|
|
|
Most Economical Thermostat Set Back for a GSHP?
Last Post 09 Dec 2008 01:07 PM by a0128958. 18 Replies. |
Sort:
|
|
 Prev Next |
You are not authorized to post a reply. |
|
Bill Neukranz
Veteran Member
Posts:1103
 |
| 05 Dec 2008 11:46 PM |
|
The question, what is the most economical thermostat set back for a GSHP, has got to be one of the most common questions I see. I've wondered about this myself. To answer TechGromit's question posed on 12/2 ("What Do You Keep Your Thermostat Set At?"), for the heating season, I use a 3° F set back for all rooms at night, and a 3° set back for bedrooms during the day. I.e., 68° is the non-set back tstat setting, and 65° is the set back setting. But I don't have fact based support to affirm this is the most economical tstat setting. It's just my guess that allowing the house structure and its contents to mildly cool 3° can be reasonably overcome later with a small amount of heating, and thus, I think I'm saving money using a set back. But in reality, I really don't know. I've been thinking about this since TechGromit posed the question. And I think I've come up with a quantitative way to figure out optimum set back from an economics point of view. I thought I'd share here what I'm thinking / doing, to encourage comment on what I might be missing, or doing right. Since I have this handy dandy WEL instrumentation system ( http://www.welserver.com ), I've got plenty of data to consider (data for my specific system is at http://welserver.com/WEL0043 ). I clearly and accurately know what my GSHP's power consumption is, real time and accumulated. Multiplying the GSHP's KWH consumption by the electric utility rate ($0.1264/KWH) fundamentally gives me cost of running my GSHP. For example, at the moment, today, 12/5, my GSHP system has consumed 27.8 KWH so far. Thus it has cost me $3.52 today to provide heat to my home. I could try different set backs, and look to see how the cost changes, but, obviously, weather has a huge influence. What I need to do is figure out a way to rationalize weather changes. I think I can do this for outside temperature. I clearly and accurately know what my Heating Degree Days are, real time and accumulated. For example, at the moment, here in Dallas where it's cold right now, there's been 20.6 HDD so far. I.e., the average temperature has been 44.4°. If it's a warmer day, the HDD number is lower, and visa versa. What I've decided to do, to rationalize out weather temperature influence, is to divide HVAC cost by HDD. For example, for today so far, this index is 0.170 $/HDD. In other words, for every degree the average temp today has been colder than 65° (the 'zero' basis for HDD calculation) , it is costing me 17.0 cents to heat my home. (I believe temperature is not the only weather related influencer - the other is wind speed. Unfortunately though, I don't have any data on wind speed, so for now, I'm going to ignore this. I'll see if the data holds up without taking wind speed into account.) So at the moment, using a 3° tstat set back, I'm averaging today $0.170/HDD. My plan is to watch the data and see if this index holds. If it does, I'll continue with a 3° set back for a week. Then I'll change to a 4° or 2° set back, for another week. Assuming a relationship continues, it should be possible to reasonably determine the most economical set back (or prove that it's most economical to leave the tstat alone with no set back). Below is the chart I've set up to track the index (it's also at http://welserver.com/WEL0043/SetbackEfficiency.gif on a real time basis) - it's only a few hours old, but I'll keep lengthening the time span as I get comfortable with the right amount of filtering/smoothing. The chart is already showing a trend at a constant $0.170/HDD as cost and HDD are continually increasing for the day. So maybe I'm on to something! I'll periodically update this thread with my findings. Please feel free to comment on where you think this thinking is good / flawed. We can all learn from each other. Best regards, Bill |
 Attachment: SetbackEfficiency copy.jpg
|
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>
|
|
|
|
|
engineer
Veteran Member
Posts:2749
|
| 06 Dec 2008 08:47 AM |
|
Good plan.
The two competing influences at work in consideration of setbacks are:
1) Decreased heat required by structure at lower interior temperatures.
2) increased cost of operation (lower COP) during recovery - this shows up in at least three ways which I'll describe in increasing order of subtlety:
a) Operating heat strips during recovery sucks the COP down so much that it overwhelms any setback. Almost everyone here who can see lightning, hear thunder, and fog a mirror understands this.
b) Reduced COP of 2nd stage operation - if recovery from setback causes stage 2 operation which would not have occurred had the setback not been made, there is still a definite penalty. Y2 COP is less than Y1 COP for a variety of reasons mostly having to do with flow rates through system components. The lower COP is detailed in manufacturer specs. Whether that is balanced by the lower total Btu requirement provided by the setback
c) Lower COP from extremely long on-cycles - This may happen in closed loop systems. If setback recovery algorithm is smart enough to start early and stay in low stage, that avoids losses from a)+b) above. However, it may be that a several-hours-long continuous on cycle might depress EWT to lower temps than would have occurred during normal intermittent operation since field doesn't get the chance to recover during off cycle.
c1) There is a further subtlety in c) - the unit is more efficient during continuous operation rather than multiple short cycles. This has to do with refrigerant pressures equalizing during off cycle and then the compressor having to rebuild operating Delta-P during the first minute or so of on cycle. This gain would tend to counteract loss of COP from temporarily depressed EWT.
Bill - I think your instrumentation would allow you to quantify much of the above. You have degree day data, although that is flawed both by wind issues but also that I think HDDs are merely calculated from averaging the day's high and low extremes. Better would be to use hourly data from a nearby airport. Software geared toward commercial system design uses hourly data and averages.
You have kwh, water flow, EWT and LWT. Taken together you can totalize heat provided as well as calculate your actual COP under various scenarios of setback - keep us advised!
|
|
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>
|
|
|
Ona
Basic Member
Posts:189
|
| 06 Dec 2008 01:03 PM |
|
This is exactly what I am trying to figure out in a very low tech way compared to you. I wasn't smart/lucky enough to get the welserver system (I still actually don't know exactly how/who installs this). What I did do is pay to have some cheap cumulative sub meters installed. I have 3:
1) Geo system
2) Heat strip & blower combo
3) Hot water heater (for later analysis on how much of my hot water is supplied by DSH)
I have been trying to track these on a daily basis compared to HDD's (see graph below). Unfortunately, for the first six weeks of operation my systems dip switches were in the improper position which did not allow stage 2 or backup heat to power up. This gives some inaccurate information, but at the same time I believe it gives a bit of insight as to how it affected my KWH usage when it ran longer at only Stage 1. Running longer only at Stage 1 made my daily KWH usage lower (but would not keep me warm when the temp got below 25F).
So, I was wondering if there may be a different relationship above a certain HDD number. For example, I may use 0.85 KWH/HDD when the HDD is below 30, but above 30 I may use 1.05 KWH/HDD?
My one suggestion would be to put it in the form of KWhrs/HDD rather than $/HDD because the cost of energy is variable.
I'm using this to track my payback period. I took my last 4 years worth of fuel oil use and HDD's and came up with a rate of approximately 0.15 gallons/HDD. Using the HDD as the link, I compare how much I would have spent using oil and how much I do spend on electric.
I think what you are doing looks great and it would be nice to hear about what you find. I know that my daily human written data is not as complete or informative, but you might want to compare some parts.
Also, I don't know how to paste something direct into this post, I had to give my spreadsheet as an attachment link, can you please let me know if you are able to open this?
|
Attachment: HDD.pdf
|
|
|
|
Masoud
Basic Member
Posts:180
|
| 06 Dec 2008 01:51 PM |
|
Ona;
Yes, the pdf file opened fine. Why electric coil use is not zero before November 24, when dip switches were changed to make aux heat available on your system?
How do you measure btus generated by DSH?
Regards, Masoud |
|
|
|
|
Ona
Basic Member
Posts:189
|
| 06 Dec 2008 02:33 PM |
|
Masoud ~
My spreadsheet is not completely clear, I apologize. My electric coil backup power use also includes the blower. So, the majority of that value so far (and hopefully in the future) is just the electricity draw of the blower. That is why it is not zero for the first 6 weeks.
Regarding the DSH. My chart only shows the electricity draw for my hot water tank. It does not actually read anything from the DSH. The reason that I am tracking the hot water tank is because I plan to track the use when my geo system is not running vs when it is. Also heating season vs cooling season. I hope to get a clear idea of how much of my hot water the DSH is actually providing. This made sense to me when I planned it, hopefully there isn't some flaw in my logic? |
|
|
|
|
TechGromit
Advanced Member
Posts:634
|
| 06 Dec 2008 09:01 PM |
|
Posted By a0128958 on 12/05/2008 11:46 PM
I clearly and accurately know what my GSHP's power consumption is, real time and accumulated. Multiplying the GSHP's KWH consumption by the electric utility rate ($0.1264/KWH) fundamentally gives me cost of running my GSHP.
Not to make things any more complicated than they already are, you have to remember most utilities charge different rates for different times of the day. So electricity used during prime time during the day is going to be more expensive then off peak electricity usage. This is why people who have solar panels do so well, they are producing power and selling it to the power company during peak hours, and using it during off peak hours. In order to get a true costs, I think you going to have to consider the electricity usage rates at different times of the day. My guess would be any setbacks performed during the day are going to offer a much larger savings than any setback at night. Also, if you have a Aux/Backup heat strip, disable it, at least until your tests are complete. Even better yet, compare the saving (or costs) both with and without the aux heat strips. I tried last year to play around with setback, (without getting to scientific thru) but i didn't really see any significant savings with or without setback. (And the old system didn't have heat strips). |
|
|
|
|
Brock
Advanced Member
Posts:599
|
| 06 Dec 2008 09:43 PM |
|
Posted By TechGromit on 12/06/2008 9:01 PM
Not to make things any more complicated than they already are, you have to remember most utilities charge different rates for different times of the day. So electricity used during prime time during the day is going to be more expensive then off peak electricity usage. This is why people who have solar panels do so well, they are producing power and selling it to the power company during peak hours, and using it during off peak hours. In order to get a true costs, I think you going to have to consider the electricity usage rates at different times of the day. Yes, yes and yes. Since we are on time of use I have never run our geo on peak (expensive rates). Our huge advantage is storing heat in our pool. I can run the geo system off peak only to heat up the pool and circulate the warm pool water through the 9 pex zones 24x7 to take that heat back out. Our heat pump runs close to 6 hours a day to maintain the pool/hot tub. Adding all the zones it runs about 11 hours a day, still all off peak. Our furnace ran for the first time last week, it was 5F and 20 mph winds out of the north. Our initial calculations had the furnace running at about 20F with the geo running 15 hours a day so I can't complain. Now I have to figure out a way to get even more heat from the pool, maybe run it through a coil in the furnace... |
|
| Green Bay, WI. - 4 ton horizontal goethermal, 16k gallon indoor pool, 3kw solar PV setup, 2 ton air to air HP, 3400 sq ft |
|
|
Bill Neukranz
Veteran Member
Posts:1103
|
| 06 Dec 2008 10:33 PM |
|
OK, these are some excellent comments - exactly what I was hoping for.
I've been out most of the day, so I won't have an opportunity to visit specifically on a number of good points until tomorrow.
Meanwhile, noting the comment about heating cost versus heating KWH, I've changed the index to Heating KWH per HDD. This removes the variability of electric rates clouding the analysis (no such thing as a peak residential rate, BTW, here in my suburb of Dallas).
So instead of the old index, which was sitting at $0.19/HDD at the moment, it's now 1.4 KWH/HDD.
I've changed tThe chart I'm using to record the index to KWH/HDD. It's still located in the same place: http://welserver.com/WEL0043/SetbackEfficiency.gif .
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>
|
|
|
Masoud
Basic Member
Posts:180
|
| 07 Dec 2008 11:57 AM |
|
Hello Bill,
94.0 KWH for yesterday is too high, and it does not fit with the calcs.
Regards, Masoud |
|
|
|
|
Bill Neukranz
Veteran Member
Posts:1103
|
| 07 Dec 2008 11:53 PM |
|
Masaod, thanks for catching this. Yes, I quoted the total KWH for the house (including HVAC, pool, misc, lighting, etc.), which was 94.0 KWH, and not the number for HVAC.
Since it's a little bit of work for me to go find the exact correct HVAC number at the time I wrote the above, I just deleted the paragraph.
Today's looking to be again an index near 1.25 KWH/HDD in spite of it being a much warmer day. Looks like this KWH/HDD index may hold up well for experimentation with tstat set backs.
Specifically at this moment at night:
HVAC Index = 1.29 KWH/HDD
HVAC energy today = 31.0 KWH
HDD today = 14.2 DD (actuall also had 0.1 CDD today too)
Tstat set back continues to remain at 3°
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>
|
|
|
Bill Neukranz
Veteran Member
Posts:1103
|
| 08 Dec 2008 01:13 PM |
|
Posted By engineer on 12/06/2008 8:47 AM
... Operating heat strips during recovery sucks the COP down so much that it overwhelms any setback.
... Reduced COP of 2nd stage operation
... However, it may be that a several-hours-long continuous on cycle might depress EWT to lower temps than would have occurred during normal intermittent operation since field doesn't get the chance to recover during off cycle.
... I think HDDs are merely calculated from averaging the day's high and low extremes. Better would be to use hourly data from a nearby airport. Software geared toward commercial system design uses hourly data and averages.
... COP under various scenarios of setback
engineer, much appreciate the comprehensive note. I don't have heat strips installed, so there won't be any influence here on my planned set back experiment. I wonder, though, if this will make my experiment less useful to others - i.e., only to those who also don't have heat strips. I'll footnote the fact that I don't have heat strips. I've already figured out via experimentation that I avoid going into 2nd stage with a 3° set back, or less. 4° or more causes some 2nd stage operation in the morning as part of the 64° to 68° warmup time period, starting at 5:30 AM. A several hour long continuous cycle significantly lowers my EWT - see another thread I started today on this subject. I'm calculating HDD on a near real time basis (6 second samples), relative to 65°. So for those days that outside temperature exhibits very unusual, very non-linear behavior, my HDD will be more accurate than the high and low average method. My WEL instrumentation system gives me outside temperature so I don't have to rely on weather data. I believe COP is a real time measurement. 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>
|
|
|
Bill Neukranz
Veteran Member
Posts:1103
|
| 08 Dec 2008 01:35 PM |
|
Posted By Ona on 12/06/2008 1:03 PM
... I wasn't smart/lucky enough to get the welserver system (I still actually don't know exactly how/who installs this).
... So, I was wondering if there may be a different relationship above a certain HDD number. For example, I may use 0.85 KWH/HDD when the HDD is below 30, but above 30 I may use 1.05 KWH/HDD?
... My one suggestion would be to put it in the form of KWhrs/HDD rather than $/HDD because the cost of energy is variable.
... I don't know how to paste something direct into this post, I had to give my spreadsheet as an attachment link, can you please let me know if you are able to open this?
A WEL unit is designed to be self installed. It's not difficult. See http://www.welserver.com for details. I'm hoping to show that at least within a reasonable 'band,' there's a reasonable relationship between KWH consumed and HDD. We'll see how the experiement goes. Thanks for the suggestion to compute KWH/HDD versus $/HDD. I've already made the change. 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>
|
|
|
Bill Neukranz
Veteran Member
Posts:1103
|
| 08 Dec 2008 01:55 PM |
|
Posted By TechGromit on 12/06/2008 9:01 PM
... most utilities charge different rates for different times of the day.
... if you have a Aux/Backup heat strip, disable it ...
... i didn't really see any significant savings with or without setback.
TechGromit, thanks!
I've now got 3 footnotes for my experiment:
1. Analysis is done in KWH. Further analysis will be needed to take into account peak/off-peak period utility billing rates.
2. Heat strips are not present / turned off.
3. Solar heating is kept consistent across various set back experiments (i.e., same curtains/blinds are opened/closed at same time of day, each day).
I think that once I establish there's a consistent relationship between KWH and HDD for my current in place 3° set back, I'm going to jump right to a no set back environment. I am most curious on whether or not money (KWH) can be saved by simply 'not touching the dial,' as many advocate.
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>
|
|
|
Bill Neukranz
Veteran Member
Posts:1103
|
| 08 Dec 2008 03:08 PM |
|
Looking at this further, I've concluded that the chart I referenced in the first posting in this thread ( http://welserver.com/WEL0043/SetbackEfficiency.gif ), showing KWH/HDD as a function of time, is of limited use. It's hard to use this chart to predict KWH usage, given outside weather HDD and inside tstat set back pattern.
It also appears to make limited sense to look at the data at any finer resolution than on a daily basis.
What is of better use, I believe, is a scattergram plot showing KWH usage as function of HDD. An example of this is shown below. This chart shows the daily KWH and HDD for the first 8 days of Dec, noting I currently have in place a 3° tstat set back pattern.
Pretty clearly there's a linear relationship between daily KWH usage and daily HDD. In fact, for my current 3° set back pattern, the linear relationship is
KWH = 1.78(HDD) - 11.3 .
This says that for every HDD for me, it's going to cost me 1.78 KWH (for the day).
Thus, it looks like I can predict how much KWH I will use (for heating) if you give to me the expected HDD for the day (assuming the HDD is in the linear part of the graph).
So it appears that it's the slope of the line that will change with different tstat set back patterns. I.e., for a zero set back (never touch the dial') strategy, the slope of the line will go up or down (don't know which yet). I.e., I'm going to consume more or less KWH, daily, per HDD, with no set back in place.
I'm going to think about this some more before I change the experiment to zero tstat set back.
Comments are welcome!
Best regards,
Bill
|
Attachment: KWH vs HDD.jpg
|
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>
|
|
|
Ona
Basic Member
Posts:189
|
| 08 Dec 2008 03:42 PM |
|
How do you get your scatter plot to show in the post?????
I've been tracking this exact data for my system.
Attached is my scatter plot. What I've noticed is that somewhere in the 25-30 HDD the slope of the "line" changes. I believe that at ~ 30 I use more KWh per HDD than below 30. |
Attachment: HDDvKWH.pdf
|
|
|
|
Bill Neukranz
Veteran Member
Posts:1103
|
| 08 Dec 2008 08:42 PM |
|
Posted By Ona on 12/08/2008 3:42 PM
How do you get your scatter plot to show in the post?????
(Using Windows) I created my scatter plot using MS Excel.
I copied the plot to the Windows scratch area.
I created a new image using Photoshop.
I pasted the plot into Photoshop from the Windows scratch area.
I saved the image as a JPEG, adjusting the compression to ensure the file size is less than 100 KB.
Here in this BBS I clicked on 'Browse' for Message Attachment, and found the JPEG file I just created.
I clicked on 'Submit.'
I'm sure there's more efficient techniques, not requiring expensive products like Excel and Photoshop, but these are the tools I have at hand, and thus how I did it.
My linear expression to relate KWH to HDD, with a 3° set back pattern, is
KWH = 1.78(HDD) - 11.3 .
But obviously if HDD = zero, KWH consumed can't be -11.3 KWH. So what this says is the relationship between KWH and HDD has to become non-linear around the 6.3 HDD point and warmer (fewer HDD). I think this is the phenomena you're observing.
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>
|
|
|
engineer
Veteran Member
Posts:2749
|
| 09 Dec 2008 07:10 AM |
|
Another way to think of this is that your system's HDDs don't correlate to the national weather service's 65 deg F basis - your house has sufficient solar gain or internal sources of heat so as to require no heating down to an average outdoor temp of 55-60. |
|
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>
|
|
|
Masoud
Basic Member
Posts:180
|
| 09 Dec 2008 10:12 AM |
|
Hello Bill / Ona:
Bill wrote: "... the relationship between KWH and HDD has to become non-linear around the 6.3 HDD point and warmer ...". It seems to me, the relationship between KWH and HDD in equation KWH = 1.78(HDD) - 11.3 stays linear, but it is subject to KWH being greater or equal to zero.
The lady at the propane company told me their estimates of customers' propane use, thus deliveries, are based on HDD and a "building factor". They are often correct to 5% of the actual use. Their projections are more accurate for newer houses. Estimates are better for customers who use propane only for heating, as compared to those also using propane for cooking, clothes drying, and power generating.
I think in an older house, like Ona's, wind has a greater influence on the building's heat loss at lower outside temperatures. Also for colder NY weather, relatively larger demand for heat to capacity ratio, due to the HVAC design specs, the heat pump operates longer in the second stage at a lower cop, when HDD is substantial. These two factors (wind effect & lower 2nd stage cop), in my opinion, contribute to a change in the slope of the imaginary plot depicting the relationship between KWH and HDD for Ona's house, around 30 HDD.
I noticed at my house, a burst of bright sunshine, within minutes noticeably reduced operating time, and extended off time, between on / off cycles of the heat pump, when outside was 15 degrees a few days ago.
Rgegards, Masoud |
|
|
|
|
Bill Neukranz
Veteran Member
Posts:1103
|
| 09 Dec 2008 01:07 PM |
|
Thanks for the further comments.
Yesterday was a warm day - producing a nonlinear KWH/HDD value. Yesterday was a 5.79 HDD day using 5.62 KWH Cold in the morning, and hot in the afternoon. In fact, yesterday also was 1.82 Cooling DD.
My list of foot notes is now up to four. A KWH/HDD index is dependent on:
1. Wind conditions (I can't do anything about this one).
2. Solar gain of the home (I can avoid this influence by opening/closing window blinds everyday at the same time).
3. Aux heat strips (not an influencer for me because I don't have them).
4. Use of 2nd stage heating (won't be an influencer for me because I have my tstat set to avoid using 2nd stage heating).
I'm getting close to being ready to change the experiment to no set back.
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:
290 |
 |
Members:
0 |
 |
Total:
290 |
|
|
|