QueenBee said:
AndyH said:
Edit... Here's a quote from Homepower's Richard Perez, speaking at the Midwest Renewable Energy Fair in 1999 about modern off-grid inverters:
In the early days a good inverter would go, eh, six months, we had some that lasted as little as three weeks before they blew up. Those Days Are Gone. These inverters are rock solid. The Traces, StatPower, the Exeltechs, they rarely fail...I asked the people at Exeltech, who keep track of this stuff, what is your mean time before failure?...How many units do you have out there, how long have they been out there, and how many come back DOA 'please fix this for me'? Thirty-two years. This is phenomenal reliability. So the days when the inverters lasted a week or two and blow up in our face is long gone. ...It means you can buy it and chances are it will never break...
QueenBee said:
I don't understand what he means by "Thirty-two years" in response to the questions he asked.
It truly means exactly what it is! Exeltech makes a lot of equipment for the military - it's built like, well, a tank. What he said in the talk (which was expanded on a bit later during Q&A) was that when the top-tier companies put an inverter in the field, it's likely to stay there a very long time unless struck by lightning. There's a good possibility that the top-tier inverter manufactures use longer life capacitors - that'll likely be the limiting factor barring lightning.
It truly means exactly what it is?
what is your mean time before failure?
Thirty-two years.
How many units do you have out there?
Thirty-two years.
how long have they been out there?
Thirty-two years.
how many come back DOA 'please fix this for me?
Thirty-two years.
I certainly hope the MTBF isn't 32 years since modules are said to have a 600 years MTBF and Enphase is above 300 years. Although what that actually tells about the three products is questionable.
You might have missed this part: "...It means you can buy it and chances are it will never break..."
I understand that it can be confusing when information intended to be used in one realm leaks into another. As I said, it's a look at expected life in the field, not engineering statistics or warranty copy or marketing-speak. That was the context of Joe Perez's "32 years" statement - give the audio a listen and evaluate it for yourself. The best advice I can offer is to contact manufacturers and ask then directly: "Once I buy your inverter and install it in accordance with applicable rules/codes, et al, how many years can I expect it to continue to work as designed (barring acts of nature, war, theft, etc.)? (And validate your answers with independent folks that monitor the industry - like HomePower, the NREL, etc.)
QueenBee said:
QueenBee said:
You can call it marketing-speak but are you telling me every grid-tied inverter manufacturer I looked at on hardysolar.com is run by people that don't know how to market their products when the reality "means you can buy it and chances are it will never break"????? If so you need to get hired on as the marketing executive for any one of these companies and convince them to offer lifetime warranties, it'll be an industry first and the industry will either laugh because the company will not be in business to live up to the lifetime warranty or it'll push them to the head of the back.
OutBack: "Standard 2 year / Optional 5 year"
Magnum Energy is similiar "New Warranty Three-year warranty standard. Five-year warranty if installed on an MP or MMP panel."
MidNite is 5 years
Solectria is 5 year and also has 10 year models
SMA Sunny Boy, Xantrax, Fronius, PV Powered, Kaco is 10 years
Beacon Power is 5 years
AndyH said:
With respect, we're still talking about two different things. I've got about 10 years of industrial sales experience and study and will try to compare/contrast the 'sales' world and the 'real' world. When I say that I'm talking about a real-world expectation, I'm speaking from the position of an end user that has an inverter and is wondering if I should buy a spare to put in the attic. A warranty, on the other hand, is a marketing tool designed to provide a purchaser enough of a comfort cushion to make it easier to buy the product, but not keep the company on the hook any longer than necessary. Regardless of warranty, a good company is going to take care of the customer because the only things that'll give them a long business life are their products, their customers, and their reputation.
I have none so I'm just an outsider but from my point of view warranties are a balancing act between making the product and/or warranty affordable/cheaper and providing the customer with confidence that your product is reliable. If a car came with a 1 year warranty I'd be pretty suspect even it was significantly cheaper. If a car came with a 20 year warranty I don't think I'd want to pay what it would cost for them to make money on that transaction. So to me the fact that central inverters come with 2/5/10 year warranties is a sign that they expect failure rates to increase after those time periods and significantly enough that they can't afford to accept responsibility for the warranties. In the end it's all about perception and my perception along with a lot of other peoples is that central inverters come with shorter warranties because they are expected to have shorter lives. For example what kind of maintenance does a PV system really need? If you use the default values of http://pvcalc.org/pvcalc" onclick="window.open(this.href);return false; they say .5% per year in maintenance, so for a $10,000 system you've got $5,000 after 10 years, which would certainly pay to replace the inverter, and I have no idea what else.... Anyway my point just being that I'm certainly not the first person to have the perception that one should plan(budget) for replacing a central inverter after 10 years.
This is EXACTLY my point! The error I see in your analysis is trying to make a connection between a marketing device (warranty) and an engineering/design device (MTTF/MTBF). Those are two completely different things managed by two completely different groups within a company. I agree with your view of how this might affect the end user's perception - that's exactly why it's touted (or ignored) by sales folk. Thank you - you've just validated that sales techniques work!
I'm not at all surprised with the PVCALC info or other references that speak of 5-10 year inverter lives. They're simply based on info that isn't yet updated to reflect equipment coming off the shelf since about 2000 or so. That's the way things work when it comes to literature reviews and products derived from industry experience - it's all either older info or info that's still skewed by older equipment still in service. If an inverter fairy replaced all inverters in the field with devices manufactured since about 2010 and then updated all the literature and spreadsheets to reflect that new reality, the pvcalc numbers would be different. That's why I think papers like the one I linked from Europe better reflect the 'new reality' of modern inverters as they've been installing a very significant number of systems over the past 5-10 years and their reliability numbers though young will be skewed toward 'modern' equipment while US numbers will be skewed to a mass of 'older' equipment.
http://www.solelprogrammet.se/Globa...PDF/herman laukamp fin_rep.pdf?epslanguage=sv
QueenBee said:
Now back to reality Enphase and other microinverter manufacturers may end up really hurting if it turns out their products don't last 25 years and they are replacing more of them than budgeted.
Yes, 'reality'
I'm aware you're a micro-inverter person, and that's fine. I agree in general with your view in this sentence, yet from the perspective of someone that's been a sales professional, and from the additional view of someone that's operated a small business manufacturing, selling, and supporting tech devices, I have to remind you that the flaw in your logic is that even if a significant percentage of inverters fail, the REALITY is that only about 20% of failures will be returned to a company for replacement! That's also part of the manufacturer's calculus when designing their warranty. Again - a warranty is a marketing device - and marketing is tax deductable by businesses.
QueenBee said:
I chose Enphase microinverters the first time around and now the second time around for various reasons. Some of them being 25 year warranty is a great sales pitch that works well on me, since it's DIY the extra complexity caused by strings, string sizes, sizing the inverter(s), picking a manufacturer, figuring out how the various monitoring solutions compare, DC wiring, additional NEC requirements for DC. Then you add in that I'm heavily shaded at different times of the day/year and it makes the decision pretty easy. Plus I love seeing the individual output of each panel. Now had I been off the grid most of the above reasoning doesn't matter any more and the important fact that it needs to work off the grid makes Outback come to the top.
I can certainly respect this! While I personally find string sizing, calculating I2R losses, etc. to be a 'fun' part of the process, I realize not all solar folks are geeks/nerds/electronics hobbyists.
All of my monitoring is handled by the charger controller, the inverter, and the TriMetric and I can simply walk down the hall to check the numbers, so I don't feel the need for a web interface to check on how each panel is performing. I'd say that your shading challenges are the number one reason to either split an array, use per-panel MPPT devices, or microinverters - even if they result in a more expensive system.
I think it's useful to point out that any inverter installed in a 75°F space at ~45% humidity is going to have a much longer life than if that same device is installed on a roof between shingles and a solar panel - that's tough duty for an electronic device. As I see it, the corollary is that a device designed to go under a panel MUST use higher quality components if it's to have a long service life. That's covered in one of the studies I linked above - there's been a LOT of progress in all inverters in the past 10 years.
A related observation/marketing example: A marketing division will tout that they use military grade hardware or long-life capacitors or don't use optical isolation and their competition does. This is a case where facts are used out of context to spin an almost-lie. If one designs an inverter to live inside a living space and last 10 years, they can use low-grade capacitors - they'll provide more than enough life to meet that goal. Those same capacitors will NOT last 10 years when put on a roof under a solar panel where it might be 160°F in the summer. To last for the same 10 years, a microinverter manufacturer MUST make different choices in order to have the same reliability. Marketing will spin 'higher quality components' to their advantage to lead or cause a consumer to assume better components equals longer life. (That might be true if the microinverters are installed inside next to a central inverter, but that's not how they're intended to be used.) The marketing folks do that on purpose, and will NOT 'correct' the customer that develops the false assumption.
QueenBee said:
AndyH said:
Whatever you use, install good surge suppressors. (I promise - I'm not affiliated with Midnite in any way. My first small grid-tied system used Delta suppressors.
)
I came across that second video when I was looking up their warranties and was pretty shocked. It made me really wonder if the GE "breaker style" surge suppressor I used is adequate enough. If a "name brand" like Delta is selling a product that is performing so poorly then I don't have much confidence in my GE one. Looks like I'll have to order one of their SPDs for my second system. Thanks for sharing this!
Absolutely! I've used a number of different whole-house suppressors and found out the hard way that they didn't live up to the sales-speak claims. I strongly suggest that anyone with questions should bypass sales/marketing and go directly to engineering. It's sad, but sales divisions take great pride in bending the truth almost to the point of breaking when writing copy, and some sales folk outright lie. I put plenty of faith in info from the design/engineering folks but zero in warranties or sales info. FWIW.