Metrics for cost per KHW hour + thoughts on power optimizers vs. microinverters

[max2015]

Just received our first quote for a 10.7 kW DC PV solar system that according to the installer has an estimated annual kWh production of 17,384 ... We have a two story house with no shade issues... Roof has a southwest orientation...

Our current power usage is 1,000 KHW per month (before PV) and will increase to 1,750 KWH per month with the addition of a recently ordered electric car.

The $50,750 proposal is as follows:

31 Sunpower X21/345 panels

1 SolarEdge SE10000A-US inverter (single phase)

31 SolarEdge P400 power optimizers (instead of using microinverters)

Monitoring is via the SolarEdge portal.

Does anyone have an experience with SolarEdge as a company? What about the use of power optimizers vs. microinverters? Based on my research, this sounds like an ongoing debate...

Final question, what's the rule of thumb for calculating the cost per KHW hour? I know Sunpower panels are more expensive but is this proposal reasonable from a cost per KHW hour?

Thanks for your feedback / advice...

[silversaver]

$$$$$$ shop around.

use search function.

there's no good microinverter support 345W panel yet. unless you wanna try new microinverter as ABB/Power-One....etc

[J.P.M.]

So you plan on driving that EV 20 - 25,000 miles/yr ? If less, that 9,000 kWh/yr. added to the current 12,000 kWh/yr. may be too large an estimate.

Do you have space constraints ? If not, depending on available area, other equipment of very similar quality and annual output can be had for a lot less $$, probably about 20% less.

No shade --->>> string inverter in my book.

Please clarify what you mean by "cost per kWh". That can have many interpretations.

In general, rules of thumb are best used for measuring thumbs.

[sensij]

No shade --->>> string inverter in my book.

I've been looking at data on PVoutput.org and I am less sure this is advice I would agree with. There seems to be a noticeable production hit (a few percent or more) on systems that use string inverters relative to those using microinverters or optimizers, and I don't think it can be explained by location, orientation, panel selection, or data accuracy. In the real world, it appears that very few arrays are truly shade free. There are definitely cost, reliability, and obsolescence risks to choosing something other than a string inverter, but without digging into it too deeply, there does appear to be some benefit to accepting those risks. I think it is worth further study, but not on the top of my list at the moment.

[DanKegel]

10.7 kW ... $50,750...
Final question, what's the rule of thumb for calculating the cost per KHW hour? I know Sunpower panels are more expensive but is this proposal reasonable from a cost per KHW hour?

The rule of thumb is: dividing the cost before subsidies ($50,850) by the rated max power generation of the array (10,700 watts) by yields the price per watt, in this case $5.

$5 is high. $3.50 is more what you should be aiming for these days.

Sunpower is for people with too much money and too little space on their roof. You might consider stepping down one level in crazy to LG 305's. i.e. go get another quote

[max2015]

So you plan on driving that EV 20 - 25,000 miles/yr ? If less, that 9,000 kWh/yr. added to the current 12,000 kWh/yr. may be too large an estimate.

Do you have space constraints ? If not, depending on available area, other equipment of very similar quality and annual output can be had for a lot less $$, probably about 20% less.

No shade --->>> string inverter in my book.

Please clarify what you mean by "cost per kWh". That can have many interpretations.

[/FONT]

Unfortunately my daily work commute is 140 miles, which means my EV daily charge will be appox. 46 kWh. This translates into approx. 1,000 kWh per month just to maintain the car. I then add the 1,000 kWh for the normal home use and I'm north of 20,000 kWh per year.

I don't have space constraints on my 2nd story roof. I was just trying to max out the install area on the SW facing side of the roof. At 31 345W panels, I'm still slightly underpowered for my annual power needs.

I'm waiting for two more quotes since spending $35,000 to $50,000 is an overwhelmingly large investment...

Really appreciate your questions!

Thanks...

[DanKegel]

Unfortunately my daily work commute is 140 miles, which means my EV daily charge will be appox. 46 kWh.

Wow. That's past the range of your average EV, isn't it?

I don't have space constraints on my 2nd story roof. I was just trying to max out the install area on the SW facing side of the roof. At 31 345W panels, I'm still slightly underpowered for my annual power needs.

Slightly underpowered is a good place to be. Replacing Tier 3 and Tier 2 electricity
with PV is more economical than replacing Tier 1 electricity.

31 305 watt panels would probably still keep you from going into Tier 2, wouldn't they?
And they'd only cost two thirds as much.

[foo1bar]

I'm waiting for two more quotes since spending $35,000 to $50,000 is an overwhelmingly large investment...

For a 10.7kw installation, $37.5k would be $3.5/watt - which is a good price.
Of that $37.5, you will get 30% back in tax credits - about $11k. (I'm assuming you're in the US)
If your income for 2014 and for 2015 are going to be roughly equal, you can look at how much you paid in income tax last year to get an idea of whether it'll all come back to you on your 2015 taxes or if it'll require 2016 tax year as well.
Remember to look at the total income tax owed - not your refund nor the balance that is due by tomorrow (4/15) - but the total income taxes (line 47 of your 1040)
Assuming that you make enough in 2015 to owe $11k in taxes, you will have an effective price of $26.5k

And $26.5k is probably less than what you paid for that EV car - so probably not as overwhelming.

If you're considering doing it in multiple pieces, (ie. 1/3rd now, 1/3rd in a year, 1/3rd in 2 years) - it's probably going to cost you quite a bit more. And probably wouldn't get done before the Dec. 2016 cutoff for fed. credits.
I doubt that'd be a good path financially...

[J.P.M.]

Unfortunately my daily work commute is 140 miles, which means my EV daily charge will be appox. 46 kWh. This translates into approx. 1,000 kWh per month just to maintain the car. I then add the 1,000 kWh for the normal home use and I'm north of 20,000 kWh per year.

I don't have space constraints on my 2nd story roof. I was just trying to max out the install area on the SW facing side of the roof. At 31 345W panels, I'm still slightly underpowered for my annual power needs.

I'm waiting for two more quotes since spending $35,000 to $50,000 is an overwhelmingly large investment...

Really appreciate your questions!

Thanks...

12,000 + 12,000 = 24,000 kWh/yr.

So, given that range, you will probably be driving a Tesla, or have charging capabilities at work ?

If charging at work, knock 6,000 kWh/yr. off the load.

Any south facing roof ?

W/out space constraints, Sunpower is like driving a Mercedes roadster as a grocery hauler when a Toyota or Chevy is as fit for purpose. Same with solar energy. Panels are a commodity, not a lifestyle.

[thejq]

I've been looking at data on PVoutput.org and I am less sure this is advice I would agree with. There seems to be a noticeable production hit (a few percent or more) on systems that use string inverters relative to those using microinverters or optimizers, and I don't think it can be explained by location, orientation, panel selection, or data accuracy. In the real world, it appears that very few arrays are truly shade free. There are definitely cost, reliability, and obsolescence risks to choosing something other than a string inverter, but without digging into it too deeply, there does appear to be some benefit to accepting those risks. I think it is worth further study, but not on the top of my list at the moment.

Another possibility is panel to panel variation. In a string configuration without optimizer, the weakest panel (lowest voltage/current) polls down the whole string. Again, no proof, just a logical guess.

[max2015]

Slightly underpowered is a good place to be. Replacing Tier 3 and Tier 2 electricity
with PV is more economical than replacing Tier 1 electricity.

31 305 watt panels would probably still keep you from going into Tier 2, wouldn't they?
And they'd only cost two thirds as much.

Thanks Dan... I'm pursuing all alternatives right now... This is a great idea... With your suggestion, I would still have a small PG&E bill but it wouldn't be a monster every month...

JPM, you asked about the south facing portion of the roof... Our house is a wide rectangle so the south facing portion is quite small and has a chimney, which is why I was trying to max out the larger, west side with 31 panels...

I'm going to build a "Chinese menu" to run the financial comparisons between different panel wattages and the costs... Although we live in the SF Bay Area, my daily commute takes me into the Central Valley... Unfortunately 240V charging at work isn't an option and the CV doesn't have a lot of public charging options either...

Agree with your comment that panels should be commodity purchases... I'm just trying to make a significant dent in my 20,000+ kWh consumption per year!

[max2015]

12,000 + 12,000 = 24,000 kWh/yr.

So, given that range, you will probably be driving a Tesla, or have charging capabilities at work ?

If charging at work, knock 6,000 kWh/yr. off the load.

Any south facing roof ?

W/out space constraints, Sunpower is like driving a Mercedes roadster as a grocery hauler when a Toyota or Chevy is as fit for purpose. Same with solar energy. Panels are a commodity, not a lifestyle.

JPM, you asked about the south facing portion of the roof... Our house is a wide rectangle so the south facing portion is quite small and has a chimney, which is why I was trying to max out the larger, west side with 31 panels...

I'm going to build a "Chinese menu" to run the financial comparisons between different panel wattages and the costs... Although we live in the SF Bay Area, my daily commute takes me into the Central Valley... Unfortunately 240V charging at work isn't an option and the CV doesn't have a lot of public charging options either...

Agree with your comment that panels should be commodity purchases... I'm just trying to make a significant dent in my 20,000+ kWh consumption per year!

[gvl]

Bay Area is more expensive no doubt, in LA I could get quotes less than $3.10/wt for a 10.2kW array with LG300 panels and Solar Edge. I would say micros are not worth it in your situation, significantly more expensive for not much benefit.

[J.P.M.]

JPM, you asked about the south facing portion of the roof... Our house is a wide rectangle so the south facing portion is quite small and has a chimney, which is why I was trying to max out the larger, west side with 31 panels...

I'm going to build a "Chinese menu" to run the financial comparisons between different panel wattages and the costs... Although we live in the SF Bay Area, my daily commute takes me into the Central Valley... Unfortunately 240V charging at work isn't an option and the CV doesn't have a lot of public charging options either...

Agree with your comment that panels should be commodity purchases... I'm just trying to make a significant dent in my 20,000+ kWh consumption per year!

Understood. For a good job on cost anal., a decent handle on POCO rates and tariffs is needed. Without it, you're one step above back of the envelope anal. Sometimes T.O.U. can ad to the dent over tiered rates, but it takes some knowledge of both and use patterns as f(time) to get an educated guess.

On sizing: have you run PVWatts yet ?

[J.P.M.]

I've been looking at data on PVoutput.org and I am less sure this is advice I would agree with. There seems to be a noticeable production hit (a few percent or more) on systems that use string inverters relative to those using microinverters or optimizers, and I don't think it can be explained by location, orientation, panel selection, or data accuracy. In the real world, it appears that very few arrays are truly shade free. There are definitely cost, reliability, and obsolescence risks to choosing something other than a string inverter, but without digging into it too deeply, there does appear to be some benefit to accepting those risks. I think it is worth further study, but not on the top of my list at the moment.

Perhaps. But each sit. is different. I'm not saying they are a bad idea, just more like a bandaid for a less than optimal exposure. I'd suggest keeping in mind that shade will reduce an arrays output, string or micro. Micros have their uses, but they will not restore what's lost due to direct shading, or condition what's not available.