Since it keeps coming up, here is a real world performance data for vertical, horizontal (slinky) and pond loop (also slinky).See attached pdf. Heat dominated climate, Western New York The vertical loop has no A/C, but year around domestic hot water. All year are over 2 years old, More details of the system in the links below. Please be aware that the scaling is different between the graphs.
http://welserver.com/WEL0448/
http://welserver.com/WEL0384/
http://welserver.com/WEL0383/

Any other data for these like what tonnage system they are attached to and house heat load and loop length of each one?

Posted By docjenser on 27 Dec 2011 01:55 AM
... here is a real world performance data for vertical, horizontal (slinky) and pond loop (also slinky) ...

Interesting to see EWT examples for a cold season climate.

Nice job producing annual EWT charts that can be used to benchmark what normal performance is, useful in case something at some point breaks with respect to the water loop.

Here are 3 suggestions for removing the occasional spikes in your data:

1. Implement a rolling 24 hour average measurement on the EWT values.  (If you already have a Treatment on EWT values, first copy EWT values to another device.  Apply treatment F and chart EWTCopy_D values.)  Now brief period outages get averaged out.

2. Implement a sample and hold function on your EWT sensor.  This way, when the temp value goes to zero, it will be ignored (since you're only sampling once every 571 minutes).

3. Put the WEL unit on a high quality sine wave UPS (I use the smallest APC SmartUPS unit available).  This will increase your cost of implementation by about $300, though.

In my case, for EWT, I do all 3 above to help ensure a real clean chart that autoranges nicely between the low and high temperatures (i.e. doesn't go to zero on the y-axis).  Here's the example:



You also might want to consider putting ambient outside temp on your EWT chart as ambient temp is the most influencincing factor on EWT.  I use a 30 day rolling average implementation (red dashed line) to ensure not being distracted by normal day-to-day variation.

Hope this helps.

Best regards,

Bill

They already have a sample and hold function. I will try your suggestion with the rolling 24h average.
The horizontal loopfield (WEL0448) is for a 2500 sqf old farmhouse with a heatloss of 58KBTU/H. 4 ton W-A. Leaky, wind sensitive. Loopfield is slinky with 4x800 ft of 0.75" pipe at 6 ft.
Pond loop (WEL0384) is has 10 ton capacity. 12 x 500ft in slinkies at 6-7 ft depth. Heatloss is 102 KBTU/H, Building is new built with 4800 sqf.Econar 10 ton W-W heatpump with hydronic air handlers for A/C.
Veritcal (0336) is a heatloss of 56 kbtu/h, 6 ton w-w heatpump. Loop is 2 boreholes, 400ft each, with 1.25" pipe. Hope this helps. Thanks