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

# Modelling enclosure losses from a packaged or built-up RTU

I need to model the impact of increased enclosure insulation on the energy used by a roof top unit. I was planning making a small zone with the an exterior boundary condition wall having UA equal to that of the purposed RTU and using that space as a supply plenum. Depending on the size of unit and the location of the coils, I would also perhaps do the same with a return plenum to capture any gain/losses driven by the return air delta T.

Assuming I can come up with a reasonable UA approximation, would this work? Is there a better way.

• Appropriately modelling the inside convective heat transfer
• Accounting for the variation of air temperature as it move though the RTU / estimating the area of enclosure that is subject to the return or supply air temperatures (or even something in between like the air between the fan and heating coil on a blow through unit).

UPDATE

Applying the above approach to enclosure losses before the coils (via dummy return air plenum) fails to account for the mixed air temperature after outdoor air has been introduced. For this I'm hoping that i can use an EMS program to calculate the mixed air temperature accounting for jacket losses and then reset the air temperature at the coil inlet node to this calculated value.

• accurately calculating the losses/gains at the mixing box
• not sure if it is possible to just reset the air temperature at the coil inlet node using EMS

Using EnergyPlus, currently 9.0 but could update if necessary.

Any advice on how best to model this would be appreciated.

edit retag close merge delete

I would also be open to modelling losses external to EnergyPlus, exporting variables that would be key drivers of enclosure losses and calculating that way, but keeping it all internal to EnergyPlus would be preferably.

( 2021-01-25 13:21:54 -0600 )edit

Sort by ยป oldest newest most voted

Well, I guess I'll answer this myself, perhaps an unimaginative answer will inspire someone else to chime in with something slicker.

I ended up making a basic EMS program that takes hard coded surface areas for different modules of the RTU, then calculates and sums the UA dT (U-value also hard coded) for each module at each time step based on the applicable air temperature and puts the load into the zone begin served as "other equipment".

I use whether the fan is running or not to determine if losses are occurring.

more

realistically, how do people think the losses though for a single zone packaged roof top unit that is not running would compare to losses though the rest of the roof? i.e should I be spending my time trying to model off cycle conduction losses as well?

( 2021-02-26 19:32:00 -0600 )edit

Your approach seems reasonable as it adds load to the HVAC system albeit within the zone instead of within the RTU itself. I've never gone through the trouble of accounting for RTU or air handler wall conduction losses. What is the difference in cooling energy and other results between a model with your EMS program and a model without?

( 2021-02-28 18:56:38 -0600 )edit

I've only been working with some basic assumptions just to get everything up and running, but I'm seeing low to mid single digit percent HVAC energy savings in cold climates (nearly all savings is in heating).

( 2021-03-01 11:52:56 -0600 )edit