I am modelling a shadowbox spandrel in EnergyPlus in order to obtain the max/min temperatures reached by the materials and then try to assess the risk of condensation. (model)

The geometry is the following: Zone1: 10cm thin shadow box (height:1m, lenght:1,5m), with a FenestrationSurface (Coated DGU) on the East side and Zone2 on the West side (trough an insulated wall). The other four surfaces have adiabatic BC. Zone2: a small box used to obtain the internal temperatures (22°C during the day and 18°C by night). At first I'd like to obtain the temperatures that the shadow box would reach without venting holes. The weather file is GBR_London.Gatwick.037760_IWEC. Running the attached model with Energyplus 8.5.0 I get several of the following warnings:

 ** Warning ** Inside surface heat balance did not converge with Max Temp Difference [C] =2.920E-003 vs Max Allowed Temp Diff [C] =2.000E-003*

and of the following severe errors:

* Severe * Temperature (low) out of bounds [-108.64] for zone="ZONE 1", for surface="SURFACE NORTH"*

Number of timesteps per hour is set to 60.

Given that running the model with ventilation (WindandStackOpenArea) I don't get severe errors I think that the problem might be the zone temperature limits of energyplus. (model) Although the Temperature out of bounds is negative I'm expecting to reach about 150°C in the shadowbox. Is there an upper temperature limit in Energy Plus? Is there a way to change it? I understand from this question that the room I'm trying to model is too small for the hypotheses of the convection correlations, could it be a reason for the missing convergence?