Variable Primary Flow

openstudio
chiller
retag add tags

I have a two part question. I have a customer that has a variable primary flow system with two air cooled chillers. They have had control issues with the flow fluctuations caused by the opening of isolation valves during the transition from one chiller to two chillers. This is a common problem with VPF systems. I have corrected the issue via slowing down the isolation valve cycle times but based upon the literature out there another possible solution is placing their chillers in series. So this is where I get to my question(s).

1) What is the best way to model variable primary flow in OpenStudio, and: 2) Can you model two chillers in series and to VPF?

355

jmoody

asked 2016-09-01 11:19:48 -0500

4.4k

__AmirRoth__

updated 2016-09-01 14:24:58 -0500

edit flag offensive 0 remove flag close merge delete

Comments

add a comment see more comments

I have not done it in series but in parallel. But I think it is possible. You have to make sure that you have a bypass in both supply side and demand side of the chilled water plant loop. Do not use the common pipe in the plantloop object. Place setpoint managers for both chillers. Setup chillers operation mode as "leaving set point modulated". And the most difficult thing for me, which I have been looking how to set it up, this kind of setup requires to maintain a minimum flow through the loop. You can still model it with 0 minimum flow at the pump, but is not how it works in reality. If you set the primary pump with a non-zero minimum flow, you have to figure out a plantloop schedule for the chilled water loop to turn on the chiller and pumps when necessary. This is the way I would do it...VPF Air Cooled Series.JPG

812

Carlos Vazquez

answered 2016-09-02 00:16:00 -0500, updated 2016-09-02 00:20:35 -0500

edit flag offensive 0 remove flag delete link

Comments

Thanks for the reply. Really got me thinking of a few things I want to test with this. I am going to change my plant to the configuration you show in the drawing and makes some runs.

One question, what is the purpose of the adiabatic pipe on the supply side around the chillers? What does it represent and do you have to show it?

jmoody (2016-09-02 13:42:30 -0500) edit

In EnergyPlus it is a recommended practice for any loop to have a bypass in both supply and demand side. Both will be more active during your simulation when having a minimum flow setup through the loop. For a variable flow primary-only loop, the bypass in the demand side will be more active. The one in the supply side might be active in any instance that both chillers are off with the pump active or when both chillers are on and unable to satisfy the load.

Carlos Vazquez (2016-09-02 19:55:13 -0500) edit

I am exploring a VPF loop as well. I can follow all of the above, but am wondering what chiller model to use? It appears that the Chiller:Electric:EIR chillers all use data at constant chilled water and condenser water flow rates to develop their curve coefficients. Yet, in VPF the chilled water flow rate is clearly not constant. To me, this suggests that ChillerCapFTemp, ChillerEIRFTemp ChillerEIRFPLR also should depend on flow rate, not just temperature?