Calculating $/W

[aliris]

Apologies: this has been discussed quite a bit even lately but probably there's a slow dribble on it always.

Still I'm not sure how (and have searched fruitlessly to find how) $/W is calculated from the CSI database because there are so many technical ways of estimating power.

Is one of the following the correct denominator? Someone elsewhere discussed that they felt AC rather than DC was a better scale but I think there's a convention to this, no?

Nameplate Rating	Direct current output of photovoltaic module under Standard Test Conditions in kilowatts
CEC PTC Rating	Alternating current output of photovoltaic module under PVUSA Test Conditions in kilowatts
Design Factor	A ratio comparing a proposed system’s expected generation output with that of a baseline system
CSI Rating	Alternating current output of photovoltaic module calculated by multiplying the CEC-PTC Rating with the Design Factor

TIA

Or instead one of these?

CEC PTC Rating Fixed	CEC PTC Rating of the fixed array portion of the project
CEC PTC Rating Single-Axis Tracking	CEC PTC Rating of the single-axis tracker portion of the project
CEC PTC Rating Dual-Axis Tracking	CEC PTC Rating of the dual-axis tracker portion of the project
CSI Rating Fixed	CSI Rating of the fixed array portion of the project
CSI Rating Single-Axis Tracking	CSI Rating of the single-axis tracker portion of the project
CSI Rating Dual-Axis Tracking	CSI Rating of the dual-axis tracker portion of the project

[aliris]

In the thread "Questions on proposal comparison" it's noted STC is conventional, so "Nameplate" seems right?

But.... the $/W from the WorkingDataSet are just too high. And... that db despite saying it is current seems to have practically no current data. More, when I use their calculators as a test they reference projects that are very recent that are not in the WorkingDataSet.

What am I missing, where is the recent data please? Anyone??

TIA

[cebury]

It's nameplate rating.
Keep in mind lots of folks are overpaying due to lack of education, competing bids etc. And lots of projects are Total Project cost in the application, not the cost for just solar. NoRmally main panel upgrades aren't included, roof work, etc. I had an optional MPU even when they could've line side tapped, also had roof work done, and extra electrical done fo new circuits. They listed the total project cost in my application, so it's impossible to do accurate $/watt based on mine. When I reviewed the database before my purchase, all were way over my estimates per watt except for one which was half my quotes. Maybe that was an inside job, because when I called that vendor and tried to negotiate a price, they wouldn't even lower it to my other quotes. They were one of the Most expensive quotes I got. No where near the database listed cost.

The database was completely useless for me to determine going rates. Getting competing quotes was the only way.

[J.P.M.]

In the thread "Questions on proposal comparison" it's noted STC is conventional, so "Nameplate" seems right?

But.... the $/W from the WorkingDataSet are just too high. And... that db despite saying it is current seems to have practically no current data. More, when I use their calculators as a test they reference projects that are very recent that are not in the WorkingDataSet.

What am I missing, where is the recent data please? Anyone??

TIA

Do the $/Watt yourself by adding a column. Divide the total $$ by the number of panels and the STC rating and put the result in that column. Then, sort out the residential jobs, sort by date, vendor, zip, or anything else. As for up to date, the last entry I saw in Col. "Y" was dated 10/25/2016. The database takes some getting used to, and it's not as good as it was before the CSI incentive program wound down, but it still contains useful information, particularly when comparing relative vendor or equipment prices. It says nothing however, about the quality or integrity of the vendor or their work. It is no more than another source of information. Some understanding of its power and limitations is helpful and necessary.

[aliris]

cebury, those databases are a puzzle.

The currently-connected db doesn't have any price-paid info. But it does seem fairly current.

The working db has incentive payment info and some recent dates but…it doesn't at all match with what you pull up through their search-by-zipcode screens. That internal pool is far more current. So even though the db says it's drawing on this data and is current, it isn't.

As for including capital improvements in the project cost, that seems odd to me because the "working" database tracks incentive money paid and I would have thought they would have been scrupulous about not including that. I agree though that by all evidence the db seems quite "unclean", containing wrong info all around.

And I can't quite imagine how using it as a bargaining chip is going to work well, because what am I to say: "I saw in the CSI db that you gave that guy in the next zip-code over a better price" and he'll just say "yeah, I have a cheaper sub there" or something. Either the vendor will or won't give you a better price but it's hard to see how this will help get to one. OTOH others are better bargainers than I.

JPM, I agree for sure the db is interesting. If you're not familiar with the totally-better-than-sliced-bread feature of excel called "pivot table", I suggest playing with it. You don't need to sort, it will automagically slice/dice contingencies and cross-tabs and filters of any parameter you wish. I'm no fan of Excel generally but that bit of it is amazing.

Now, if only the CSI db were as amazing…. I don't trust it to give me the info I'm looking for. I can tell, for example. that in my area Enphase is used twice as much as SE optimizers. Again, whether that's significant I just don't know. It is clear that SunPower is an order of magnitude more popular. And more expensive. Per the estimate of earlier.

But all this changes with time and these data are old. So… I agree with cebury nearby vendor's estimates are a better metric.

[cebury]

I download the db and then IIRC I had to export to smaller files then I imported into MSAccess and ran queries against it. I added multiple layers of queries to fix data integrity issues, just because I had time to do it. Like combine all "SolarCity" and "Solar City" etc. I was curious about vendor totals for my area.

It sounds reasonable the preCSI could've been reimburseable costs only. Sounds like it's changed since I last looked, but that working-db is probably only exported from their searchable, perhaps at intervals?

Good luck to you..

[aliris]

Right on the terrible shape of the data! I wasted an hour or so cleaning things up like Solar City==SolarCity. Then the computer crashed and lost it all reminding me what a waste of my time! I get the point. There are several orders of magnitude more vendors out there than I had any idea of, and some of them do several orders of magnitude more business than others. It's a whole political movement I really didn't know about. And combining public policy with private enterprise is a wild endeavor.

$/W makes a lot more sense to me now than formerly. It's been said here repeatedly "it's just business" and I get that, narrowly, now more than formerly. So... it wasn't wasted time. Learned a little more about excel too.