[TRNSYS-users] Surface Gain Inside - Pane Temp Error

Weber Daniel webedani at student.ethz.ch
Wed Aug 1 16:18:58 PDT 2018


Dear TRNSYS users,

I am modelling a building with Type 56 where I input a heat gain from a MATLAB TABS model to a ceiling surface as "Surface Gain Inside". Now the simulation gives me the error "... pane temperature out of bounds at time [second time step]" of the two windows in the zone (in the first timestep there is no heat flux as all initial temperatures are 20°C).
But when I divide my heat gain surface input by a factor of 10 in the MATLAB model it works. The total heat gain is then listed as roughly 750 kJ/h which is about 200 W (not much for a heating system, right?). The thing is, if I now input a constant heat gain I can go up to values of 7.5e6 kJ/h and the simulation starts running (later there is of course an error as something isn't realistic). But somehow if I input my MATLAB model output without dividing by a factor of 10 it doesn't work. I would expect my real heat gain without the division by 10 to be 7500 kJ/h (~2 kW) which is also realistic and nowhere near the possible constant 7.5 kJ/h I've tried.

Additionally, when I receive the error I get no output data for the second timestep, so I can't see how much my heat gain would've been. It seems TRNSYS runs into an error even before incorporating the heat gain from my MATLAB model.

I've tried the first of the following steps suggested in the archive but it didn't work. I haven't tried 2 and 3 as I don't think they could change anything:


"1) Set the TSKY= Tamb the most common mistake is a wrong sky temperature
2) look in the bui-file which orientation the window  (surface17 zone 2) has.
   Set the radiation for this orientation to 0 to check if there is a wrong connection.
3) check the raditive part of your heating and gains"


Any help is greatly appreciated.

Best Regards,

Daniel Weber
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.onebuilding.org/pipermail/trnsys-users-onebuilding.org/attachments/20180801/5ca0bbb6/attachment.html>


More information about the TRNSYS-users mailing list