Is this a reasonable solar bid? — and other questions
Howdy folks,
I got my first solar bid today, from Consolidated Solar Technologies. For a 3.2 kW ground-mount array in Albuquerque, New Mexico, the total cost before the federal tax credit is $16,300, and $11,410 after the credit. This works out to $3.56/watt.
One thing I worry about is that their calculator shows a higher projected output than PVWatts does. Their tool showed 6,500 kWh yearly, but PVWatts shows 5,844 kWh. This is supposed to be a net-zero-enabling array, so I don’t want to underproduce. I’m using 6,500 kWh for my yearly usage
Another concern: my electrical panel has four slots left, and the estimator told me they generally need four slots to interconnect. This is an issue because I have yet to add the 240v breaker for an electric heat pump, which would leave me with only two slots–not enough.
Finally, I worry that my estimating is off; I haven’t yet installed the heat pump and am only estimating its use. Last year was really cold and we used 176 therms (delivered, not purchased–accounting for 80% efficient equipment) to heat the home for the whole heating season. That translates into 17,600,000 BTUs. At a COP of 3.5, that requires 1,474 kWh for the season. I figure 3.5 is reasonable since much of the season features sunny daytime temperatures that rise into the 40s and 50s where the unit operates at high efficiency, and we can be trusted to keep the thermostat low at night and only let the machine turn on once outside temperatures have risen a bit (also the outdoor unit will be on the east side and getting bathed in sunlight most mornings). Summer AC usage will be much lower since we already have a swamp cooler that works fine most of the time.
So here are my questions for you guys:
- Does $3.56/watt sound reasonable?
- Should I trust PVWatts, or their calculator?
- Are their installers/companies/equipment that can use only two slots in an electrical panel?
- Is my estimate of heat pump electricity usage sane?
Thanks!
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
Prices in most of the US are around $3 per watt, before federal tax credit. I'd ask them how much an additional kw would cost, because their production estimate seems very high. One reason for the very high per watt cost may be the small size of the system
My pv system production was very close to the PVWatts estimate, for whatever that's worth.
I'm in the Chicago suburbs, and the company we're thinking about using quoted:
2.9 kW system (11 panels on the roof)
$12,500.00 (before credits)
$8,700 (after federal tax credit)
In Illinois, we also have Solar Renewable Energy Credits, so this reduces costs further, over time, by an estimated $3,800 (you get paid for every 1,000 kW produced).
I've gotten several quotes, and they've ranged in price (higher and lower than what's outlined above). I had pretty good luck using Energy Sage (energysage.com) as a resource for facilitating quotes (although some installers opted out, saying they were too busy to bid).
The national average as of November is $2.89 watt installed before ITC. According to the latest GTM stats.
But $2.89 that's the national average. Some markets are more competitive or higher volume than others and even within lower priced PV areas there is price variance. At the retail installed pricing end for ~20% efficiency panels from a first tier vendor you'll typically pay a 25-30 cent premium over 15% panels from a first tier vendor, and an even higher premium over lower tier panels & inverter vendors. Consolidated uses only Sunpower panels, a US company whose claim to fame is their 21% efficiency Maxeon line, which will be more expensive than a 15-17% efficiency first tier Chinese panel.
Hopefully they fully specified the panels & inverters in the quote(?).
A COP of 3.5 is HSPF ~12. The labeled HSPF of the heat pump is reasonably valid for zone IV on this map:
http://www.fsec.ucf.edu/en/publications/html/FSEC-PF-413-04/images/Figure5_lg.gif
You'll note that Albuquerque NM does indeed fall within zone IV on that map, so hopefully the heat pump you are specifying is labeled HSPF 12.0 or higher(?).
Most ducted heat pumps are backed up with resistance heating strips. If that's the case the LAST thing you would want to do is set back the thermostat overnight, since that forces the heat pump into burning power in the heat strips during the recovery ramp which is dramatically less efficient than just letting the heat pump keep up over night with a "set and forget" approach. What heat pumps are you looking at?
You should be able to size the heat pump by running a heat load calculation based on fuel use, as outlined in this bit o' bloggery:
https://www.greenbuildingadvisor.com/blogs/dept/guest-blogs/out-old-new
Oversizing a heat pump by more than 1.5x at the 99% outside design temp an result in lower comfort as well as lower efficiency. With modulating heat pumps there can be some efficiency gains with oversizing depending on where the minimum modulated output is relative to your shoulder season heat load. (It's complicated, but not impossible to figure out where the sweet spot roughly lies.)
Sounds like I should get some more bids. I don't need Cadillac panels. Watts are watts.
I made my own manual J spreadsheet. For my 1300 sf slightly sub-pretty-good-house, the 99% heat load is about 20,000 BTUs if I keep the attic ducts, and 15,000 if and when I build new ductwork under a dropped ceiling in the main hallway (I'm assigning the house a 25% penalty for ducts in the attic).
I'm primarily looking at American conventional heat pumps; mostly the Carrier Infinity, Bryant Evolution, and various companies' IQ drive units. Japanese mini-splits are rare here, especially mini-duct units, and I'm not real impressed with the on-paper specs of the mini-duct units compared to conventional heat pumps.
With a conventional unit, I would set it up to disable strip heat except for when the emergency heat button is pressed, and operate a wood stove for the one or two nights per decade when it plunges to -20 and kills all the trees in the neighborhood.
How about the whole "need four slots on an electrical panel" thing? Is that universally true, or just a CST thing?
Nate,
I can't answer the "need four slots" question.
I would trust the PVwatts calculator. Of course, PV production will vary from year to year, sometime significantly. If you are using PVwatts, make sure that the orientation and angle you entered are correct.
If the PVwatts prediction differs from that of the solar contractor providing a bid, it's perfectly reasonable to call up the contractor and point that out, asking why their estimate is different.
Nate,
I can easily see two-three slots being taken up by the breaker that either feeds/backfeeds whatever panel is associated with the PV system. There could also easily be other breakers associated with powering controls or something similar for when the PV array isn't powering itself. If that was also a 220v circuit, there would be your four slots.
That price seems reasonable. The lower prices described above I'm assuming are for roof-mounted arrays which are less expensive. Ground-mounted systems are more expensive than roof-mounted systems because of the racking.
Aiming for net zero is tricky. You could
1) Aim to underproduce by a bit, to make sure you don't waste money on a bigger than needed array.
2) Aim to overproduce by a bit, to ensure you meet the net zero target,
3) Design your system to be easily expandable if needed, e.g. using micro-inverters to make it more modular.
4) Wait until after you have operated for a year with your final heating system installed before adding PV.
There are various ways to gain space on your panel. A moderately expensive options that is does well at removing constraints is to add a sub-panel--you use one pair of pole positions to feed the sub panel, from which you can break out as many additional circuits as you need. But you can often gain some space at lower cost by joining two circuits in a single position using tandem circuit breakers. The rules about when and how much of that is allowed are a little complicated. See http://www.ashireporter.org/homeinspection/articles/inspecting-tandem-circuit-breakers/2047 or check with a good electrician.
I would also consider how old your panel is. If it's very old, from the early days of circuit breakers, you might even consider a completely new panel.
One thing to think about is whether there might be an electric car in your future. If so, you might want another high-current circuit for that, which might push you towards installing a sub-panel.
But all of that said, a lot of solar system designs would only need one pole pair.
Also note that your natural gas heating system uses some electricity, and that will go away when you replace it with a heat pump.
Most US brand name modulating ducted heat pump lineups start at 2 tons, and only have a 2.5:1 turn down ratio, which limits the comfort & efficiency factors.
eg. The smallest of the line 2-ton Carrier Greenspeed can deliver ~24,000 BTU/hr @ +18F (Albuquerque's 99% outside design temp) and 18,000 BTU/hr even at 0F, but won't be able to modulate down to your average wintertime load. Even at 18F it's minimum modulated output would be ~10,000 BTU/hr, and when it's in the high 20s or 30s outside (when your load drops to 10K or less) it's minimum output would be even higher. In combination with many of the air handler options it would have an HSPF of 12+, but it's sub-optimally oversized for a 15,000 BTU/hr load. You can play around with this online tool a bit (click on the "Heating Capacities" tab to be able to select different air handlers & compressors.):
http://www.tools.carrier.com/greenspeed/
For the published HSPF of different combinations, see the tables beginning on p15:
http://dms.hvacpartners.com/docs/1009/Public/01/25VNA-01PD.pdf
The as-used efficiency depends on both the on-off cycling, and the fraction of operating time that it's running at it's minimum modulation. Again, "set and forget" is the better strategy, even with the heat strips disabled. Letting it modulate (or cycle at minimum modulation) will use less total power than recovering from setback having to run at it's maximum speed/lowest efficiency.
For a ground mounted systems there is no savings to be had by going with 15% efficiency panels- anything you would save in panel cost would go directly into the increased size of the structural mounting. Going with a first tier panel manufacturer can be important for long term reliability. It doesn't hurt to get multiple bids though, and compare the equipment line items.
Right now I have a single-stage, non-modulating, 80% efficiency 125,000 BTU furnace that is oversized by a factor of five, so my comfort requirements are not especially high. :)
What equipment would you suggest?
The 1.5 ton Fujitsu 18RLFCD has a better than your average mini-duct cassette air handler, and is good for 21,600 BTU/hr @ +17F, yet can still throttle back to as low as 3100 BTUh/r @ +47F, and tests at HSPF 11.3. (=COP 3.3)
http://portal.fujitsugeneral.com/files/catalog/files/18RLFCD1.pdf
Unlike the rest of the pack, the Fujitsu mini-duct cassettes can be mounted either vertically or horizontally, which can prove useful for figuring out somewhere that it can fit.
If last year's quotes from different contractors in MA are any indication, the raw difference in upfront cost between that and a 2-ton modulating HSPF 12+ Carrier Greenspeed could pay for a good chunk of the PV array. The Fujitsu + ducts was quoted under $8K, the Greenspeed + ducts over $15K. (The sample size is small though, one quote each, different contractors. The homeowners eventually opted for a multi-split wall coil solution instead of ducts due to a desire for micro-zoned control.)
Finding somebody competent to design the ducts for the thing could be an issue. Even though it has more push than most other mini-duct cassettes you can't use the rules of thumb HVAC hacks are comfortable with when using bigger-deal air handlers. You'll want it to be hard-piped, no flex, with radiused ells, no sharp throated turns. In a "almost pretty good house" you don't necessarily need to run the supply ducts over to the windows old-school style either.
Plan-B would be the bigger deal 1.5 ton Mistubishi MVZ-A18AA4 air handler married to the 1.7 ton MXZ-2C20NAHZ multi-split, which should be able to deliver the 20,000 BTU/hr even at +17F (you'd have to consult the engineering manuals), and modulate down to about 8K out @ 47F.
https://meus.mylinkdrive.com/files/MXZ-2C20NAHZ_Submittal.pdf
That will likely be more expensive, and lower efficiency than the Fujitsu.
IIRC Daikin now has a 1.5 ton mini-duct cassette that would cut it on BTUs too, but at slightly lower efficiency and with a wimpier air handler than the Fujitsu.
Yes, I keep coming back to the 18RLFCD in my plans but I can't get over these issues:
- Nobody I have found is willing to install it or build ducts for it; they don't even bid. It would be DIY all the way.
- Contractor support for myself or future owners will be close to nil.
- I am not sufficiently confident in my own manual-J-ish spreadsheet to buy a unit only capable of meeting the load at full capacity, without it being a tiny bit oversized. I was extremely aggressive, including R-values for flooring materials and subtracting the heat load from interior electrical loads. With such little excess capacity, at not-uncommon temperatures it wouldn't be able to keep up.
- Efficiencies are not amazing, and the best conventional models do better with more capacity and contractor support.
Honestly my ideal case would be per-room DIYable ductless mini-splits that you can plug into a regular wall outlet that modulate between 1000 and 5000 BTUs of heating and cooling, at high efficiency. Boy that would be amazing. Two such units are http://www.unionaire.us/product.php?productId=3145 and http://mrcool.com/mrcool-diy/, but their efficiencies are not amazing and they'd be hugely oversized. Does anybody know of such a product, or will I be dreaming forever? Basically the mini-split version of a small PTAC/PTHP.