reflected radiation, so it should not be directly connected with
the total radiation input of the solar collector type. This term can be
difficult to calculate when the obstacle is close to the collector (in
this case the radiation reflected by the obstacle can be significant),
but shall be included anyway.

I hope this helps.

Fernando Domínguez Muñoz
University of Málaga (Spain)