First time here? Check out the Help page!
1 | initial version |
Looks like the culprit is an EnergyPlus bug.
I took the CondFD1ZonePurchAirAutoSizeWithPCM.idf example file and made a few changes to it to simplify the model, the most important deleting the MaterialProperty:PhaseChange object in the IDF.
The model showcasing the bug has HeatBalanceAlgorithm = ConductionFiniteDifference. I made a sum of the heat flux into and out of the wall using
Output:Variable,*,Surface Outside Face Conduction Heat Transfer Rate per Area,Hourly;
Output:Variable,*,Surface Inside Face Conduction Heat Transfer Rate per Area,Hourly;
which results in the graph below in a five year simulation. As you can see, there is a trend of accumulation of heat in the wall. Keep in mind that this wall has NO PCM in it.
The bug disappears by changing to HeatBalanceAlgorithm = ConductionTransferFunction, which is the default choice in the IDFEditor. The graph is below and heat storage in the wall is cyclical, as it should be.
2 | No.2 Revision |
Looks like the culprit is an EnergyPlus bug. Unfortunately for @DerricksRoss, MaterialProperty:PhaseChange objects must use the HeatBalanceAlorithm to simulate, which appears to be causing the bug.
Here is a google drive link to the files.
Here is a link to the github issue I posted.
I took the CondFD1ZonePurchAirAutoSizeWithPCM.idf example file and made a few changes to it to simplify the model, the most important deleting the MaterialProperty:PhaseChange object in the IDF.
The model showcasing the bug has HeatBalanceAlgorithm = ConductionFiniteDifference. I made a sum of the heat flux into and out of the wall using
Output:Variable,*,Surface Outside Face Conduction Heat Transfer Rate per Area,Hourly;
Output:Variable,*,Surface Inside Face Conduction Heat Transfer Rate per Area,Hourly;
which results in the graph below in a five year simulation. As you can see, there is a trend of accumulation of heat in the wall. Keep in mind that this wall has NO PCM in it.
The bug disappears by changing to HeatBalanceAlgorithm = ConductionTransferFunction, which is the default choice in the IDFEditor. The graph is below and heat storage in the wall is cyclical, as it should be.
3 | No.3 Revision |
Looks like the culprit is an EnergyPlus bug. Unfortunately for @DerricksRoss, MaterialProperty:PhaseChange objects must use the HeatBalanceAlorithm = ConductionFiniteDifference to simulate, which appears to be causing the bug.
Here is a google drive link to the files.
Here is a link to the github issue I posted.
I took the CondFD1ZonePurchAirAutoSizeWithPCM.idf example file and made a few changes to it to simplify the model, the most important deleting the MaterialProperty:PhaseChange object in the IDF.
The model showcasing the bug has HeatBalanceAlgorithm = ConductionFiniteDifference. I made a sum of the heat flux into and out of the wall using
Output:Variable,*,Surface Outside Face Conduction Heat Transfer Rate per Area,Hourly;
Output:Variable,*,Surface Inside Face Conduction Heat Transfer Rate per Area,Hourly;
which results in the graph below in a five year simulation. As you can see, there is a trend of accumulation of heat in the wall. Keep in mind that this wall has NO PCM in it.
The bug disappears by changing to HeatBalanceAlgorithm = ConductionTransferFunction, which is the default choice in the IDFEditor. The graph is below and heat storage in the wall is cyclical, as it should be.