|
|
|
Measuring COP of geothermal heat pump
Last Post 26 Feb 2012 01:43 PM by a0128958. 16 Replies. |
Sort:
|
|
 Prev Next |
You are not authorized to post a reply. |
|
dlloyd1
New Member
Posts:4
 |
| 14 Jan 2012 03:37 PM |
|
I would appreciate comments on the following:
Current measurements of COp require an ammeter, voltmeter, flow rate and temperature. It is clunky and only provides one number at that moment.
What is needed is a continuous display of COP to monitor performance. Wikipedia describes the COP another way as the theoretical max by measuring refrigerant temperatures. T (cold) entering the evaporative heat exchanger and T (hot) entering the condensing heat exchanger. Change the F to Kelvin and:
T (hot)
COP =—————————
T (hot)—T (cold)
If T (hot) = 165 F ( 347K) and T (cold)= 28F (271K), this gives a COP of 4.5
Why cannot this be a way to get a continuous COP? The only problem is how to get those two Ts. This could be a simple monitor on the unit or even on the computer. Your thoughts, experts. |
|
|
|
|
|
|
dlloyd1
New Member
Posts:4
|
| 14 Jan 2012 03:39 PM |
|
Posted By dlloyd1 on 14 Jan 2012 03:37 PM
I would appreciate comments on the following:
Current measurements of COp require an ammeter, voltmeter, flow rate and temperature. It is clunky and only provides one number at that moment.
What is needed is a continuous display of COP to monitor performance. Wikipedia describes the COP another way as the theoretical max by measuring refrigerant temperatures. T (cold) entering the evaporative heat exchanger and T (hot) entering the condensing heat exchanger. Change the F to Kelvin and:
COP =T (hot) divided by [T (hot)—T (cold)]
If T (hot) = 165 F ( 347K) and T (cold)= 28F (271K), this gives a COP of 4.5
Why cannot this be a way to get a continuous COP? The only problem is how to get those two Ts. This could be a simple monitor on the unit or even on the computer. Your thoughts, experts.
|
|
|
|
|
Bill Neukranz
Veteran Member
Posts:1103
|
| 14 Jan 2012 07:01 PM |
|
Posted By dlloyd1 on 14 Jan 2012 03:37 PM
I would appreciate comments on the following: Current measurements of COp require an ammeter, voltmeter, flow rate and temperature. It is clunky and only provides one number at that moment. What is needed is a continuous display of COP to monitor performance. Wikipedia describes the COP another way as the theoretical max by measuring refrigerant temperatures. T (cold) entering the evaporative heat exchanger and T (hot) entering the condensing heat exchanger. Change the F to Kelvin and: T (hot) COP =————————— T (hot)—T (cold) If T (hot) = 165 F ( 347K) and T (cold)= 28F (271K), this gives a COP of 4.5 Why cannot this be a way to get a continuous COP? The only problem is how to get those two Ts. This could be a simple monitor on the unit or even on the computer. Your thoughts, experts. Having good familiarity with COP, and having experience measuring it real time and for historical analysis, and with no knowledge of measuring COP via refrigereant teamperatures, I looked in Wikipedea for the information you cited above. Unfortunately, I couldn't find it. I would appreciate a link please - I'd like to go read this. Many thanks! WaterFurnace, and perhaps other major name producers, uses this formula to compute COP: ((EWT - LWT)*GPM*500) + P*3.416) / P*3.416, where EWT and LWT are in °F, the fluid is water (affects the 500 constant otherwise), and P is in kW and does not include pumping power. WF advertises a 5.0 COP for its Envision series, measured at 68° or 70° EAT (can't tell which), and 400 CFM/ton. Perfect measurement of in-the-field COP requires use of an RMS power meter, installation of a flow sensor into the loop (along with finishing up with a loop purge), and installation of temp sensors on the EWT and LWT loop lines. Obviously the flow sensor installation is too much work for one time in-the-field measurement, so, some reasonable compromise is in order. BTW, it's tempting to substitute voltage and current measurement for RMS power measurement, but, there's enough variability with PF such that I wouldn't advise doing so. An instrument to accurately measure RMS power is not significantly more expensive than one that separately measures voltage and current. Where compromise is really needed is with the flow sensor. It's reasonable to assume that at least during the period of measurements, dynamic, steady state loop water pressures are constant. Thus, it's reasonable to manually measure Entering Water Pressure, and Leaving Water Pressure (via PT ports on the loop pipe right at the unit), subtract the two, and look up the flow rate from the pressure delta via charts supplied by the manufacturer. Indeed, for example, WF publishes delta pressure vs flow rate charts for this very purpose. So, with a reasonably accurate RMS power meter, and a manually calculated flow rate, now an in-the-field COP measurement can be reasonably accurate without going to significant expense. I have 2 WF Envision GSHPs, and I measure COP, real time, in the above manner, with my monitoring system that I installed. It shows real time COP on the first page here: http://www.welserver.com/WEL0043/ . For my 3 ton unit, for example today, steady state COP is running at about 4.8, in first stage. (For COP, it's not difficult for me to obtain near advertised values since I'm in a warm season climate - i.e. my EWT is currently running at about 66°). Real time, instantaneous COP measurement can be interesting, but, there's a lot of variability in the measured numbers, particularly power due to compressor warm up time. Thus, a better COP 'health' indicator can be obtained when it's looked at statistically over an entire month period. Here's an example:  Here we see that for this particular month, for this measurement period, statistically COP is about 4.8. Hope this helps. Again, I'm not aware of a technique to obtain COP simply via sole measurement of refrigerant line temperatures. Best regards, Bill |
|
Energy reduction & monitoring</br>
American Energy Efficiencies, Inc - Dallas, TX <A
href="http://www.americaneei.com">
(www.americaneei.com)</A></br>
Example monitoring system: <A href="http://www.welserver.com/WEL0043"> www.welserver.com/WEL0043</A>
|
|
|
dlloyd1
New Member
Posts:4
|
| 15 Jan 2012 12:25 PM |
|
Here is the website address at Wikipedia as requested:
http://en.wikipedia.org/wiki/Coefficient_of_performance |
|
|
|
|
jonr
Senior Member
Posts:5341
|
| 15 Jan 2012 05:20 PM |
|
Q is heat, not temperature and a HP never works at maximum theoretical efficiency. So you can't use the formulas. |
|
|
|
|
Bill Neukranz
Veteran Member
Posts:1103
|
| 15 Jan 2012 06:14 PM |
|
Posted By jonr on 15 Jan 2012 05:20 PM
Q is heat, not temperature and a HP never works at maximum theoretical efficiency. So you can't use the formulas. My conclusion too. Best regards, Bill |
|
Energy reduction & monitoring</br>
American Energy Efficiencies, Inc - Dallas, TX <A
href="http://www.americaneei.com">
(www.americaneei.com)</A></br>
Example monitoring system: <A href="http://www.welserver.com/WEL0043"> www.welserver.com/WEL0043</A>
|
|
|
docjenser
Veteran Member
Posts:1400
|
| 16 Jan 2012 01:22 AM |
|
Voltage should be rather constant, flow also. Temp sensors measure the delta T for the source, and amp sensor uses the different draw of the HP during different working conditions. Fluids containing antifreeze should be calculated with 485 instead of 500 (for water without antifreeze).
Instead of voltage and amperage you can use a watts meter. On one of the WELs we compare the wattsmeter and its accuracy to the amp sensor multiplies with a constant voltage measured once.
http://welserver.com/WEL0396/
As you see the accuracy of the 2 is within 1.5%, $200 for the wattsmeter, or $35 for the current sensor.
The only other thing you need is the power factor of your compressor. |
|
| www.buffalogeothermalheating.com |
|
|
dlloyd1
New Member
Posts:4
|
| 16 Jan 2012 07:53 AM |
|
I know that using refrigerant temperatures is a radical approach but just hang with me for a while. Assume for the moment that it will work. Then find a way to measure refrigerant temperatures. The Wiki system should give a lower and more accurate reading of COP than the conventional approach because there are several losses not included in the standard since it is measured early in the cycle. The heat exchanger is a slight inefficiency. The compressor must also circulate the refrigerant, there is friction that is a heat loss, there is a slight noise that is an acoustic loss. These are all included in the Wiki approach.
Why not find a way to measure the refrigerant Ts and do a test using both approaches? |
|
|
|
|
jonr
Senior Member
Posts:5341
|
| 16 Jan 2012 08:03 AM |
|
I think trying it would be educational. |
|
|
|
|
joe.ami
Veteran Member
Posts:4377
|
| 16 Jan 2012 09:35 AM |
|
IMHO COP is a number embraced by salesman.
While there is nothing wrong with it a true average COP would be measured over years since system was designed (most commonly) by a model that uses a 20 year weather average. Peak COP is for the marketing dept.
My customers are simply interested in how much they will save V propane. |
|
Joe Hardin
www.amicontracting.com
We Dig Comfort!
www.doityourselfgeothermal.com
Dig Your Own Comfort! |
|
|
docjenser
Veteran Member
Posts:1400
|
| 18 Jan 2012 03:22 AM |
|
Posted By joe.ami on 16 Jan 2012 09:35 AM
IMHO COP is a number embraced by salesman.
While there is nothing wrong with it a true average COP would be measured over years since system was designed (most commonly) by a model that uses a 20 year weather average. Peak COP is for the marketing dept.
My customers are simply interested in how much they will save V propane.
I agree. We only monitor COP on one system, the rest we do not bother. 1st, 2nd ,3rd stage run time, KW usage, operational dollars spend matters much more in the real world. |
|
| www.buffalogeothermalheating.com |
|
|
Bill Neukranz
Veteran Member
Posts:1103
|
| 18 Jan 2012 10:16 AM |
|
Posted By docjenser on 18 Jan 2012 03:22 AM
... operational dollars spend ... Docjenser, I'm looking at adding this for a couple of WEL systems that I support. I'm curious how you're doing this for your systems: 1. Assuming an average $/kWh rate constant (i.e. using a 'device slot' in the WEL) and leaving it as such without future change? 2. Assuming an average $/kWh constant and updating it once per year? 3. Assuming an average $/kWh constant and updating it monthly? 4. Some technique where the WEL reads an input for the $/kWh rate? 5. Something else? At your convenience, would appreciate some comment. Many thanks! Best regards, Bill |
|
Energy reduction & monitoring</br>
American Energy Efficiencies, Inc - Dallas, TX <A
href="http://www.americaneei.com">
(www.americaneei.com)</A></br>
Example monitoring system: <A href="http://www.welserver.com/WEL0043"> www.welserver.com/WEL0043</A>
|
|
|
Traloch
New Member
Posts:1
|
| 25 Feb 2012 07:51 AM |
|
Hi, I'm measuring COP of air to water heat pump for college project. Can someone tell me what I need to change in the above formula if i'm using Litre/min for flow and degrees Celsius for temperature? What are the constants 500 and 3416? |
|
|
|
|
Looby
Basic Member
Posts:401
|
| 25 Feb 2012 08:19 PM |
|
Posted By Traloch on 25 Feb 2012 07:51 AM
What are the constants 500 and 3416?
Nope, I won't do your homework for you, but here are two
crumbs that might help get you started on the right path:
500 converts flow rate from GPM to pounds per hour, that is:
1 gallon per minute = 500 pounds per hour (of pure water)
3146 converts power units from kW to BTU/hr, that is:
1 kilowatt = 3416 BTU per hour (?? or maybe 3412 ??)
...good luck,
Looby |
|
| One measurement is worth a thousand expert opinions. |
|
|
engineer
Veteran Member
Posts:2749
|
| 25 Feb 2012 09:34 PM |
|
Is Mr Lloyd confusing Carnot (max theoretical) efficiency for a given pair of temperatures with actual real world COP? |
|
Curt Kinder <br><br>
The truth is incontrovertible. Malice may attack it, ignorance may deride it, but in the end, there it is - Winston Churchill <br><br><a href="http://www.greenersolutionsair.com">www.greenersolutionsair.com</a>
|
|
|
docjenser
Veteran Member
Posts:1400
|
| 26 Feb 2012 01:21 PM |
|
Posted By a0128958 on 18 Jan 2012 10:16 AM
Posted By docjenser on 18 Jan 2012 03:22 AM
... operational dollars spend ... Docjenser, I'm looking at adding this for a couple of WEL systems that I support. I'm curious how you're doing this for your systems:
1. Assuming an average $/kWh rate constant (i.e. using a 'device slot' in the WEL) and leaving it as such without future change?
2. Assuming an average $/kWh constant and updating it once per year?
3. Assuming an average $/kWh constant and updating it monthly?
4. Some technique where the WEL reads an input for the $/kWh rate?
5. Something else?
At your convenience, would appreciate some comment. Many thanks!
Best regards,
Bill
Sorry, I just saw this. We are using a constant, namely 13.5 cents/kwh, which is the pretty constant rate here over the last 2 years. Power prices fluctuate daily, or even within a day, and you can go crazy over this, but over the year it pretty much evens out. That is why we also display the KWH used, and everyone can do his/her own math. I find it more important to calibrate your sensors and equipment, at 5F delta T, 1 degree off means 20% off in the calcs. |
|
| www.buffalogeothermalheating.com |
|
|
Bill Neukranz
Veteran Member
Posts:1103
|
| 26 Feb 2012 01:43 PM |
|
Sorry, I just saw this. We are using a constant, namely 13.5 cents/kwh, which is the pretty constant rate here over the last 2 years. Power prices fluctuate daily, or even within a day, and you can go crazy over this, but over the year it pretty much evens out. That is why we also display the KWH used, and everyone can do his/her own math. I find it more important to calibrate your sensors and equipment, at 5F delta T, 1 degree off means 20% off in the calcs. Thanks. I decided to proceed with using a 12 month avg, updated on an annual basis. For my area, last year's avg kWh rate for the year (2011) was 9.9 ¢/kWh (not including benefits of solar PV). I'm using this rate throughout this year and will update it at the end of the year. I also put an historical 6 year rate chart on my example site ( WEL0043). I agree on your calibration comment. In the old days, in fact, I found the DS1820 sensors drifted over time. Once the S version became available, and now particularly the B version, I find very little drift over time, and very little calibration error. After a calibration of the water loop sensors in particular, I check the calibration of the DeltaT, using 2 temp propes for my Fluke meter to directly measure DeltaT. With the 'B' sensors now, I'm only having to adjust calibration (in the WEL) occasionally. Thanks for taking a moment to comment. Best regards, Bill |
|
Energy reduction & monitoring</br>
American Energy Efficiencies, Inc - Dallas, TX <A
href="http://www.americaneei.com">
(www.americaneei.com)</A></br>
Example monitoring system: <A href="http://www.welserver.com/WEL0043"> www.welserver.com/WEL0043</A>
|
|
|
| You are not authorized to post a reply. |
|
Active Forums 4.1
 |
Membership: |
 |
Latest:
croccohvacusa |
 |
New Today:
0 |
 |
New Yesterday:
0 |
 |
Overall:
35027 |
 |
People Online: |
 |
Visitors:
159 |
 |
Members:
0 |
 |
Total:
159 |
|
|
|