Question-and-Answer Resource for the Building Energy Modeling Community
Get started with the Help page
Ask Your Question

VRF & DOAS Design & Operation

asked 2018-03-29 11:17:06 -0600

Matt Koch gravatar image

updated 2018-03-29 14:26:22 -0600

So, I tried the VRF & DOAS measure from BCL in OpenStudio 2.4. I have a few questions.

1) Where in the EnergyPlus output, can I find the VRF outdoor unit capacity (similar to an RTU capacity)? Is the only way to get these to add some VRF-related output variables to the OpenStudio 2.4 reports section and then look at design day numbers in DView or ResultsViewer?

2) On the DOAS side of things, i.e. the "traditional" air loop, rather than the VRF structure, each zone has a field "Account for Dedicated Outdoor Air System", which appears to be set to "No" by default. But since VRF and DOAS do not "communicate" with each other shouldn't this be set to "Yes", so that the outdoor air coming into the zone is viewed as a (mild) load on the VRF terminal for that zone? The effect may be small if the OA is space neutral, but depending on what the various zones are doing, I could imagine OA temperature being off from zone set point by a few degrees?

3) I am doing a ASHRAE 90.102997 Appendix G analysis, requiring to compare a "System 6" (RTUs with PFPB) to a VRF & DOAS system. I always thought that the latter is much more efficient than the former, but I get much more annual cooling and heating energy for VRF & DOAS than for RTU & PFPB. Something is amiss, I am sure. Anyone seen this before?

edit retag flag offensive close merge delete



compare the VRF curves in your model to the energyplus example/default curves. I have found openstudio curves to result in unexpectedly high energy consumption.

TomB gravatar imageTomB ( 2018-03-29 15:48:18 -0600 )edit

Thank you kindly, TomB. I am looking in EnergyPlus-8.7.0\ExampleFiles\VariableRefrigerantFlow_5Zone.idf and finding things like VRFHeatCapFT, VRFHeatCapFTHi, VRFHeatEIRFT, VRFHeatEIRFTHi, HeatingEIRLowPLR,HeatingEIRHiPLR. Are these the curves I need to be moving over to OpenStudio?

Matt Koch gravatar imageMatt Koch ( 2018-04-03 16:44:19 -0600 )edit

That sounds right. I didn't do anything fancy here, I just copied the values in the curve objects (manually) from e+ examples into the appropriate edit field in my OpenStudio model.

TomB gravatar imageTomB ( 2018-04-03 17:22:04 -0600 )edit

I finally found time to graft the VRF systems from EnergyPlus-8.7.0\ExampleFiles\VariableRefrigerantFlow_5Zone.idf into a "Mitsubishi" library. However, I still get no energy advantage of the VRF/DOAS system over the RTU/VAV system. I do not think that is correct. I saw an advantage only when using the "hack" library I grafted from Daikin eQuest/Trace-700 into a "Daikin" library. Unfortunately, other than a rough comparison of a Trace-700 to an OpenStudio model, I have nothing that would prove I grafted this correctly. Is there no on out there who had gotten to the bottom of this?

Matt Koch gravatar imageMatt Koch ( 2018-08-11 21:34:01 -0600 )edit

1 Answer

Sort by ยป oldest newest most voted

answered 2018-03-30 08:08:24 -0600

updated 2018-03-30 08:33:13 -0600

1) If you are looking for the size of the outdoor unit then this information is in the eio file or table reports (html file) if that input is autosized.

  VRF Heat Pump,           !- Heat Pump Name
  autosize,                !- Gross Rated Total Cooling Capacity {W}
  autosize,                !- Gross Rated Heating Capacity {W}

! <Component Sizing Information>, Component Type, Component Name, Input Field Description, Value
Component Sizing Information, AirConditioner:VariableRefrigerantFlow, VRF HEAT PUMP, Design Size Rated Total Cooling Capacity (gross) [W], 37248.19743
Component Sizing Information, AirConditioner:VariableRefrigerantFlow, VRF HEAT PUMP, Design Size Rated Total Heating Capacity [W], 37248.19743

If you are talking about the capacity during the simulation then that information would be available using the Output:Variable object defined in your input file.

Output:Variable, *, VRF Heat Pump Total Cooling Rate, timestep;
Output:Variable, *, VRF Heat Pump Total Heating Rate, timestep;

2) The Account for Dedicated Outdoor Air System input field is in the Sizing:Zone object. Turning this input on will impact the zone loads used during sizing calculations. The zone loads used during sizing calculations primarily affect the zone air flow rate in cooling and heating mode, which in turn are used to size the HVAC system. The VRF system terminal units have a dedicated mixer object which accounts for outdoor air during operation. When the terminal unit is on the mixer provides the specified outdoor air.

3) The higher efficiency of the VRF system is attributed to lower fan power when some terminal units turn off when not needed and lower external static pressure of the non-ducted terminal units. Purported higher efficiency of the outdoor unit can also reduce energy consumption when some terminal units turn off. And of course there is the heat recovery aspect of VRF systems. Fan power savings is a given in the VRF system while higher operating efficiency at part-load is reflected in the performance curve for EIR as a function of PLR.

  VRF Heat Pump,           !- Heat Pump Name
  CoolingEIRLowPLR,        !- Cooling Energy Input Ratio Modifier Function of Low Part-Load Ratio Curve Name
  CoolingEIRHiPLR,         !- Cooling Energy Input Ratio Modifier Function of High Part-Load Ratio Curve Name
  HeatingEIRLowPLR,        !- Heating Energy Input Ratio Modifier Function of Low Part-Load Ratio Curve Name
  HeatingEIRHiPLR,         !- Heating Energy Input Ratio Modifier Function of High Part-Load Ratio Curve Name

I have tested these systems in both lab and field environments and the efficiency of those systems did not represent what was reported in the product literature. The reason for this difference is wrapped up in the control logic used by each manufacturer. As these tests were years ago maybe things have gotten better over time. A discussion of the EIRfPLR performance curve is presented in the publication here in Figure 10 on page 32.

edit flag offensive delete link more


Thank you kindly rraustad.

I found the answers to item 1) exactly as you suggested. I guess EnergyPlus, let alone OpenStudio, does not report these like it does the traditional "coil" numbers, so one has to go digging? Is there a reason?

I agree that VRF should be more efficient because of what you state in item 3). Are you saying, though, that the real life performance trails the lab performance? Is this similar to advertised mpg on cars and actual mpg obtained by various drivers? Is that why OpenStudio shows "unexpectedly high energy consumption" as TomB indicated above?

Matt Koch gravatar imageMatt Koch ( 2018-04-03 16:50:50 -0600 )edit

And, I am still confused on item 2) above. I have a separate ("traditional" OpenStudio, for lack of a better term) air loop for a DOAS system, to which I connect the same zones that then also have the VRF equipment. Do I need to turn the Account for Dedicated Outdoor Air System in the zone on the air loop screen to Yes, so that the OA from the DOAS is treated as an extra load to the VRF equipment - which it probably is when the OA is not exactly "neutral"?

Matt Koch gravatar imageMatt Koch ( 2018-04-03 16:54:16 -0600 )edit

It's unfortunate that the burden is on the user to setup the performance of VRF systems. The user knows nothing about the controls and that can impact performance a great deal. I would suggest a conservative approach to modeling VRF performance. The reports for the "coil" numbers are segregated by coil type so yes you have to search for your type of coil. If you are providing OA directly to the zone then I would use "Yes" for account for dedicated outdoor air system. Check the results to ensure coil sizing is as you expect.

rraustad gravatar imagerraustad ( 2018-04-06 09:29:44 -0600 )edit

Your Answer

Please start posting anonymously - your entry will be published after you log in or create a new account.

Add Answer


Question Tools



Asked: 2018-03-29 11:17:06 -0600

Seen: 528 times

Last updated: Mar 30 '18