RKStrand's profile - activity

I saw the issue on GitHub and have taken a quick look at it. I'll also post an answer there as well. The Input-Output

This is definitely a problem and is being worked on. When you selected the CondFD solution, that triggered an initializ

Sorry to hear that you are having issues with your radiant system model in EnergyPlus. It's hard to tell what exactly i

In response to the hints that you have been given, there needs to be some clarifications made as to what the EnergyPlus radiant model is doing.

1. The notion that there is no feedback between the radiant system, the wall, and the zone is absolutely false. There is feedback. While the radiant system, the wall, and the zone are modeled in different parts of the code, the results of the radiant system simulation is incorporated into the wall and zone calculation automatically via iteration. This is necessary to avoid any heat imbalances in the overall EnergyPlus heat balance based solution. This was one of the many advances over its predecessor programs like BLAST and DOE-2--that EnergyPlus integrated various solution parts. The EnergyPlus radiant model has always been integrated with both the zone and HVAC portions of the simulation.

2. The CTF method does approach some limits when dealing with very massive elements and other unusual constructions. The question is: how massive is your slab and what is the rest of your construction? Are you getting any warnings about not completing the warm-up period? The CTF approach as well as its extension to include heat sources/sinks present in radiant systems is fairly robust so I'm not sure that is the source of any differences you are seeing.

3. I completely agree with the comment about making sure that EnergyPlus is actually using the manual inputs you are trying to feed it. This is where I would start--making sure that your flow and water temperature data is matching what EnergyPlus is using. If it isn't, there shouldn't be an expectation that EnergyPlus surface temperatures are in agreement with your data.

I cannot speak to your experience of success with using an assumption of a 4-5K delta T other than to say that I am glad to hear that you were able to get things to work and were able to match the data in the cases you ran. What I will say though is that there is nothing built into EnergyPlus that makes a 10K delta T assumption. The sizing calculations in EnergyPlus are based on the delta T parameter that the user enters for the sizing of the plant loop in the input file. One can definitely adjust that parameter, but using autosizing with radiant systems requires more than force air systems

Strange. I'm not sure what could be causing this but perhaps we can still figure it out. So, if the constructions, materials, and internal gains are the same as the baseline model that is running without producing this error, what are the remaining differences between the baseline input file and the input file that is causing this error?

You said that you checked your model against a baseline model. Does this mean that you started with the baseline input and then made changes to it? If so, what changes did you make? It seems like there is some sort of instability that has been introduced. From what I have seen, this is particularly possible with unusual constructions. The error "moving around" simply means that another surface tripped up the temperature limit. It is highly possible that it is actually not a problem with the exterior or interior surface. Unfortunately, it is hard to tell without seeing your input file(s).

What about the constructions for various surfaces (the one that caused the temperature out of bounds error and also any others in the zone)? Something very lightweight/thin could cause a problem like this.