We charge about $2425K for a turn key horizontal geosystem, with HWG, 2 tanks and dual stage compressor. That is $6K/ton, not the $9K/ton you keep referring to. A 3 ton system is about $7-7.5K/ton.

One price point says little about other areas, but based on this progression, $9k/ton for a 2.25 ton system is quite accurate. But if someone has absolutely decided that they are going to use geo (a very odd decision), then one should compare *incremental* cost per ton (plus operating costs and lifetime) to more or less insulation. The big savings with more insulation comes not from downsizing the geo system but when one decides that a small geo system isn't cost effective and replaces it completely (with something less expensive).

Again assuming you are going to have some heating system then 6K ton or 9K a ton are not reflective of the additional cost of geo, as pre tax credit price of geo is little more than the cost of the loops higher than a conventional system. Post tax credit costs are often similar between geo and fossil in my AO. However, I'm not a code minimum builder so if we were to get compared to lowest bidder (code minimum means one central return and poor duct design execution as well as 80% furnace and 13 seer AC), we might be 2K/ton higher tops (after tax credits) or 4K pre tax credit.

The source of the $9K/ton reference/(rumor?) is the the average cost for ~5-6 ton average sized systems in nearby CT under their state rebate programs. (Is it reasonable to assume Hudson Valley prices differ dramatically from the CT average?) That number is also pretty consistent with 4 ton+ system quotes I've seen (sample size very small) in MA, where 2-3 ton systems are quoted come in around $11-13K/ton.

But if you can get competent installations at a better price point than that, by all means DO!

FWIW: (and it ain't worth much), the daughter of a friend of mine recently had a tiny 1-ton system installed in North Carolina at a pre-subsidy cost of about $12K.

Posted By jonr on 13 Aug 2013 09:55 AM

We charge about $2425K for a turn key horizontal geosystem, with HWG, 2 tanks and dual stage compressor. That is $6K/ton, not the $9K/ton you keep referring to. A 3 ton system is about $7-7.5K/ton.

One price point says little about other areas, but based on this progression, $9k/ton for a 2.25 ton system is quite accurate. But if someone has absolutely decided that they are going to use geo (a very odd decision), then one should compare *incremental* cost per ton (plus operating costs and lifetime) to more or less insulation. The big savings with more insulation comes not from downsizing the geo system but when one decides that a small geo system isn't cost effective and replaces it completely (with something less expensive).

At IRC2012 code min (R20+5, which is about R20 whole-wall) 5000 square feet of wall would have a heat load of about 20,000BTU/hr.  At ~$2/square foot of taking it up to R40 (2x code) is $10K, and it takes 10,000 BTU/hr off the heat load, or about $12K/ ton, which is still pretty reasonable even if it's only offsetting $7-8K of cost on the mechanicals since the avoided ton used no power.

But if one cheaped out and only went 1.5x code for R30, it takes ~7500BTU/hr off the heat load for only ~$5K of additional cost, or $8K/ton, which is cheaper than the average per-ton cost average for geo in CT, and clearly a good investment.

You can drive yourself nuts on the penny-accounting on competing subsidies, etc. but currently unsubsidized grid-tied solar PV + air source heat pumps is a comparable or better investment than a $9K ton GSHP in MA, but still not as good as going 1.5x code min on R-values.  Using the "Pretty Good house" rule of thumb on R-values is usually pretty close to or slightly over the long term financial-rationale limits for US Climate zone 5, but of course that depends on the cost of the methods you use to get there.  (At 17-18 cents/R-foot for closed cell foam it breaks down a lot sooner than 3 cents/R-foot cellulose, with 10-11cents/R-foot open cell or EPS/polyiso somewhere in between.)

But if one cheaped out and only went 1.5x code for R30, it takes ~7500BTU/hr off the heat load for only ~$5K of additional cost, or $8K/ton, which is cheaper than the average per-ton cost average for geo in CT, and clearly a good investment.

Using incremental cost, it's something more along the lines of spending $5K to save $2-3K up front (reduced geo system cost) and $70/year (both are illustrative guesses). ~36 year payback isn't so good.

Something like that. In almost every case in this zone 5/6 location there will be a long term rationale for R30 whole wall, and with GSHP @ CT pricing it's a no brainer (until you confuse the issue with different subsidy equations), whereas the next R10 taking it to R40 may cross over the financial model limits even for a comparatively longer term.

At R40 the NPV of the last R5 may be "neverland" if you went with natural gas in anticipation that the folks drinking the frack-water are right, and that natural gas will cost $2-3/MMBTU for the next century. The sweet spot is probably between R30 & R40 in most cases, but it depends on what you include in the financial model for energy inflation and discount rates.

The Building Science Corp analysis comes up with ~R35 whole-wall for climate zone 6, R30 in climate zone 5, but that's for more conventional mechanical system costs. Adding R5 to those numbers when GSHP is in the mix isn't insane. See Table 2, p.10 of this document:

http://www.buildingscience.com/documents/reports/rr-1005-building-america-high-r-value-high-performance-residential-buildings-all-climate-zones

The upper Hudson Valley is the cold edge of zone 5, warm edge of zone 6.

Adding R5 to those numbers when GSHP is in the mix isn't insane.

My point is that once you have decided to put in a GSHP, you have relatively low incremental costs for more capacity and you have low operating costs. Both tip the scale towards relatively less, not more, insulation being cost effective.

If 2 ton systems are ~$12K/ton and 5 ton systems are ~$9K/ton (as would seem to be the case in New England- clearly YMMV), there's still plenty of room for insulation upgrades above code-min on the $32K differential in system cost for downsizing the GSHP from 5 tons, (code min house with lots of pretty window views) down to 2 (1.5-2x code min house with judicious use & types of glazing.)

It's not as if a gas fired system or an air source heat pump is going to be free, and probably isn't more than a $10K savings (if that) below 2-ton GSHP. But the IRR of a $10KUSD difference based on the lower operating cost of the GSHP isn't very much compared to the IRR of the 2 kilowatts of grid tied PV that net metered at residential retail at 15 cents/kwh.

It's never a simple model, and it has many moving subsidy factors. The cost of PV is still crashing (~$4-4.50/watt installed unsubsidized price regionally when I last checked), but the net metering models will surely be changing, now that it's passed a sub-retail inflection point on lifecycle per kwh cost. PV too has a 30% federal tax credit at least through 2016. The rate of increase of the installed base of distributed small scale PV in the US is on an exponential trajectory, doubling about every 2.5 years.

http://dqbasmyouzti2.cloudfront.net/assets/content/cache/made/content/images/articles/Cumulative_US_DG_PV_to_1H_2013_580_333.png

At R-values sufficient to get to Net-Zero-Energy with rooftop arrays, PV & air source heat pumps is poised to eat GSHP lunch, unless the installed price in New England comes down substantially. In Germany where the permitting issues are streamlined and the utilty rates are 2x those of NY you can get 2-6kw systems installed at under $2.50/watt, using the same panels, and (essentially) the same inverters. (And it's not because labor in Germany is so cheap.) The US will very likely hit that price point before 2020, a point at which the lifecycle cost per kwh would be sub-retail almost everywhere in the US (and dirt-cheap in NY/NE.)

FWIW: At the moment most of NY still follows the IRC 2009 code, which would mean R15 whole-wall type construction even for Climate zone 6, and R30 whole-wall would be 2x code-min. Most of the Hudson Valley is US Climate zone 5, but Essex, Warren & Ulster counties are zone 6 (Westchester & the Bronx are zone 4).

http://energycode.pnl.gov/EnergyCod...New%20York

Dane, just curious. Is labor in your area union?

Would there be no heating system in the absence of geo? Of course there would be. So in the discussion of new builds the cost per ton would be total installed cost, minus fossil system cost, minus tax credits. No one is going to pay 12K or 9K/ton if you use fair math.

Posted By joe.ami on 14 Aug 2013 09:33 AM
Would there be no heating system in the absence of geo? Of course there would be. So in the discussion of new builds the cost per ton would be total installed cost, minus fossil system cost, minus tax credits. No one is going to pay 12K or 9K/ton if you use fair math.

And if you use the same fair math for the 30% tax credit + local percs for ~$4/watt PV, the 2-ton GSHP  @ 0.7 x $24K= $16.8K isn't the clear winner by any means (and sometimes loses), against $12K of air source heat pump + 1.5-1.7KW of (30% tax subsidized) grid-tied PV.  If you stick with the code-min ~50KBTU/hr load scenario and 4-5 tons of GSHP for $25-30K post-subsidy price  vs. say $18K of ASHP + 10KW (more, if the utility will allow it) of PV the argument isn't any better, but it might be tough to optimally fit the 10KW array on your roof (but not impossible, if you design for it)

Like I said, competing subsidies make for moving targets and a messy financial model, but with the 30% tax credit people are already paying ~$3/watt, sometimes less, post-subsidy.  With some local subsidies fattening the pie it's already approaching German-style costs in those locations.

In the upper Hudson valley, with decent shading factors you get about 1.2-1.3kwh annually per watt of PV- a 10kw array would deliver 12,000kwh/year or better, and would still usually cover the efficiency delta in power use between a pretty-good ASHP and a pretty-good GSHP, and significantly more than the delta between best-in-class ASHP technology vs. merely pretty-good GSHP.

As US costs match or beat the Y2012 unsubsidized cost of <$2.50/watt Germans were paying for rooftop PV, if combined with rough-justice net-metering @ retail it gets even harder to make the case for GSHP, at least in climates where the air source solutions still work (which would include the upper Hudson valley.)  If/when you can get to net-zero heating & cooling cost (if not total net-zero-energy) for even the current subsidized cost of the GSHP system the competition will have to become blistering fierce. With the German existence proof of system costs in that range, and projected panel costs of 35-36cents/watt by 2017 from PV industry analysts, you have to believe German-style installed costs for PV are coming to the US, and soon (much sooner than I would have anticipated even three years ago.) 

There's reason to believe that net metering terms won't look exactly the same in 2020 as they do in 2013, and the subsidy for PV (like that for GSHP) is set to expire in 2016, but odds are good that PV won't need it.  The the PV train has already left the station: In CA last year there were more than 3000 of grid tied roof top PV installed beyond what was covered by any kind of subsidy, and that's at the full $4/watt. Care to guess how fast the stuff will go up when the unsubsidized price point is bouncing on $2/watt, in a retail-electricity market as expensive as NY?  Utilities caught unprepared for this level of distributed generation built into their operating & business models are going to be in for a rough ride, if not road kill.  But there's no stopping it at this point- it's only a matter of "when", not "if". My best guess is that it'll happen before 2020, with a temporary change in slope in 2016/2017 as the subsidy goes away. (Expect a PV-rush in 2016 as fence-sitters make a last-ditch dive for the subsidy, unless it gets extended.)

If 2 ton systems are ~$12K/ton and 5 ton systems are ~$9K/ton (as would seem to be the case in New England- clearly YMMV), there's still plenty of room for insulation upgrades above code-min on the $32K differential in system cost for downsizing the GSHP from 5 tons, (code min house with lots of pretty window views) down to 2

YMMV = your math may vary? (5x9)-(2x12) = $21K not $32K

I think it's a reasonable summary to say that one should run software like BEopt (or hire someone who can do an accurate and unbiased job) to analyze all the options using local quotes. Solar, geo, nat gas, hydronic, insulation ranging from code minimum to 2x, off peak storage, etc. I suggest a payback period not exceeding how long you expect to live there (or be prepared to lose money on most better than code min energy savers).

Most custom built high performance luxury house isn't typically a 3-years & flip kind of deal going in, it's more often a commitment to decades. But you can't make a financial rationale for even code min in a single decade time frame, often not even in two decades. Using a payback period against the anticipated dwelling period is more than just a bit silly.

Time of use rates aren't universally available in NY, but could be a real factor in the financial modeling. If you assume the GSHP runs at an annualized average of 4 and a mid-range ASHP option only gets 2, paying off the difference in upfront cost takes a lot longer if you're paying 8 cents for most of the heating power use vs. 15. And if you're net-metered for the PV at the peak-power retail the payback on the PV is better than net metered at fixed-rate, since most of the output occurs during demand-hours, and a significant fraction during the extreme-demand hours when time of use rates are the highest.

But most of these decisions are never purely (often not even barely) an exercise in pure financial rationality. Comfort is a real issue most luxury home buyers are willing to pay something for when they understand the options. R30+ (whole wall) walls are noticeably more comfortable than R15/20 at the temperature extremes. U0.20 windows offer an even greater comfort factor over code-min. Radiant floors are noticeably more comfortable in a code-min house, but that factor is greatly diminished in a high-R house due to low to non-existent heat loads during the daylight hours.

At $25K for a 4 ton system, and an heat extraction from the ground of 27 KBTU/H (about 8 KW), geo runs about $3.14/watt before subsidies. Add 30% tax credit and some local subsidies to the equation, and you get down to $1.50 - $2.20/watt installed capacity. The beauty about it is that the sun does not have to shine, and you can extract the heat on demand. It does not abuse the grid for storage.
Now, solar on the roof producing the power to run the geo, why does that sound familiar?

http://hpac.com/heating/church-takes-nontraditional-approach-heating-installs-geothermal

Dana in your solar scenario you would have to buy solar plus a conventional heating system. Your suggesting only about 6K to heat and cool 5000 SF. I don't think so.

geo runs about $3.14/watt

~$4/watt PV

Comparing $/watt between solar and geo ignores the critical issue of how many hours they run. Stick with ROI for comparisons. Eg, 20 years for solar vs no solar compared to n years for geo vs nat gas.

Given that one can go here and buy panels for $.70/watt, I have no doubt that $2.50/watt total systems cost and perhaps a 13 year ROI is coming soon.

Posted By joe.ami on 15 Aug 2013 08:59 AM
Dana in your solar scenario you would have to buy solar plus a conventional heating system. Your suggesting only about 6K to heat and cool 5000 SF. I don't think so.

Without re-reading my prior posts, IIRC I included an air source heat pump the tune of something like $15K for the 5 ton GSHP @ $9K/ton pre-scenario, and worked in the solar on the delta in cost. But  or yuks, lets re-run the numbers using the (fedral only) subsidized costs for both the GSHP and PV.

GSHP:  0.7 x 5 x $9K= 31K

...less $15K for the ASHP leaves $16K...

... a $3/watt you get ~5300 w-pk of PV, delivering 6400 kwh/year.

Is that enough to pay for the difference in systems operating efficiency? (Sometimes yes, sometimes not- probably not in a code min house with 8-9 months of heating season.)

In the 2-ton $12K/ton scenario, the subsidized GSHP costs 0.7 x $24K= $16.8K.

If 2-tons of ASHP costs more than $12K I'm in the rong bizness (or it's a Daikin Altherma) leaving $4.8K on the table.

At a subsidized $3/watt that buys 1.6kw of PV that delivers ~2000 kwh/year.

Is that enough to pay for the difference in systems operating efficiency?

Sometimes yes, sometimes not.  Marc Rosenbaum lives in a ~1200' fairly tight house at R-values below the Pretty Good House levels I was recommending, but still above code-min, in a somewhat warmer climate (Martha's Vineyard MA). The separately metered power used in his 1-ton Fujitsu is well under 2000 kwh/year total!

He is an energy nerd with very MISERLY use of plug load power (not much help on the heat load there), and his girlfriend still won't let them freeze in the dark to save power, (though I expect Marc would, if left to his own devices. :-) )

At 2 tons peak load in a Pretty Good House we're talking a short and shallow heating season- something like 6 months-7 at most, and with a bit of passive solar considerations the duty cycle on the heat pump stays pretty low, even in the middle of winter, since the plug loads & hot bodies are a significant fraction of the average load, and the sunny winter day heating load is negative (it gains enough passive heat to coast through most of the night. This can be optimized by design, but it's true even for non-optimized homes.)

At typical PassiveHouse heated with a mini-split uses about 1.5kwh/per square foot per year on the mini-split.   Even at 1.5x PassiveHouse power use in a heat pump, a 5000' house would only be using  ~11,000 kwh/year with ductless heat pumps, and a better-class ducted air source heat pump might use 13,000 kwh/year, whereas a best-in-class GSHP maybe 9,000.  If that's the case, best in class GSHP would be the better deal (if best-in-class really only costs $12k/ton installed.) At a more typical average system COP as-installed by the less-than-meticulous designers it's close to a wash.

What I'm saying is that as PV hits $2/watt and lower it's going to become an even bigger challenge to make the case for GSHP on pure financial grounds, and that IS where it's going.  Utility scale PV in the US is already coming in at a contracted paid for and built $1.7-1.8/watt, and many states are sweetening the pot for the residential scale systems substantially on PV, well beyond the 30% federal kick back (in much the same way that CT was kicking in another ~$1K/ton for GSPH until the funds were depleted), bringing the total cost down to the German unsubsidized levels (but SFAIK not $2/w yet.)  When there are competing subsidies the model gets messy, but we're already in that grey area, moving closer to the PV/ASHP every year.

In smaller houses built to the Pretty Good House levels (not Passivehouse) builders are shrugging, throwing in 1.5-2 tons of mini-split (multiple heads for distribution reasons, not total heat load) and calling it a day, and more often than not it works. With a ~2-ton heat load and ducted air source heat pump it would still be pretty easy to fit a big enough array on the roof of a 5000' house to get to net-zero (if that were a design goal) whereas with a 5 ton heat load it's impossible.  (Heating a 5000' house with a tractor-trailer-load of 3/4 ton Mr. Slims might be possible too, but I'm not willing to go there on a luxury house that size the way I might on a 2000' rancher with an open floor plan.)

Posted By docjenser on 14 Aug 2013 11:44 PM
At $25K for a 4 ton system, and an heat extraction from the ground of 27 KBTU/H (about 8 KW), geo runs about $3.14/watt before subsidies. Add 30% tax credit and some local subsidies to the equation, and you get down to $1.50 - $2.20/watt installed capacity. The beauty about it is that the sun does not have to shine, and you can extract the heat on demand. It does not abuse the grid for storage.
Now, solar on the roof producing the power to run the geo, why does that sound familiar?

http://hpac.com/heating/church-take...geothermal

I keep reading about these beasts, but they're like sasquatches in my area- everybody has heard of them, some even believe they exist, but nobody I know has actually seen one in this state, which is closer to the Hudson Valley than western NY.  Without a quote on the table I surely wouldn't budget it. The CT rebate average seems to be a more realistic budgetary number based on what people are quoted here. YMMV.

Conflating the PV peak-watts with GSHP output watts is more than just a bit silly!  They're not the same thing. 

PV has no operating cost, lasts for 40 years with minimal maintenance, and is putting out a premium product- electricity, something that can be used in a heat pump to deliver 2-5x that amount of power as low-grade heat in a water heater, air conditioner, or even a GSHP. (A kilowatt of grid tied PV and a $100 window shaker moves 3-4 kilowatts of low grade heat.)

The GSHP's output IS the low-grade heat, has a much shorter equipment life, and has a significant operating cost (the high grade energy in the form of electricity).

So, (even assuming the sasquatch price point on the GSHP is realized) what's the point of comparing their price/watt?

(And that's not even getting into the fact that one rightly needs to subtract the input power from the GSHP and only use net-watts, or that the PV output is it's output at a specified incident irradiance, not an operational average, etc.  These are completely orthogonal numbers that no bearing on one another.)

Posted By Dana1 on 15 Aug 2013 05:22 PM

Posted By docjenser on 14 Aug 2013 11:44 PM
At $25K for a 4 ton system, and an heat extraction from the ground of 27 KBTU/H (about 8 KW), geo runs about $3.14/watt before subsidies. Add 30% tax credit and some local subsidies to the equation, and you get down to $1.50 - $2.20/watt installed capacity. The beauty about it is that the sun does not have to shine, and you can extract the heat on demand. It does not abuse the grid for storage.
Now, solar on the roof producing the power to run the geo, why does that sound familiar?

http://hpac.com/heating/church-take...geothermal

I keep reading about these beasts, but they're like sasquatches in my area- everybody has heard of them, some even believe they exist, but nobody I know has actually seen one in this state, which is closer to the Hudson Valley than western NY.  Without a quote on the table I surely wouldn't budget it. The CT rebate average seems to be a more realistic budgetary number based on what people are quoted here. YMMV.

Conflating the PV peak-watts with GSHP output watts is more than just a bit silly!  They're not the same thing. 

PV has no operating cost, lasts for 40 years with minimal maintenance, and is putting out a premium product- electricity, something that can be used in a heat pump to deliver 2-5x that amount of power as low-grade heat in a water heater, air conditioner, or even a GSHP. (A kilowatt of grid tied PV and a $100 window shaker moves 3-4 kilowatts of low grade heat.)

The GSHP's output IS the low-grade heat, has a much shorter equipment life, and has a significant operating cost (the high grade energy in the form of electricity).

So, (even assuming the sasquatch price point on the GSHP is realized) what's the point of comparing their price/watt?

(And that's not even getting into the fact that one rightly needs to subtract the input power from the GSHP and only use net-watts, or that the PV output is it's output at a specified incident irradiance, not an operational average, etc.  These are completely orthogonal numbers that no bearing on one another.)

The church in the link was 28 tons and the price was $154,000 turn key, that is $5,500/ton with very expensive high temp heat pumps and expensive Wilo pumps for dual redundancy.

The comparison is not silly. Both systems extract solar energy, the geosystem needs electricity to run it, but does not require and external storage system, which is the biggest problem. We simply use the grid for that. PV does not always work when you need it, geo is on demand. And how does the PV system 40 years life expectancy works after a hailstorm (we had 2 here in the last 8 years), not speaking about the performance after 1 inch of snowfall. (See attachment)
The numbers I gave you were heat extracted from the ground, not BTUs delivered to the space. Both systems can work nicely together, since one can make the electricity the other one needs to extract even more energy out of the ground for space conditioning. Still the biggest problem is the lack of energy storage with PV. Your posts are perfect examples what I call you living in an theoretical world. For all the knowledge you have, it does not account for the real world, the hailstorms, the snow, the real installation costs. You simply continue to make up numbers and scenarios which are far away from the real world to make them fit your thinkings and your philosophies.

I'll agree with one thing, the comparison is silly. 25K for 4 tons is quite common around here, 3/watt not so much.