Fix VRF Heat Pump Total Heating Rate not match the sum of coil heating rate (with piping loss) #10627

yujiex · 2024-07-26T23:59:47Z

Pull request overview

Fixes VRF FluidTCtrl heat pump total heating rate doesn't match zone VRF air terminal total heating rate #10625

Details on the fix of heat pump heating rate not matching total heating coil heating rate

The issue of heat pump heating rate not matching total heating coil heating rate, is caused by three factors

In function ControlVRFIUCoil, the capacity calculation method in FanSpdResidualHeat uses temperature difference (TotCap = FanSpdRto * Garate * 1005.0 * (Tout - Tin)) rather enthalpy difference (TotCap = FanSpdRto * RatedAirMassFlowRate * (PsyHFnTdbW(Tout, Wout) - PsyHFnTdbW(Tin, Win))). This is different from the coil heating rate calculation, causing some mismatches. So a function FanSpdResidualHeatUsingH is used to compute the capacity instead.

after using FanSpdResidualHeatUsingH in the coil heating rate calculation, the mismatch reduction is as follows

// before
      diff          percent.diff     
 Min.   :-73.044   Min.   :-1.03251  
 1st Qu.:  0.000   1st Qu.: 0.00000  
 Median :  0.000   Median : 0.00000  
 Mean   : -3.613   Mean   :-0.05856  
 3rd Qu.:  0.000   3rd Qu.: 0.00000  
 Max.   :  0.000   Max.   : 0.00000  

// after
      diff         percent.diff     
 Min.   :-24.84   Min.   :-0.41946  
 1st Qu.:  0.00   1st Qu.: 0.00000  
 Median :  0.00   Median : 0.00000  
 Mean   : -1.14   Mean   :-0.01852  
 3rd Qu.:  0.00   3rd Qu.: 0.00000  
 Max.   :  0.00   Max.   : 0.00000

Within each time step, this->HeatingCapacity, this->PipingCorrectionHeating, HeatingPLR are involved in the calculation of VRF heat pump heating rate, this->TotalHeatingCapacity. In each time step, this->HeatingCapacity and this->PipingCorrectionHeating is evaluated twice. The coil heating rate calculation uses the values from the first round while the output of heat pump heating rate and piping correction factor uses the values from the second round. By re-evaluating coil heating rate using the updated OU capacity and heating rate, and removing the 100W lower bound on coil heating rate calculation, the difference between the coil heating rate and heat pump heating rate reduced to below 1E-10

// before
      diff         percent.diff     
 Min.   :-24.84   Min.   :-0.41946  
 1st Qu.:  0.00   1st Qu.: 0.00000  
 Median :  0.00   Median : 0.00000  
 Mean   : -1.14   Mean   :-0.01852  
 3rd Qu.:  0.00   3rd Qu.: 0.00000  
 Max.   :  0.00   Max.   : 0.00000  

// after
      diff             percent.diff       
 Min.   :-6.366e-12   Min.   :-1.820e-13  
 1st Qu.:-4.547e-13   1st Qu.:-1.515e-14  
 Median : 0.000e+00   Median : 0.000e+00  
 Mean   :-2.070e-14   Mean   :-4.444e-16  
 3rd Qu.: 4.547e-13   3rd Qu.: 1.479e-14  
 Max.   : 6.366e-12   Max.   : 1.554e-13

The following is relevant debugging prints

// the following is a timestep, heating coil heating rate is 5849.6, heat pump heating rate is 5921, 
reported piping correction factor is 0.984. However 5921 x 0.984 is about 5826, 
different from coil heating rate of 5849.6.
   timestamp           heat.pump.heating.rate piping.correction.heating coil.heating.rate  diff percent.diff
   <dttm>                               <dbl>                     <dbl>             <dbl> <dbl>        <dbl>
 1 2024-01-06 22:40:00                  5921.                     0.984             5849.566. -24.8       -0.419

// There are two rounds of evaluations of this->HeatingCapacity and this->PipingCorrectionHeating
01/06 22-4,this->HeatingCapacity=5930.29974800236,PipingCor=0.9863862943052852
=7000.000000000003/(7000.000000000003+96.61117597961002)
01/06 22-4,this->HeatingCapacity=5921.339831212836,PipingCor=0.9837523896967507
=5849.5663925516155/(5849.5663925516155+96.61117592655613)

// coil TotalHeatingEnergyRate 5849.6 uses the OU capacity of 5930.3 and piping correction of 0.986. 
This is from the first round of this->HeatingCapacity evaluation. 
The following are some more details
// thisDXCoil.TotalHeatingEnergyRate = AirMassFlow * (OutletAirEnthalpy - InletAirEnthalpy) * PartLoadRatio
=5849.5663925516155
=0.4523946203083688 * (41284.34272289792 - 28354.112941287483) * 1,
Here the demand QCoilReq=5849.566392551615, it gets value from the MaxHeatCap, which is a result of LimitTUCapacity. 
This function restricts coil output to not exceed OU capacity. 
The TotalTUHeatingCapacity (OU capacity) value is calculated using 5930.3 x 0.986, 
corresponding to first round of this->HeatingCapacity evaluation
01/06 22-4,TotalTUHeatingCapacity=5849.566392551615
=5930.29974800236 * 0.9863862943052852

// Heat pump heating rate (this->TotalHeatingCapacity) uses value from the second round
// VRF heat pump heating rate is computed here
this->TotalHeatingCapacity = TotalCondHeatingCapacity * HeatingPLR;
// TotalCondHeatingCapacity is 5921.3 (which gets value from this->HeatingCapacity), 
so is this->TotalHeatingCapacity (as HeatingPLR is 1 here), piping correction is 0.9838. 
01/06 22-4,
TotalTUHeatingCapacity = TotalCondHeatingCapacity * this->PipingCorrectionHeating
=5825.132209162181=5921.339831212836 * 0.9837523896967507

before fixing
eplusout_develop.csv

after fixing
eplusout_recalcCoilHeat.csv

Regression diffs

Regression diffs happen in four files

US+SF+CZ4A+hp+crawlspace+IECC_2006_VRF
VariableRefrigerantFlow_FluidTCtrl_5Zone
VariableRefrigerantFlow_FluidTCtrl_HR_5Zone
VariableRefrigerantFlow_FluidTCtrl_wSuppHeater_5Zone

Meter diffs happen in VariableRefrigerantFlow_FluidTCtrl_5Zone.idf and
VariableRefrigerantFlow_FluidTCtrl_HR_5Zone.idf. It happens in the following fields. It's reasonable as heating coil heating rate is updated in this PR. The magnitude of the maximum percent difference is around 1e-5 to 1e-6.

HeatingCoils:EnergyTransfer J
HeatingCoils:EnergyTransfer J
Electricity:Facility J
Electricity:HVAC J
Electricity:Facility J
Electricity:HVAC J

Table diffs happen in

VariableRefrigerantFlow_FluidTCtrl_5Zone
VariableRefrigerantFlow_FluidTCtrl_wSuppHeater_5Zone

Table diffs are in Coil Total Capacity at Rating Conditions and Coil Sensible Capacity at Rating Conditions for three heating coils, as well as Heating:EnergyTransfer, General:Heating:EnergyTransfer, HeatingCoils:EnergyTransfer, Heating:EnergyTransfer:Zone*. These are also expected as coil heating rate is updated.

ESO output happens in all four files

US+SF+CZ4A+hp+crawlspace+IECC_2006_VRF
VariableRefrigerantFlow_FluidTCtrl_5Zone
VariableRefrigerantFlow_FluidTCtrl_HR_5Zone
VariableRefrigerantFlow_FluidTCtrl_wSuppHeater_5Zone

Take the US+SF+CZ4A+hp+crawlspace+IECC_2006_VRF.idf as example, the Heating Coil Heating Rate is different (0.005 relative diff) as expected. Zone heating demand is relative stable (1e-7 relative diff), the supplemental heating coil heating rate also differs as a result of the heating coil heating rate change. The diffs in cooling rate and related fields are very small (1e-12 relative diff). Heat Pump Terminal Unit Heating Load Rate (0.003 relative diff).

regression diff unfinished

NOTE: ENHANCEMENTS MUST FOLLOW A SUBMISSION PROCESS INCLUDING A FEATURE PROPOSAL AND DESIGN DOCUMENT PRIOR TO SUBMITTING CODE

Pull Request Author

Add to this list or remove from it as applicable. This is a simple templated set of guidelines.

Reviewer

This will not be exhaustively relevant to every PR.

previously it says HP total heating rate is the sum of air terminal heating rate (which includes supplemental heating coil heating rate). This is wrong for both the curve-based model and FluidTCtrl model. FluidTCtrl model doesn't have an entry to this output field, so one is added.

coil heating rate calculation is like the following AirMassFlow * (OutletAirEnthalpy - InletAirEnthalpy) * PartLoadRatio In the computation of fan speed ratio, the coil heating capacity is FanSpdRto * Garate * 1005.0 * (Tout - Tin) The difference between 1005.0 * (Tout - Tin) and (OutletAirEnthalpy - InletAirEnthalpy) can cause mismatches between coil heating rate and heating demand Here we change the fan speed ratio caculation to also use the enthalpy difference way to compute its heating capacity to match the demand

mjwitte · 2024-08-16T15:09:37Z

doc/input-output-reference/src/overview/group-variable-refrigerant-flow-equipment.tex

 \paragraph{VRF Heat Pump Total Heating Rate {[}W{]}}\label{vrf-heat-pump-total-heating-rate-w}

-This output field is the operating total heating capacity of the variable refrigerant flow heat pump in Watts. The capacity includes any degradation due to defrost mode. This value is calculated for each HVAC system time step being simulated, and the results are averaged for the time step being reported. This value should match the sum of the individual zone terminal unit output variables for Zone VRF Air Terminal Total Heating Rate.
+This output field is the operating total heating capacity of the variable refrigerant flow heat pump in Watts. The capacity includes any degradation due to defrost mode. This value is calculated for each HVAC system time step being simulated, and the results are averaged for the time step being reported. This value should match the sum of the individual zone terminal unit heating coil output variables for Heating Coil Heating Rate.


It thought this should say "sum of the individual zone terminal unit heating coil output variables for Heating Coil Heating Rate less any piping loss."

It might just be the total of heating coil heating rate rather than total coil heating rate - piping loss.

in CalcVRFCondenser, vrf.TotalHeatingCapacity (VRF Heat Pump Total Heating Rate output variable) is computed like this

vrf.TotalHeatingCapacity = TotalCondHeatingCapacity * HeatingPLR * CyclingRatio;

And in HeatingPLR is like this

if (TotalCondHeatingCapacity > 0.0) { HeatingPLR = min(1.0, (TUHeatingLoad / vrf.PipingCorrectionHeating) / TotalCondHeatingCapacity); } else { HeatingPLR = 0.0; }

For normal cases where TotalCondHeatingCapacity > 0.0,

vrf.TotalHeatingCapacity would just be

TotalCondHeatingCapacity * ((TUHeatingLoad / vrf.PipingCorrectionHeating) / TotalCondHeatingCapacity) * CyclingRatio = TUHeatingLoad / vrf.PipingCorrectionHeating * CyclingRatio

TUHeatingLoad / vrf.PipingCorrectionHeating would be coil heating rate plus the piping loss.

Right, I meant coil heating rates plus piping loss is the output from the outdoor unit.

I see. Sorry I misunderstood earlier. I will modify the description.

mjwitte · 2024-08-16T15:10:21Z

doc/input-output-reference/src/overview/group-variable-refrigerant-flow-equipment.tex


 Note: refer to the rdd file after a simulation for exact output variable names

+\paragraph{VRF Heat Pump Total Heating Rate {[}W{]}}\label{vrf-heat-pump-total-heating-rate-w}


This needs a unique label (there are duplicate label warnings when file changes are viewed in github).

Good catch. I'll add a suffix to it

mjwitte · 2024-08-16T15:11:53Z

src/EnergyPlus/DXCoils.cc

    Garate = state.dataDXCoils->DXCoil(CoilIndex).RatedAirMassFlowRate(1);
    // why always limit the minimum fan speed ratio to 0.65?
-    FanSpdRatioMin = min(max(OAMassFlow / Garate, 0.65), 1.0); // ensure that coil flow rate is higher than OA flow rate
+    FanSpdRatioMin = min(max(OAMassFlow / Garate, 0.0), 1.0); // ensure that coil flow rate is higher than OA flow rate


This seems like a big change, unrelated to the heating rate outputs. Does this branch include some changes from another branch?

I might have left it in by accident. Just reverted it.

…eatRate

yujiex · 2024-08-26T21:53:49Z

@mjwitte @rraustad would you mind taking another look at this PR? Thanks. I was wondering whether the type of adjustment in HVACVariableRefrigerantFlow.cc look Okay.

rraustad · 2024-08-26T22:40:50Z

src/EnergyPlus/DXCoils.cc

+    if (std::abs(ZnSenLoad) < 100.0) ZnSenLoad = sign(100.0, ZnSenLoad);
+    Real64 Wout = Win;
+    Real64 TotCap = FanSpdRto * RatedAirMassFlowRate * (PsyHFnTdbW(Tout, Wout) - PsyHFnTdbW(Tin, Win));
+    return (TotCap - ZnSenLoad) / ZnSenLoad;


Do you really need to iterate on this? There's only 1 unknown.., FanSpdRto.

I guess not. I will change it to directly calculate FanSpdRto

rraustad · 2024-08-26T22:46:00Z

src/EnergyPlus/DXCoils.cc

+            Tout = Tin + (Ts_1 - Tin) * (1 - BF);
+            Real64 RatedAirMassFlowRate = state.dataDXCoils->DXCoil(CoilIndex).RatedAirMassFlowRate[0];
+            auto f = [QCoilSenHeatingLoad, RatedAirMassFlowRate, Tout, Tin, Win](Real64 FanSpdRto) {
+                return FanSpdResidualHeatUsingH(FanSpdRto, QCoilSenHeatingLoad, RatedAirMassFlowRate, Tout, Tin, Win);


I am not saying to change this but it just seems odd to me when meeting a load to modulate the fan based on suction temperature (which meets the load using air flow) instead of modulating the compressor at some known fan speed. I guess this is an artifact of using VS fan. I would hope in the case of a VS fan that the refrigerant suction T is relatively constant.

It seems that modulating the compressor is at the next step after the calculation of the TU's are finished. In this function, I don't think refrigerant suction temperature changes

rraustad · 2024-08-26T23:05:10Z

src/EnergyPlus/HVACVariableRefrigerantFlow.cc

+    }
+    if (this->TUHeatingLoad / this->PipingCorrectionHeating > TotalCondHeatingCapacity) {
+        this->TotalHeatingCapacity = this->HeatingCapacityPrev * HeatingPLR * this->PipingCorrectionHeatingPrev;
+    }


TotalHeatingCapacity should match the sum of the TU capacity + piping losses. So isn't it just that? TotalHeatingCapacity = Q_h_TU_PL = TU_HeatingLoad + Pipe_Q_h?

yes, it is matching this.

rraustad · 2024-08-26T23:07:00Z

testfiles/US+SF+CZ4A+hp+crawlspace+IECC_2006_VRF.idf

-    autosize,                !- Cooling Outdoor Air Flow Rate {m3/s}
-    autosize,                !- Heating Outdoor Air Flow Rate {m3/s}
+    0,                !- Cooling Outdoor Air Flow Rate {m3/s}
+    0,                !- Heating Outdoor Air Flow Rate {m3/s}


Does including OA flow cause a difference between the sum of TU heating capacity + piping losses and condenser total heating capacity?

it does cause some more difference

@rraustad Here is the output file when the heating cooling air flow rate is autosize
eplusout_when heating cooling air flow rate autosize.xlsx

The following is a snapshot sorted with column O in descending order

If L x M = 9083.83288 * 0.98793126 = 8974.2, which closely matches column K at 8973.19993, then why is the TU heating coil "allowed" to provide more than that "max" heating rate? Is the TU heating coil capacity getting limited by the MaxHeatingCapacity variable? I recall discussing that function LimitTUCapacity should include piping losses.

Coil capacity should have been limited. Maybe there's some lingering issues there. I will check on that.

@rraustad I've changed to just using TU_HeatingLoad for this->TotalHeatingCapacity. Now the total "Heating Coil Heating Rate" and the "VRF Heat Pump Total Heating Rate" equals (regardless of whether the "Heating/Cooling Outdoor Air Flow Rate" is autosize or 0). The following are the outputs for these two cases.

eplusout_0 heating cooling air flow.csv
eplusout_autosize heating cooling air flow.csv

The coil capacity is indeed limited. I set the OU evaporative capacity to 5000W (coil capacity is 10023W). Coil heating rate is less than OU capacity at max speed.

…eatRate

github-actions · 2024-09-03T20:58:01Z

⚠️ Regressions detected on macos-14 for commit `b9a481a`

Regression Summary

EIO: 3
ESO Big Diffs: 3
Table Small Diffs: 1
MTR Small Diffs: 1
Table Big Diffs: 2
Table String Diffs: 1
ESO Small Diffs: 1

dareumnam · 2024-09-10T21:05:29Z

@rraustad @mjwitte Do you have any further comments on this? If this PR needs more time for discussion, we can move it to the 25.1 bug freeze milestone, as we're planning to tag the first RC for 24.2 tomorrow.

rraustad · 2024-09-11T18:29:55Z

doc/input-output-reference/src/overview/group-variable-refrigerant-flow-equipment.tex

 \paragraph{VRF Heat Pump Total Heating Rate {[}W{]}}\label{vrf-heat-pump-total-heating-rate-w}

-This output field is the operating total heating capacity of the variable refrigerant flow heat pump in Watts. The capacity includes any degradation due to defrost mode. This value is calculated for each HVAC system time step being simulated, and the results are averaged for the time step being reported. This value should match the sum of the individual zone terminal unit output variables for Zone VRF Air Terminal Total Heating Rate.
+This output field is the operating total heating capacity of the variable refrigerant flow heat pump in Watts. The capacity includes any degradation due to defrost mode. This value is calculated for each HVAC system time step being simulated, and the results are averaged for the time step being reported. This value should match the sum of the individual zone terminal unit heating coil output variables for Heating Coil Heating Rate plus any piping loss.


This does seem like the correct definition for the VRF heating capacity.

rraustad · 2024-09-11T18:40:48Z

src/EnergyPlus/HVACVariableRefrigerantFlow.cc

    this->TotalCoolingCapacity = TotalCondCoolingCapacity * CoolingPLR;
-    this->TotalHeatingCapacity = TotalCondHeatingCapacity * HeatingPLR;
+    // adjustment for matching HP heating rate and coil heating rate
+    this->TotalHeatingCapacity = TU_HeatingLoad;


This is not actually correct. The TU_HeatingLoad is the target (coil load + piping loss) that the model will iterate on to find a solution. The TotalHeatingCapacity is the result. With a tolerance on the iterations these 2 variables should not match. They should be very close (like within 0.001%), but not match. How far off TotalHeatingCapacity is from TU_HeatingLoad tells you how well the model did to calculate the final result. To equate these at the end will hide any issues with the model. TU_HeatingLoad should match exactly with the coil loads + piping loss. TotalHeatingCapacity should represent the refrigerant side heating capacity. Since this model has refrigerant properties could TotalHeatingCapacity = Mdot deltaH on the refrigerant side? Then you would know how well the model did to meet the load.

I see.

I think the piping correction is not consistent in the IU and OU calculation, which might have partly caused the discrepancy. See comment here: #10625 (comment)

There might be other places contributing to it as well. I've been playing around with adjusting stuff but so far I've not been able to figure out how to effectively resolve the discrepancy.

Maybe just revert this one change and revisit how different these 2 results are. TotalHeatingCapacity versus TU_HeatingLoad. Maybe also add another report for refrigerant side capacity?

I will do that

Myoldmopar · 2024-09-11T19:43:00Z

Thank you both for the comments this afternoon. I'm going to mark, for now, that this is not going into 24.2. If things get resolved and we agree, we can certainly go ahead and drop it in. But I'm going to change it to focus on other things.

yujiex · 2025-02-06T22:30:12Z

@yujiex just touching base on here. It seems like this had momentum, but then it got slowed down trying to get some things to agree. Could you comment on the latest status? If it's just that the current code could be merged as an improvement, with a follow-up to continue the investigation, then I'll work to get this merged in. If the current code is not necessarily the right changes, then obviously we'll just keep holding. Also, if you want to get the documentation changes in while you continue investigating code, we can trim this down to just documentation changes.

Hi @Myoldmopar, the current one still can't fix the issue of heat pump heating rate not matching the heating coil heating rate. I think the major thing is in the piping loss correction. But I still can't figure out how it should be fixed. So maybe we can merge in the documentation change first? Do I create a new PR with documentation changes in the new PR?

nrel-bot-2 · 2025-02-14T01:00:25Z

@yujiex it has been 7 days since this pull request was last updated.

nrel-bot-2 · 2025-02-21T00:19:51Z

@yujiex it has been 7 days since this pull request was last updated.

yujiex · 2025-02-25T23:43:41Z

@Myoldmopar I put the documentation update in a new PR (#10957). Maybe we can merge that one in first? I will continue investigating the remaining issues in this PR.

…eatRate

github-actions · 2025-02-26T18:12:17Z

⚠️ Regressions detected on macos-14 for commit `e333a28`

Regression Summary

EIO: 1
ESO Big Diffs: 4
Table Big Diffs: 3
MTR Small Diffs: 2
Table Small Diffs: 1
Table String Diffs: 1

…eatRate

yujiex · 2025-02-27T21:56:25Z

@mjwitte and @rraustad I modified the code to update the coil heating rate after the latest piping correction and heat pump capacity is computed. The 100W capacity limit is removed as there was not such a limit on the OU side: if OU was less than 100W or when the piping correction is very low, there will be some coil and OU mismatch. Some details are added in the PR description to explain the fixes. Please see if they make sense. Thanks.

github-actions · 2025-02-27T22:46:25Z

⚠️ Regressions detected on macos-14 for commit `bb3d010`

Regression Summary

EIO: 1
ESO Big Diffs: 4
Table Small Diffs: 2
MTR Small Diffs: 2
Table Big Diffs: 2
Table String Diffs: 1

rraustad · 2025-02-28T18:40:23Z

@yujiex this looks very close now with very little difference (1E-11) between the TU heating rate and the HP heating rate adjusted for piping losses. Are you finished with this branch?

yujiex · 2025-02-28T18:51:43Z

@yujiex this looks very close now with very little difference (1E-11) between the TU heating rate and the HP heating rate adjusted for piping losses. Are you finished with this branch?

@rraustad Mostly finished. I will need to look into the regression diffs in more detail but the list of files that has diffs are reasonable.

rraustad · 2025-02-28T20:10:25Z

src/EnergyPlus/HVACVariableRefrigerantFlow.cc

+                                              ActualSC);
+                    Real64 AirMassFlow = FanSpdRatio * heatingCoil.RatedAirMassFlowRate(1); // heating mode
+                    // fixme: temporary remove multiply of coil PLR
+                    heatingCoil.TotalHeatingEnergyRate = AirMassFlow * (OutletAirEnthalpy - heatingCoil.InletAirEnthalpy);


Why is the parent writing to the child? The parent should call the child and the child should calculate it's own operation. Isn't the child heating coil TotalHeatingEnergyRate here already equal to AirMassFlow * (OutletAirEnthalpy - heatingCoil.InletAirEnthalpy)? i.e., wasn't TotalHeatingEnergyRate set in the child on the call above?

The coil TotalHeatingEnergyRate was set when the demand QCoilReq was calculated with the old piping loss (calculated with rated values) and OU capacity. The following shows the section in the PR description with details on this. 0.986 piping loss used the rated capacity of 7000W in the calculation. So the above is added to recalculate the coil heating rate with the updated piping loss.

When the parent changes the child's TotalHeatingEnergyRate, should the heating coil outlet air condition also change?

oh probably that too. I'll add these to save the updated outlet values

heatingCoil.OutletAirTemp = OutletAirTemp; heatingCoil.OutletAirHumRat = OutletAirHumRat; heatingCoil.OutletAirEnthalpy = OutletAirEnthalpy; heatingCoil.ActualSH = ActualSH; heatingCoil.ActualSC = ActualSC;

And if the heating coil outlet condition changes, then doesn't the fan speed need to change to meet the same load? And if the fan speed changes doesn't the fan heat also change? This is a good example of why the parent shouldn't update child component performance. Couldn't you just send the piping loss adjusted QCoilReq over to the heating coil?

I don't see piping losses used in LimitCoilCapacity to calculate the limit, if piping losses were used there would that correct the problem?

RemainingCapacity = TotalCapacity * PipingLosses;

When the TU capacity is limited the OU will be running at full load. I think you are saying you don't know the operating full load capacity yet. I think that's OK and the next iteration will include the correct piping losses that were calculated at the end of the previous iteration. I think there will be, within a time step's iteration sequence, a FirstHVACIteration = true, FirstHVACIteration = false, and then a final FirstHVACIteration = false (although I am not positive about this last one happening all the time). So the piping losses should be correct on the second iteration and the final iteration should have the correct answer? I do still think PipingLosses needs to be included in LimitCoilCapacity.

I just checked the calls into LimitCoilCapacity and what is passed in as TotalCapacity is already corrected for piping losses. That makes me wonder what is actually happening for this issue.

TotalTUCoolingCapacity = TotalCondCoolingCapacity * this->PipingCorrectionCooling; if (TU_CoolingLoad > TotalTUCoolingCapacity) { LimitTUCapacity(state, VRFCond, NumTUInList, TotalTUCoolingCapacity,

Does this issue only happen in heating mode?

it did correct for piping loss, but the piping correction factor was computed using the rated capacity, not the actual capacity, so that piping correction is also not right at the time of limitTUCapacity

This issue also happens for cooling mode

nrel-bot-2 · 2025-04-10T00:24:40Z