Enphase inverter MTBF calculation thread

RegGuheert
RegGuheert Solar Expert Posts: 102 ✭✭✭✭
The purpose of this thread is to attempt to create an independent calculation of the Mean Time Between Failures (MTBF) for Enphase inverters operating in the field. If you are unfamiliar with the concept of MTBF, please read Enphase's Reliability White Paper on this topic. Enphase is currently claiming an MTBF of 331 years.

I believe that there is a fairly significant and growing number of Enphase inverters represented by the posters here, so the idea is to create a list of those inverters and calculate our MTBF or at least the amount of time we have operated without any failures. I will start the list in this post, but if I did not include your system, please post and provide a link to your system. We can then update the tally occasionally to find out how things are going.

Of course we also need to keep track of failures. Since Enphase claims that infant mortalities are done after two or three weeks, let's be generous and ignore all failures (and not accumulate time) during the first month in service for all inverters. But beyond that, please report ANY failure that has resulted in the need to replace an inverter.

So here are the systems that I know about so far on this site (I tried to create a table, but this site does not seem to support tables):

System Link|Quantity|Service Entry Date|Years|Total Years|Failures
RegGuheert|42|01/2011|0|0|0
mr.radon|45|10/2010|0.25|11.25|0
Peter V|15|06/2010|0.5833|8.75|0
Ralph Day|52|07/2010|0.5|26|0
drees|18|04/2010|0.666|12|0
neelix|16|06/2010|0.5833|9.33|0
jcgee88|15|07/2010|0.5|7.5|0

(Of course we may need to drop some of these systems if the posters never return and we have no way to find out if there were any failures.)

So far that comes to an MTBF of greater than 75 years, assuming that none of us have had any failures so far. Not bad. Let's see what it comes out to with more data!
«13

Comments

  • System2
    System2 Posts: 6,290 admin
    Re: Enphase inverter MTBF calculation thread

    6|2010/.41|?|0
    I have 6,installed 8/26/10, none failed
  • Peter_V
    Peter_V Solar Expert Posts: 226 ✭✭✭
    Re: Enphase inverter MTBF calculation thread

    So far so good, no failures yet.
  • Ralph Day
    Ralph Day Solar Expert Posts: 1,019 ✭✭✭✭
    Re: Enphase inverter MTBF calculation thread

    My installer had one inverter failure out of about 200 inverters installed. Not one of mine.

    Ralph
  • RegGuheert
    RegGuheert Solar Expert Posts: 102 ✭✭✭✭
    Re: Enphase inverter MTBF calculation thread

    Interesting, Ralph! Do you know if that inverter failed within the first month or sometime after that?

    With zank's six, we now have a total of 209 Enphase microinverters between us. As such, if Enphase's MTBF calculations are anywhere near correct on average we should expect to have about one of these 209 microinverters fail every 18 months or so.
  • Ralph Day
    Ralph Day Solar Expert Posts: 1,019 ✭✭✭✭
    Re: Enphase inverter MTBF calculation thread

    Reg
    I think the failure was within the first 2 months, not far along. That person has 3 microFITs installed on his farm, all identical to mine (almost, the second and third ones have the big steel within the concrete pad, no anchors and nuts)

    Ralph
  • SolarBuff
    SolarBuff Registered Users Posts: 2
    Re: Enphase inverter MTBF calculation thread

    You can add me to the accumulation. I have had two Enphase Inverters working since June 2010.
    SolarBuff:
  • soylentgreen
    soylentgreen Solar Expert Posts: 111 ✭✭
    Re: Enphase inverter MTBF calculation thread

    I don't have a dog in this fight, just a curious bystander thinking about solar for my home, and like others, trying to make the decision between regular inverters or microinverters.

    I'm having trouble reconciling what I read in this thread (which seems to suggest MTBF is not noticeably worse than published specs) and other anecdotes such as this one:
    Of the several dozen systems we have installed with micro-inverters, at the request of the customer, ALL have had inverter failures. This makes sense and why we were reluctant to adopt the technology for all customers
    which comes from here:
    http://www.sullivansolarpower.com/about/solar-power-blog/daniel-sullivan/dangers-of-micro-inverters who claim to have 1000s of installations overall, so if they are seeing problems with "all" of the microinverter installations, that sounds like a big deal to me. But that claim is hard to square with the data in this thread...
  • fix_it_guy
    fix_it_guy Solar Expert Posts: 26
    Re: Enphase inverter MTBF calculation thread

    Hopefully next year I will have an Enphase based system installed so I'm curious about this thread as well. This year we are weatherizing and updating.
  • Ralph Day
    Ralph Day Solar Expert Posts: 1,019 ✭✭✭✭
    Re: Enphase inverter MTBF calculation thread

    Personally, I'd never bury inverters under the panels on a roof...replacement would be problematic to say the least. Groundmounted, no problem, but removing panels safely, without breakage, that's not a DIY project I'd want to tackle.

    Ralph
  • Blackcherry04
    Blackcherry04 Solar Expert Posts: 2,490 ✭✭✭
    Re: Enphase inverter MTBF calculation thread

    Personally, I have 35 of them, 2 years never a issue except for 22 MWH of electricity. I moved a panel right from the middle of the array that was shaded by a stand pipe, 4 screws with a nut driver it was a done deal in 15 minutes, it's hard to see a problem to replace a inverter. Sounds more head in the sand from naysayers. Anyone want a buggy whip ??
  • ggunn
    ggunn Solar Expert Posts: 1,973 ✭✭✭
    Re: Enphase inverter MTBF calculation thread
    RegGuheert wrote: »
    Interesting, Ralph! Do you know if that inverter failed within the first month or sometime after that?

    With zank's six, we now have a total of 209 Enphase microinverters between us. As such, if Enphase's MTBF calculations are anywhere near correct on average we should expect to have about one of these 209 microinverters fail every 18 months or so.
    I don't see how you are going to derive any useful information from this. You are comparing small sample size data with large sample size data. Even if you get 2X or 3X the failure rate you are expecting from the published data, it really doesn't mean anything.

    If you get 100X, then you may have something. :D
  • boB
    boB Solar Expert Posts: 1,030 ✭✭✭✭
    Re: Enphase inverter MTBF calculation thread

    Let's all check back in 330 years or so and see how many of them are still working.

    See you all in 2342 !!

    boB
  • RCinFLA
    RCinFLA Solar Expert Posts: 1,484 ✭✭✭✭
    Re: Enphase inverter MTBF calculation thread

    Just to add to the mix, newer Enphase units has dumped the controversial large low voltage electrolytics across the panel voltage in favor of the industry trend of putting non-polarized caps (with higher reliability record) on the high voltage booster. So you need to track the particular model.

    A reasonably large storage cap must be placed somewhere in the GTI system for smoothing the power flow (DC from the panel) in a single AC phase implementation (smooth out the AC 120 Hz ripple power). Allowing some of this 120 Hz ripple current to show up at the panel reduces panel production.

    Three phase GT inverter systems (true three phase GTI, not three separate single phase inverters) are easier since the AC power flow is almost constant so the large storage caps are not necessary. So there is another variant to keep track off.
  • Blackcherry04
    Blackcherry04 Solar Expert Posts: 2,490 ✭✭✭
    Re: Enphase inverter MTBF calculation thread

    While I have Enphase Inverters , I am really not a fan boy. I also have 4 other Inverters. 2 Xantrex, 1 Magnum and a Mean Well. They all have their quirks and I have had one of the Xantrex back to the service center twice and it still has a issue they cannot seem to be able to correct.

    Enphase has a answer for the naysayers, people can take it for what's worth. Self serving ?? yeah, I guess. When I put the Grid Tie in I bought 2 extra Inverters for $120.00 each, they are still in the box.


    http://enphase.com/wp-uploads/enphase.com/2011/03/Electrolytic_Capacitor_Expert_Report.pdf
  • Rick1
    Rick1 Registered Users Posts: 24
    Re: Enphase inverter MTBF calculation thread

    46 Enphase M215 installed 04/10/2012 no failures as of 09/01/2012, just 8.4 mw of power.
  • solarix
    solarix Solar Expert Posts: 713 ✭✭
    Re: Enphase inverter MTBF calculation thread

    My 37 years of electronics says that Emphase inverters will start dropping like flies after about five years. See you in 2015.
  • niel
    niel Solar Expert Posts: 10,300 ✭✭✭✭
    Re: Enphase inverter MTBF calculation thread
    solarix wrote: »
    My 37 years of electronics says that Emphase inverters will start dropping like flies after about five years. See you in 2015.

    and in the spirit of the late johnny carson the question would be, why would they move to china?:roll:
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: Enphase inverter MTBF calculation thread

    Maybe seven to ten years. I am not sure that most of the inverter's are running very hot under the solar panels. Also depends on the quality of their components, design, and manufacturing.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • stephendv
    stephendv Solar Expert Posts: 1,571 ✭✭
    Re: Enphase inverter MTBF calculation thread

    They come with a 25 year limited warranty, so I would have thought that their engineers and actuaries would have done their homework and established that the MTBF was well above 25 years... no?
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: Enphase inverter MTBF calculation thread

    I have tested many a computer hard drive in decades past with 5-10 year design life (according to the manual) to destruction in just 2 weeks with an environmental chamber cycling 2 cycles per day between 40F and 121F (as I remember). This was not even HALT (highly accelerated life testing).

    The bad designs failed in 2 weeks or less, the good designs did not fail in 4-6 weeks of cycling.

    Hopefully their new design (without the electrolytic capacitors) will last longer. But beyond 10 year life in a harsh environment will be difficult for typical commercial products.

    Wish them the best of luck. I have never seen an Enphase device--So it is purely a guess on my part. But even my new and improved Xantrex GT inverter failed inside of a year (and my BP Solar panels failed in less than six). Fortunately, both vendors did honor their warranties.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • RCinFLA
    RCinFLA Solar Expert Posts: 1,484 ✭✭✭✭
    Re: Enphase inverter MTBF calculation thread

    I happened to see an episode of "how's it made" on Discovery channel. It was on Smart Meters. It showed their assembly with three PCB's. One thing that surprised me was there was about five regular aluminum oxide electrolytic surface mount caps reflowed on one of the PCB's. It is going to be interesting to see how these meters survive in the long run.

    My smart meter is mounted on the outside southwest wall of the garage and it get baked by the sun each afternoon.
  • kevin
    kevin Registered Users Posts: 9
    Re: Enphase inverter MTBF calculation thread

    i think it very difficult to realize 25 years. you know,life of most component is 100000 hours. but of the CAP is very short. especially electrolytic capacitor, perhaps the famous company can promise the life is 500000 hours. if in hot surrounding,the life will shorter. we can calculate 10 hours for 1 day. so 50000/10/365=13.69863013698630136986301369863 years!
    this is just my own opinion,if they can promise,you can have a try!
  • RegGuheert
    RegGuheert Solar Expert Posts: 102 ✭✭✭✭
    Re: Enphase inverter MTBF calculation thread
    RCinFLA wrote: »
    Just to add to the mix, newer Enphase units has dumped the controversial large low voltage electrolytics across the panel voltage in favor of the industry trend of putting non-polarized caps (with higher reliability record) on the high voltage booster. So you need to track the particular model.
    That's interesting. I haven't read that anywhere else. That would explain how they have managed to increase the warranty to 25 years on the M215s. Do you have any links to information about this change?

    Anyway, here is a new Google doc containing the latest update on Enphase inverter MTBF:

    https://docs.google.com/spreadsheet/ccc?key=0AqhUVOciAXVhdEJRM2xXSFVZY3Vnak53ZFAyMHhuc3c

    I am trying to track the inverter type, but I have not (yet) calculated MTBFs separately. There are no failures of M215s shown to date.

    Anyway, there appear to be about 4 failures to date after over 1200 device-years of operation, (only 3 if you don't count the likely infant mortality of one of mine). Interestingly, the three other failures are in AZ, two of them on an array of 10 inverters owned by a poster here.

    If you have an array which is not included here, please provide a link for it an I will add it. If your array is included, please check that all of the information is correct, including failures.

    Unfortunately, I will not be able to track this much longer as the new public websites will not allow failures to be seen.
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: Enphase inverter MTBF calculation thread

    We have had at least one report of failures by "boB"--Who is "in the solar industry":
    boB wrote: »
    Either one has a good opportunity to fail because they are mounted outside in fairly harsh hot-cold-hot-cold temperature cycling environment.

    If there are more optimizers on the system than there would be micro-inverters, then statistically, I would expect an optimizer to fail first.

    I do know a lot of the older Enphase micro-inverters have failed. Problem is that when one or the other does fail, it is harder
    to replace because you have to go back up to the roof to troubleshoot the problem.

    boB

    Don't know anything more that the above.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • RegGuheert
    RegGuheert Solar Expert Posts: 102 ✭✭✭✭
    Re: Enphase inverter MTBF calculation thread
    BB. wrote: »
    We have had at least one report of failures by "boB"--Who is "in the solar industry":

    Don't know anything more that the above.

    -Bill
    Thanks, Bill!

    boB, do you have links to the public websites for any Enphase arrays that I can add into the spreadsheet? TIA!
  • Rick1
    Rick1 Registered Users Posts: 24
    Re: Enphase inverter MTBF calculation thread

    Update

    04/02/2013 16.7 MWHs produced from the 46 enphase M215 microinverters. No failures. produced 81.9 KWHs today.
  • Blackcherry04
    Blackcherry04 Solar Expert Posts: 2,490 ✭✭✭
    Re: Enphase inverter MTBF calculation thread
    RegGuheert wrote: »
    Thanks, Bill!

    boB, do you have links to the public websites for any Enphase arrays that I can add into the spreadsheet? TIA!
    Public Websites, how about a list of all of them ?? This use to come up with a list of all the links you could copy and paste, now it's a interactive map. It takes a long time to load, anyway is what it is. Click on the blue, then to " E " logo to system level. If it's a Public System in the list it's underlined and you can click on it. There is probably a way to export the data, but it's over my head.

    http://enlighten.enphaseenergy.com/all_public_systems

    Here is another list with Public sites, no data.

    http://www.batchgeo.com/map/b156ed39f7dcc5977507bda00d61ebf6

    This guy might be able to tell you how to get the data list of public sites, he has twitter, he keeps up the map above.

    http://community.enphaseenergy.com/people/casey_whiteley
    .
  • RegGuheert
    RegGuheert Solar Expert Posts: 102 ✭✭✭✭
    Re: Enphase inverter MTBF calculation thread
    Public Websites, how about a list of all of them ?? This use to come up with a list of all the links you could copy and paste, now it's a interactive map. It takes a long time to load, anyway is what it is. Click on the blue, then to " E " logo to system level. If it's a Public System in the list it's underlined and you can click on it. There is probably a way to export the data, but it's over my head.

    http://enlighten.enphaseenergy.com/all_public_systems

    Here is another list with Public sites, no data.

    http://www.batchgeo.com/map/b156ed39f7dcc5977507bda00d61ebf6

    This guy might be able to tell you how to get the data list of public sites, he has twitter, he keeps up the map above.

    http://community.enphaseenergy.com/people/casey_whiteley
    .
    Wow! Thanks!

    I didn't know about that! That is particularly useful for anyone wanting to know what the MTBF looks like in their local area. When I have some time I might grab some for my area and a couple of other locations and see if I can spot differences in MTBF in different climates.

    Thanks again!
  • RegGuheert
    RegGuheert Solar Expert Posts: 102 ✭✭✭✭
    Re: Enphase inverter MTBF calculation thread
    Rick1 wrote: »
    Update

    04/02/2013 16.7 MWHs produced from the 46 enphase M215 microinverters. No failures. produced 81.9 KWHs today.
    Thanks, Rick1!

    Can you please provide the public URL for your array? I prefer to use the URL so that I can update it independently. TIA!
  • andyman
    andyman Solar Expert Posts: 32
    Re: Enphase inverter MTBF calculation thread

    I'd attach a picture if I knew how but my enphase M190 exploded. Installed in 2011. So MTBF ? 2 years roughly. I also replaced 2 other inverters already. When I say exploded, I mean the case split in half and it ripped one of the bolts apart. Black dust under my panels and thru one of the holes in the inverter.