Tinkeringcaveman said:
I read this thread with interest as I have a solar array with 16 enphase microinverters in Troy, Illinois. I installed the array in May/June 2011. As of June 2013, I have had 5 microinverters fail. I believe I have the M190 inverters.
Here's a web page I use to track my system (I haven't updated it lately):
https://sites.google.com/site/tinkeringcavemansolarproject/" onclick="window.open(this.href);return false;
Here's a link to the enphase site:
https://enlighten.enphaseenergy.com/public/systems/JSQ217175" onclick="window.open(this.href);return false;
Welcome, and thanks for the heads-up on your system, Tinkeringcaveman! You have the second-highest rate of failures that I have found to date. Please keep us informed as you learn more.
Tinkeringcaveman said:
It does not appear that this site was used in the above-reference study of inverter failures.
No, this study is not all-inclusive by any means. The data-mining is done manually, so I can only track a relatively small number of sites.
I have added your system to the spreadsheet and will track it, but I should note that it is the ONLY site included so far that was self-selected because of a high failure rate. All others have been added because they were "discovered", regardless of the number of failures or lack of failures. While this certainly corrupts the MTBF calculation, I have all but given up on using that number (as I originally intended) since the failure rates between sites vary so widely. For instance, the MTBF for my system currently sits at 105 years, while yours currently sits at only 7 years. Another system near here has a failure rate of 387 years (although it is not roof-mounted). The failure rate for all of the M215s is currently unknown, since none have failed within the 345 device-years that I have tracked so far. However, these systems are younger, so they have not yet experienced as many thermal cycles as the M190s. In any case, I need to modify my spreadsheet so that I can track those separately.
Tinkeringcaveman said:
Yesterday, I did some spot checks into the inverter-reported temperatures for last summer. I looked at one of the failed modules, and one that has been operating without any problems. I would like to spend more time this week compiling data.
In both cases I examined, microinverter temperatures climbed about 150 degrees on a small number of occasions during the hottest weather. On one day, the failed inverter registered 158 degrees, while the ambient reported air temperature for the local weather station was 108 degrees.
The M190s are rated for up to 65 degrees Centrigrade--149 degrees Fahrenheit. My preliminary review of just those two modules identified seven days in 2012 when the inverter temperatures met or exceeded the design temperature.
Thanks for the temperature data. While I agree that your peak temperature has exceeded the rating of the M190 microinverters, it is far from clear that this would result in an immediate or even imminent failure. IMO, it is more likely this rating is needed for Enphase to be able to guarantee a 15-year life for the modules. The electrolytic capacitors are the most heat-sensitive components used in the M190s are rated for operation for 8000 hours at 221F. At all lower temperatures, their rate of degradation will be reduced, meaning even longer life (up to 15 years, according to Nichicon, the manufacturer).
IMO, a more likely explanation is related to either the site or some issue with a defect in the batch of inverters installed at a site. One site I am tracking has two failures AT THE SAME LOCATION, which makes me believe the site is the issue. Possibly lightning combined with installation details is involved, but I do not know.
Tinkeringcaveman said:
BTW - the lower performing module you mentioned is a damaged PV module- the glass has been shattered, but it has steadily produced 80% or so of its rated output since I shattered it during installation.
O.K. I was going to ask the same question about that one as drees did. But I have a similar question: Why does the microinverter fourth from the left on the top row produce MORE electricity than all the others? I don't see any obvious shading of the other modules. Does it have a higher-power PV module mounted above it than the other microinverters? Or perhaps it sits in front of a window that reflects more sunlight onto the PV module?
drees said:
Tinkeringcaveman said:
I have more work to do, but still preliminarily believe heat is the cuplrit.
If this was the case, we should be seeing a lot more failures in Arizona.
Agreed. In other words, the premise of the thread starter has proven to be false. While there are a couple of failed inverters in the Phoenix area, I have found many more failures in the immediate vicinity of where I live.
drees said:
QueenBee said:
Hear's an early release of the new grounding guidance.
http://www.ecodirect.com/Enphase-M250-Micro-Inverter-s/856.htm" onclick="window.open(this.href);return false;
That's a great write-up!
Thanks for all the information on the M250s, guys!
Unfortunately, all the various grounding schemes simply add to the confusion installers and inspectors have about what is best. This is a challenging topic which requires a certain amount of faith in what is communicated by the NEC, the manufacturers and the inspectors to accept. While some inspectors will be open to new ways to do things, others may not be.
megalo said:
Maybe they'll have a firesale on the M215s now...
That was exactly my first thought! I have an old off-grid array which would do nicely with 12 M190s (since they are 72-cell modules). I would love to upgrade 14 of my M190s to M215s and move 12 of the inverters over to this other array. At some point, swapping out these inverters might make sense.