Trace 700 - Modeling Geothermal VRF Systems

4 posts / 0 new
Last post

Any advice on how to model a geo VRF system in Trace?
Trace seems to fall apart when you select the cooling plant for VRF. It assumes that a VRF system is an air cooled unitary type system.

I wonder, is there a workaround, so that I can use the geothermal input / output features that are availabe for geo heat pumps? These are critical in sizing the geo field.

Thanks,
Eric

ebroemel's picture
Offline
Joined: 2011-05-06
Reputation: 0

Hmmm. Tough question.

There's a few key things. A standard ground source heatpump is essentially the same as geothermal vrf except for the fluid. The key differences would be:

1) The thermal storage capacity (you'd need to enter this based on the storage of the refrigerant

2) The loop pump power (probably less for pumping refrigerant than water)

3 The heat transfer with the ground (this will be largely based on #1)

The other main difference you can note is in the library. If you look at the air cooled VRF Digitial scroll, you will note that it has some advanced unloading curves. It would be ideal to have these curves for a geothermal system but they are based on air temperatures and not ground temperatures - so you can't simply copy them.. You might be able to get the curve coefficients from energy-plus?

The other main point is to adjust the minimum loop temp - which I imagine is around 0 degrees Fahrenheit.

The bottom line is that you'd want to start with a ground source heat pump, and then make your adjustments for refrigerant vs water glycol mix. You will not be able to tie the VRF system into the ground as a workaround. Instead, tie a ground system into a vrf system.

Bob's picture
Bob
Offline
eQUEST UserLEED AP BD+CTRACE 700 User
Joined: 2010-06-30
Reputation: 475

Thanks Bob for the thought provoking reply.
It is comforting that someone as experienced as you with trace find the subject complicated as well.

I think I follow most of what you have posted.

Here is a link to technical resource papers on the LG website. The fourth document in the list is a .pdf that describes how to model VRF in Trace. The file goes through air cooled and water cooled VRF. It also refers to a custom Trace library file which I have acquired from our rep. When you read this document (particularly the water cooled section, which could be adjusted for geothermal), would that change your approach?

Let me know and I can get you the library file if you are interested.

Thanks
Eric

http://www.lg-vrf.com/technical-literature.aspx?resourcesType=White%20Papers

ebroemel's picture
Offline
Joined: 2011-05-06
Reputation: 0

Thanks for the post Eric-

I would say that confirms the approach that I mentioned (and also confirms the issue of the 3 variables I mentioned). They probably provide better unloading curves and the proper temperature limits in the library (like the minimum loop temperature) - which is much better than any other option I was aware of (As opposed to getting them from energyplus)

They don't seem to mention connecting with the geothermal loop, but that's just 1 extra step.

I can take a look at the library for you. (i'll email you privately)
I would argue that you can simply leave the system type as Variable refrigerant in either case (water cooled or air cooled). I'm guessing, for water cooled, they are just playing it safe in their instruction. However, I think the Variable refrigerant SYSTEM will do better in terms of fan power than a heat pump system, but there's a lot of dependent variables there.

Bob's picture
Bob
Offline
eQUEST UserLEED AP BD+CTRACE 700 User
Joined: 2010-06-30
Reputation: 475