Why are turnkey Solar PV systems so ridiculously overpriced?

[nwdiver]

Do you have any suggestions on a good book/site to learn end-end? I've browsed a few things on youtube but some of the info seems dodgy.

Hmmm... oddly I can't recall any good single sources... I kinda learned the hard way as I went along...

Your best source will probably be the installation videos for each component... PV Quickmount has some great videos on how to find rafters , IronRidge has good videos on how modules rack to their rails and SMA has videos detailing the wiring of their inverters. There's also a lot of great blogs about separate topics like the benefits of oversizing.

If you have shading issues or a complex roof Enphase might be a good fit but it you've got a simple roof with a lot of sun a string inverter will save you a lot of $$$. Working with DC isn't too bad... the only rules to keep in mind are panels in series make a string and add voltage; strings in parallel add amperage. Use #10 wire for all strings and you'll be fine

 

[jeremyz]

I used option 3 for my solar install (in between paying somebody else and DIY). I found a small installer who was willing to allow me to be a helper monkey and pay him an hourly rate when he needed backfill work. I bought most of the materials myself, so he didn't have to float the materials. My installation was super easy since I have a low slope standing seam metal roof. I used Enphase micro-inverters because it seemed like the expected lifetime cost was lower. At the time, Enphase had a 25 year warranty and the monolithic inverters had 10 years at best. I also like their flexibility and better tolerance for some slight shading that I have.

I would say that most of the work is easy to do by yourself if you are technically proficient and have some knowledge of electrical work. The interconnection with your main panel is something that is probably best left to a professional. As an electrical engineer, I found working with an electrician to be interesting. We kind of speak a different dialect of the same language; you learn words like ampacity (current capacity of a conductor given its type, insulation type and environmental conditions). Electricians work by lots of rules of thumb.

 

[nwdiver]

I used option 3 for my solar install (in between paying somebody else and DIY). I found a small installer who was willing to allow me to be a helper monkey and pay him an hourly rate when he needed backfill work. I bought most of the materials myself, so he didn't have to float the materials. My installation was super easy since I have a low slope standing seam metal roof. I used Enphase micro-inverters because it seemed like the expected lifetime cost was lower. At the time, Enphase had a 25 year warranty and the monolithic inverters had 10 years at best. I also like their flexibility and better tolerance for some slight shading that I have.

For my first install I took a similar route... I had my house re-roofed and paid the roofers an extra $700 to install the feet in the new roof... I racked the panels myself (which IMO is the easy part) then I hired an electrician to wire the panels to the inverter for $1k.

For Enphase vs String inverters... I wouldn't take the warranty period as indication that they necessarily last longer... Enphase likely pairs the warranty of their inverters to the warranty of the panels intentionally... a 20 year warranty for string inverters is also available usually for ~50% the cost of the string inverter. Their life is limited by the electrolytic capacitors they each use... a component that degrades based on the temperatures it's exposed to. So a string inverter located in a climate controlled space will likely out live a micro inverter located on the roof.

 

[jeremyz]

I asked my installer about his experience with Enphase microinverters (about 6 months ago). A substantial part of his business is doing maintenance on other installers' installations. I think he said he's seen 1 or 2 Enphase micro-inverters that have failed in the field. He said that it's really common for monolithic inverters to crap out at or before 10 years. I think the steep premium for the 20 year warranty is indicative of that. If your monolithic inverter dies, you absolutely need to replace it. If one of my micro-inverters dies, I can live with the 2.9% power output loss.

https://enphase.com/sites/default/files/EnphaseElectrolyticCapacitorLife.pdf

 

[nwdiver]

It's hard to find an unbiased source of information in this debate since most manufacturers have a core business around one or the other... SMA does make both but they mostly focus on string inverters...

String vs Micro

There's very little reason to believe that micros have a durability advantage over string inverters (keep in mind that most micros aren't yet 5 years old)... they each contain the same components and perform the same function. It's no secret that heat accelerates the aging of any electronics so an inverter that is kept cool should last longer than one that commonly sees temperatures >120F. The 25 year warranty period of micros is more a reflection of necessity than reality. Nearly all string inverters carry a 10 year warranty from the generic brands to the I've heard stories from installers on both sides of the fence... some swear by Enphase while others will never use them again.

IMO it boils down to a cost issue... if you have a complicated roof with shading issue then micros are a good fit.

Another advantage the latest generation of multiple MPPT string inverters is the ability to 'share' inverter capacity. You can have 5kW facing East and 5kW facing West tied to a 6kW inverter. 10kW of micros would cost ~$7000 vs ~$2200 for a string inverter; you would save ~60% on inverter costs with almost identical production numbers...

There are also more sophisticated grid edge services that are going to become very beneficial which micros will have a difficult time squeezing into such a small device. The new Fronius inverters now offer reactive power management and line balancing...

 

[jeremyz]

That's an excellent point about the advantage of using a smaller monolithic inverter on a roof that has east and west facing panels. I have a shed roof that faces south, so I never thought of that.

History hasn't written whether or not Enphase's MTBF computations are correct or whether they will be around in 25 years to honor their warranty. You are correct that five years in the field is inadequate to test that.

In that SMA link, I can't believe that they're claiming 150 years MTBF for monolithic inverters (then they say they don't quote their own MTBF). That's not consistent with what installers have told me about their observed lifetimes in the field and it doesn't seem consistent with their 50% premium to extend the warranty from 10 years to 20 years. Enphase quotes their MTBF as 300+ years. In 25 years, I should expect 2.8 failing micro-inverters in my system which would be a loss of 8.2% power production.

 

[omgwtfbyobbq]

That's an excellent point about the advantage of using a smaller monolithic inverter on a roof that has east and west facing panels. I have a shed roof that faces south, so I never thought of that.

History hasn't written whether or not Enphase's MTBF computations are correct or whether they will be around in 25 years to honor their warranty. You are correct that five years in the field is inadequate to test that.

In that SMA link, I can't believe that they're claiming 150 years MTBF for monolithic inverters (then they say they don't quote their own MTBF). That's not consistent with what installers have told me about their observed lifetimes in the field and it doesn't seem consistent with their 50% premium to extend the warranty from 10 years to 20 years. Enphase quotes their MTBF as 300+ years. In 25 years, I should expect 2.8 failing micro-inverters in my system which would be a loss of 8.2% power production.

I believe SMA (and other manufacturers) are using MTBF in the context of their own warranty period, which is weird, but I've seen a bunch of other manufacturers do that with other products. I think a MTBF of 150 years from them means that 1 monolithic inverter out of 15 will fail within the 10-year warranty period. After that, who knows. It'd be nice if they published that info, but I doubt they will.

Mean time between failure (MTBF) is a statistical measure used to estimate likely inverter failures in a system over its warranty period.

String versus Micro – Which is the Right Choice? | Sunny. The SMA Corporate Blog

IRL, some components will likely last >20 years, and others likely won't.

http://energy.sandia.gov/wp-content//gallery/uploads/3-Component-Reliability.pdf

 

[neroden]

DIY is ~$1.20... pre-subsidy...

Yeah, but I value my labor at $100/hr *and* I don't wanna do the clambering around myself. So, no thanks on the DIY.

Thing is, for a 14 kW system, I could pay those DIY costs plus pay someone else about $2000 in labor and still be below my target price. Only that doesn't appear to be an option...

 

[strider]

I also like that the Sunny Boy string inverters have the Secure Power Supply feature which gives you a 120V/15A outlet when the grid is down. That's enough for us to run the fridge and charge flashlight/phone/etc batteries during the day without having to invest in a proper battery system. It feels like we're on the cusp (where cusp is <5 years) of a big drop in battery tech/prices and I don't want to invest in batteries to just cover the odd power outage only to have the prices drop massively in a few years. It's also something that can be added later.

 

[omgwtfbyobbq]

Yeah, but I value my labor at $100/hr *and* I don't wanna do the clambering around myself. So, no thanks on the DIY.

Thing is, for a 14 kW system, I could pay those DIY costs plus pay someone else about $2000 in labor and still be below my target price. Only that doesn't appear to be an option...

Is that pre-tax or post-tax? IME, state/federal taxes and the time/money I spend getting ready for working/commuting/ etc... really hit my hourly rate.

 

[FlasherZ]

Working with DC isn't too bad... the only rules to keep in mind are panels in series make a string and add voltage; strings in parallel add amperage. Use #10 wire for all strings and you'll be fine

And look closely at the ratings of components. I've found some people attempting to use circuit breakers in combiner boxes with grid-tie inverters on higher-voltage strings... but most DC circuit breakers are max 125 VDC (about 3 panels in series)... that's why you use fuses and fuse holders, good to 1000 VDC.

Also, from experience - remember that one side of the DC connection is grounded... if you're using a metal combiner box to support you on a high-pitch roof, it might not be good to grab an exposed conductor from the positive side of the string. Just sayin'.

 

[FlasherZ]

Yeah, but I value my labor at $100/hr *and* I don't wanna do the clambering around myself. So, no thanks on the DIY.

Thing is, for a 14 kW system, I could pay those DIY costs plus pay someone else about $2000 in labor and still be below my target price. Only that doesn't appear to be an option...

I agree with you, although if things around here are any indication, there's still a supply gap of people who know how to do this stuff and that's keeping the cost of labor high when there is enough demand for installing solar at nearly any price. It's similar to the electricians who charge twice as much for a 14-50 for Tesla charging vs. a 14-50 for a range or RV.

It really is simple with some of the tools. The big three things you need to learn are 1) using design tools to ensure that you are safely loading the structure with panels (IronRidge has a design tool that I really, really like), 2) poking holes in roofs without worry, and 3) electrical requirements and basics, like array grounding requirements.

I also view my own installs as a way to learn additional skills, so while I may value my time at a certain amount I do offset some of it as my own development time.

 

[strider]

I also view my own installs as a way to learn additional skills, so while I may value my time at a certain amount I do offset some of it as my own development time.

This is where I am. While I could afford to have someone do it I view this as a great way to spend time with my old man. He's one of those people who never went to college but just knows a ton about a lot of stuff (especially electricity/electronics - was a microwave tech in the USAF, worked on the gun guidance system for the Apache while at Raytheon, worked for Honeywell repairing mainframes, etc,etc). He's not going to be around forever so I need to soak up all the knowledge I can. And spending time with family is, as they say, priceless.

But it extends to other parts of life too. Does it take a ton of extra time to run my own web and e-mail server when I could I just use gmail? Of course. But I wouldn't learn anything (and Google would have all my e-mail ) But seriously, everyone has different things they like to do with their "free" time. If scrambling around a roof doesn't sound fun then don't do it

FlasherZ, QuickMountPV's tool looks really cool as well.

 

[ohmman]

while I may value my time at a certain amount I do offset some of it as my own development time.

It's also only fair to value your time at some level when it's something you don't take enjoyment from. If you're a project person, there's a lot of joy to be taken from doing your own PV install.

 

[neroden]

I agree with you, although if things around here are any indication, there's still a supply gap of people who know how to do this stuff and that's keeping the cost of labor high when there is enough demand for installing solar at nearly any price.

So the main problem here is, perhaps simply not enough people who have trained to install solar panels; the supply shortage allows the trained electricians to charge "what the market will bear". This is kind of what I thought...

I like doing the design. I do not like the clambering around.

 

[Topher]

Price around here is $2.85/watt or so.
One thing to remember is workman's comp for roof work.

Thank you kindly.

 

[nwdiver]

And look closely at the ratings of components. I've found some people attempting to use circuit breakers in combiner boxes with grid-tie inverters on higher-voltage strings... but most DC circuit breakers are max 125 VDC (about 3 panels in series)... that's why you use fuses and fuse holders, good to 1000 VDC.

Also, from experience - remember that one side of the DC connection is grounded... if you're using a metal combiner box to support you on a high-pitch roof, it might not be good to grab an exposed conductor from the positive side of the string. Just sayin'.

Another good tip... you don't need fuses unless you're combining >2 strings... Some panels don't need to be fused unless you're combining 4 or more strings...

The reason is simple... Solar Panels are naturally current limiting. A panel with an open circuit current of 9A really can't go higher than that since current is based on light (unless the Sun goes Nova in which case you'll probably have bigger problems)... IIRC ALL Solar panels are rated to a minimum of 15A. The fuse is to protect the string from a fault in which current is shunted back through a string. Since a string can't backed itself the most current you can see in any string with 2 strings is 9A.... BUT... with 3 strings there is the potential for ~18A since 2 strings can potentially back-feed the third.

 

[renim]

The ITC rewards highest cost installers, so in a market where there are price setters, the smaller players will price match upwards. result is what is seen in USA residential market.

but in utility scale, the price setter is gas power, so the solar market price match downwards, result is the 3.8 - 4.5c/kWh seen in USA utility market.

same subsidy, vastly different results.

FWIW, professionally installed, domestic solar for a 6kW system using chinese panels and german SMA invertor and no subsidy at all but with a favourable roof should be doable for about USD $7,200.

 

[Sharkbait]

Didn't read every post here but my 6.6 kW system after federal and state tax credits and incentives was $17,500. That worked out to $2.65/kW, installed here in California. Got top quality equipment. Saved at least another $4,000 in power bills during the last 2 years. Solar is the gift that keeps on giving, warranted in my case to 90% power output for first 10 years and then 80% through year 20.

Our federal government prevented cheaper panels from being imported by imposing a stiff import tax (up to 78%). Less costly materials could have lowered the total cost of installation. Our government accused the Chinese of 'dumping' panels in the US. I don't think the Chinese were 'dumping'; I think their labor costs are lower.

 

[renim]

Didn't read every post here but my 6.6 kW system after federal and state tax credits and incentives was $17,500. That worked out to $2.65/kW, installed here in California. Got top quality equipment. Saved at least another $4,000 in power bills during the last 2 years. Solar is the gift that keeps on giving, warranted in my case to 90% power output for first 10 years and then 80% through year 20.

Our federal government prevented cheaper panels from being imported by imposing a stiff import tax (up to 78%). Less costly materials could have lowered the total cost of installation. Our government accused the Chinese of 'dumping' panels in the US. I don't think the Chinese were 'dumping'; I think their labor costs are lower.

In my country the federal subsidy varies based upon demand (workload) and power to be produced (location, angle, etc) net result is that sunnier places have cheaper PV than cloudier places. I was careful to net out the subsidies, with subsidies the cost for solar in Australia is significantly less. Local installers become very competitive when there is something like a per kWH bounty instead of a per $$$ subsidy.

Our electrician cost is very similar to USA, so its a decent target excluding aussie/USA subsidies.