Undo Button

8 posts / 0 new
Last post

So wouldn't that be the greatest function ever? Well, the best lesson learned is read the Manuals since I find that what I once glanced over, actually had the key words that told me I was doing it all wrong. For instance creating separate heat pump plants and linking them together with the same geoloop. Major errors, cannot be done. Well after rereading I created the 1st heat pump (opt-vent units plant coil at System level) as the 1st plant and put in the capacity, then created the GSHP's to all the sleeping quarters and left that open capacity with the coil in the rooms, then tied the two together in Parallel-Decoupled. So the 1st HP did what it could do up to the capacity specified, then the 2nd kicked in to handle the remaining load. No errors. The greatest lesson-learned through this mental process was backing up to where I was before I mucked it up. So I got in the habit of saving to a "testing" file folder and run it, if it looked like I made forward progress, I'd save it back over the original, if not I made a note in a Readme.txt file to refer to later and went back to the original to try the next 'tweak'. Also, check out the Scan for Errors to get clues as to what the program has interpreted your input as.

Be Sustainable -- Never let today use up tomorrow!.

Bobba_Fett's picture
Offline
eQUEST UserLEED EA Credit 1 TrainingTRACE 700 User
Joined: 2010-10-01
Reputation: 229

HI Bobba_Fet,


 


    So you created one plant with more than 1 heat pump in it?  I have had to model several projects with GSHP and have been told numerous times when I call CDS for help that I could only have 1 plant and to creat 1 HP that was a weighted average of all of the ones actually being used.  This has driven me nuts every time I have done it as I did not see why I couldn't have one plant with numerous pieces of equipment in it that all share the same loop. 


   It sounds like that is exactly what you did, am I correct?

calipidgious's picture
Offline
TRACE 700 User
Joined: 2011-03-17
Reputation: 0

Cal, I've been told the same thing. So, I tend to do a quick and dirty overall KW/Ton based on selections. I typically try and get an overall EER and COP of all the heatpumps from the HP vendor when they provide heat pump selections, as most of their selection software will do this (or the rep will by hand, etc.). Obviously, this kw/t is only for the heatpumps themselves not ancillary equipment.


Before I got to this point, I also had been creating seperate plants for each size heatpump connected to the geoloop to be able to model each HP efficency. I've done it both ways and there is a difference, but I've not done enough testing to figure out why...if it's a smaller wellfield and pump due to trace not modeling correctly or the more granular efficencies.


It doesn't matter to me really anymore...I model by modleing software directions so it's not an issue during later review. Energy modeling is like predicting weather, only we have worse tools to do our guessing...

GoogleNerd's picture
Offline
Joined: 2011-04-20
Reputation: 2

Hey GoogleNerd,

    I do actually go through the process of creating a weighted average which is not generally a problem but it does get tedious when you have 30 or so heat pumps.  The thing that frosts my flakes is that not all of the heat pumps are necessarily going to have the same features/options, etc.  Its either all or none in the model so we take more credit for energy efficiency or less than we deserve.  I don't see why I can't create one plant with 100 Heat pumps in the the same plant.  I don't see why they can't share the same loop.  It just annoys me more than anything.

   I honestly have yet to find a piece of software that doesn't, well, suck in some way.  I guess its the lesser of annoying evils in the long run.

calipidgious's picture
Offline
TRACE 700 User
Joined: 2011-03-17
Reputation: 0

Well, if we are speaking of flaky...try developing your own unloading curves, or fan curves.

Or, how do you model a two stage or two compressor heat pump in TT? CDS doesn't have a clue (or at least one they will share) and manufacturers look at me funny...so that leaves the guessing to me...hmmmm. Storm front approaching...

GoogleNerd's picture
Offline
Joined: 2011-04-20
Reputation: 2

You all have a nice convo going here. And I agree, heat pumps are one of the worst things to model, really in any software. I think energy-plus has a way to model all the heat pumps individually. However, the front end stinks to eplus. Anyway, I have done some experimenting in TRACE with the same issues you all are encountering.

  • So, why can't you use 100 heat pumps in one plant? Well, there's nothing in Trace that stops you from doing this. However, Trace will have zero control of which heat pump corresponds to which room. So you will essentially end up with 50 heat pumps on, and 50 heat pumps off at half load. Therein lies the problem. It can't put all 100 heat pumps on at part load. Trace simply takes the hourly load from the system and dumps it off to the plant, so the plant has no idea which heat pump coordinates to which room/zone.
  • Does the weighted average method work? This might save you some time. I've actually created multiple plants and multiple systems to try to model these heat pumps separately. At the time, TRACE actually shared the geothermal well among all plants, so I used gshp's (I am not sure if it still works this way). Anyway, surprisingly, once you get passed 5 heat pumps or so, the statistical variance among the multiple plants makes it very close to a plant that has a weighted average. At 10 or more, the results are basically identical -Provided that the same unloading curves were used for the weighted average plant, and the multiple plants.
  • How do I model a two stage compressor? There are two ways to do this. You could really call it two pieces of equipment (in the same plant), since each piece of equipment really represents one compressor (and the heat rejection is based on kw/ton anyway). Then, set the sequencing to however the two compressors sequence in reality (I'd guess parallel).
  • 2nd way to model multiple compressors: make a list of tons and corresponding watts. At a certain tonnage, the second compressor will kick on and the wattage should reflect that. Then, you can make an unloading curve for power consumed versus tonnage. Don't even worry about making another ambient relief curve, as you can simply use one from a piece of equipment you copied.

 

I hope that makes sense. The other thing you don't want to forget to do is to copy a gshp, and set the minimum operating temperature to say 20 degrees (if you are using glycol). That way, the loop temperature can get way more capacity out of the ground, as is reality.

Bob's picture
Bob
Offline
eQUEST UserLEED AP BD+CTRACE 700 User
Joined: 2010-06-30
Reputation: 275

Hi Bob,

  Thanks for the information!  Its helpful to know and good points to keep in mind for ways to model heat pumps effectively.  I find that they are often in the models we use and I find them a real pain but they are effective!

  BTW,  I posted some today (one is above from anonymous responding to google) directly from email and didn't realize I wasn't signed in so I have some unclaimed posts, sorry [fixed this for you - Bob_f]

calipidgious's picture
Offline
TRACE 700 User
Joined: 2011-03-17
Reputation: 0

Interesting with the two stage or compressor heat pump modeling. I never thought of setting them up like parallel chillers...

I guess if you have a mixture of single stage and two stage HPs you would weight the first and second stages...say 60% (just a guess) for the total tonnage on the first stages plus all of the single stage units...then add in the other 40% (again, guess). So for a 100T system the first 60T is the part load efficiency the next 40% full load........very interesting!

Thanks for the brain food...

GoogleNerd's picture
Offline
Joined: 2011-04-20
Reputation: 2