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

VRF water cooled plant temperatures far too cold

asked 2016-01-07 08:19:35 -0500

antonszilasi's avatar

updated 2016-01-07 11:29:15 -0500

Hi Everyone,

This is part two of my original question posted here:

@rraustad was kind enough to post a very helpful reply. After I read his reply I investigated my results and discovered that the unmet hours were the result of the VRF system disabling itself often throughout the year as both the heating and cooling mode limits were exceeded.

I tried to fix this issue by setting the minimum outdoor temperature in heating and cooling mode setpoints to the maximum and minmum (100 C and -100 C respectively) so that the VRF would never be disabled.

However when I did this the simulation simply crashed as the plant temperatures "were getting far too cold".

Using the base case model with a 100,000 W cooling tower and the VRF system Water Condensor Volume Flow rate of 10 gallons a minute.

I tested a range of options to try to solve this issue (and in each option I made sure to output the output variables suggested here . These options consisted of:

  • Increasing the nominal cooling tower capacity (from 100,000 W to 5275 Kw) with the VRF system Water Condensor Volume Flow rate of 10 gallons a minute .
  • Increasing the nominal cooling tower capacity (from 100,000 W to 5275 Kw) with the VRF system Water Condensor Volume Flow rate set to autosize.

With these two models the result was the same unmet hours issue due to the VRF disabling itself. The only difference was that the the VRF disabled itself a few hours later with the autosize flow rate than with the 10 gallons a minute flowrate. Furthermore in both cases the VRF disabled itself a few hours later compared to the basecase which seems to disable itself immediately (Janurary 1st at 1 am). Note that I considered the VRF to disable itself when the CONDENSER LOOP:Plant Supply Side Outlet Temperature [C] plateued as seen below.

image description

  • Increasing the Condensor Volume Flow rate from 10 gallons a minute to 50 gallons a minute (With a cooling tower capacity of 100,000 W). Again this only delayed the VRF system disabling itself by one hour. Compare to the basecase.

With the help of the designbuilder team I've submitted a support ticket to the Energy Plus team and I'm hoping we can conduct the discussion here for the benefit of everyone.

The E+ files along with the results for each simulation can be seen in the dropbox here. Does anyone have any thoughts on the issue while we wait for the E+ support team? Thank you!

edit retag flag offensive close merge delete

2 Answers

Sort by ยป oldest newest most voted

answered 2016-01-07 16:01:29 -0500

antonszilasi's avatar

updated 2016-01-07 17:34:30 -0500

Thank you! @rraustad

I've fixed this issue by:

  1. Converting the condensor loop containing the cooling tower to a plant loop.

  2. Adding a boiler to that plant loop which then heats the plant loop to keep it from falling below 15 C. This means that the plant loop will never become too cold.

The thing I'm having trouble understanding now is. Why does this VRF system with a boiler and cooling tower consume more energy than a air cooled VRF system. Furthermore why does the VRF air cooled system meet all the setpoint hours while the water cooled VRF with a cooling tower doesn't?

The idfs for the air cooled VRF, the water cooled VRF with a cooling tower and the water cooled VRF with a cooling tower and boiler can be seen here.

Also I am using the weather file Newmark internal airport New Jersey.

Any insights would be most appreciated!

edit flag offensive delete link more



Heat pumps work by removing heat from a large space (outside air, or in your case a water loop) and moving it to the indoor space. An air-cooled unit is able to absorb heat from the outside air when providing heating. Your water cooled VRF is absorbing heat from your loop, which then must be made up by the boiler. Essentially it is the boiler that is providing the heating for your building, and the VRF units are just moving that heat around throughout the building, this is why the energy is so much greater. Water-cooled units normally work best with a geo-thermal loop to reject/absorb heat to.

StefanG's avatar StefanG  ( 2016-01-07 18:29:55 -0500 )edit

Run your 2 input files, air-cooled and water-cooled, and compare cooling, heating, water pumping, tower fan, and boiler energy.

  • is water pumping energy excessive
  • does cooling and heating energy compare
  • are all other energy consumers the same
  • are the VRF performance curves accurate for air- and water-cooled equipment (check manufacturers data)
rraustad's avatar rraustad  ( 2016-01-07 18:33:31 -0500 )edit

answered 2016-01-07 08:55:02 -0500

updated 2016-01-07 09:02:17 -0500

Check plant equipment operation to see if these temperatures make sense. If the plant heating equipment is not operating, these temperatures do make sense. I do not see a boiler on your plant.

@antonszilasi, send me a private email and I'll send you an example plant with a tower and boiler. My email is

edit flag offensive delete link more


@rraustad is a VRF system required to have a boiler? (This is my first VRF simulation). When you refer to heating equipment are you refering to the Plant-Condensor control objects? There is only a PlantEquipmentOperation object for cooling load and not heating load.

antonszilasi's avatar antonszilasi  ( 2016-01-07 09:21:16 -0500 )edit

For water cooled equipment, when cooling the building the condenser rejects heat to the water loop and you need a tower to get rid of that heat. In heating mode, the condenser absorbs heat from the water and you need a boiler to replace that heat.

rraustad's avatar rraustad  ( 2016-01-07 11:03:14 -0500 )edit

Your Answer

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

Add Answer

Training Workshops

Question Tools



Asked: 2016-01-07 08:19:35 -0500

Seen: 555 times

Last updated: Jan 07 '16