Enphase field MTBF: M190: ~36 Years M215: ~316 Years M250: >357 Years

[madbrain]

Doh, you're right. I wonder if I saw that somewhere else or if I'm just imagining things...

I don't think you are imagining it, I think Enphase really is doing it in some cases, based on what they told me on the phone. But it's up to their "escalation team" to decide.

Edit: I subscribed to the thread and got a message about someone who got his D380s replaced with M215. The post was deleted from the Enphase board today, probably by Enphase. They probably don't want to offer this option to everybody. I won't be too upset if they don't for me, as long as they make good on providing replacements and paying for labor.

You'd have to lose either 4 M190s or 3 M215s.

I wonder if Enphase can software limit the maximum power with a firmware update so that the M215s are limited to the same maximum power as the M190 - 0.8A nominal instead of 0.9A nominal)?

No idea. Maybe they can. But it probably still wouldn't pass inspection.

Do all of your panels point in the same direction? I have heard of people succesfully arguing for a lower limit circuit breaker so they could install more panels/inverters since not all of them would be hitting peak power at the same time so there would be no issue of nuisance trips - but it looks like all your panels are facing the same direction right now.

Not exactly the same direction, the ones on the D380 are at an angled, while the ones on the M190s are more flat. But they still tend to peak around the same time.

Heat could be an issue - but I doubt it's the real reason your D380s have failed. There really seems to be some sort of issue with the D380s - just by reading the Enphase community board, the D380s seem to have the highest rate of failures. M190s seem to be generally good unless you happened to get yours out of a bad batch where they can also have high failures rates. The M215s seem to be very good with only very rare failures.

OK, that's good to know that it's probably not my install. The temperature doesn't look excessive when I have monitored it on the D380s, even at the time of the intermittent 0W events.

The M215s are fairly new, perhaps too new to be able to accurately assess the failure rate.

 

[RegGuheert]

The M215s are fairly new, perhaps too new to be able to accurately assess the failure rate.

Perhaps, but note that some of the M190s failed within a year of installation. My one failure was within about six months. But I have not yet found a failed M215. Some have been out there for over two years. I may try to add some more system to the spreadsheet this weekend to see if I can find an M215 failure.

We won't know the LIFE of any of these inverters for many more years (hopefully).

 

[DaveEV]

I don't think you are imagining it, I think Enphase really is doing it in some cases, based on what they told me on the phone. But it's up to their "escalation team" to decide.

Edit: I subscribed to the thread and got a message about someone who got his D380s replaced with M215. The post was deleted from the Enphase board today, probably by Enphase. They probably don't want to offer this option to everybody. I won't be too upset if they don't for me, as long as they make good on providing replacements and paying for labor.

Yep, I saw the same message which was later deleted by Enphase.

 

[RegGuheert]

OK, just mentioning it because there have been people who have replaced all their D380s with M215s after experiencing a high rate of failures. Someone just reported it today - they are having all their 25 D380s replaced after experiencing 5 failures after 2.5 years.
http://community.enphaseenergy.com/enphase_energy_community/topics/my_190s_are_failing_at_an_alarming_rate_is_anyone_else_experiencing_this_problem#reply_12797151" onclick="window.open(this.href);return false;

Thinking about this a bit, it makes sense that Enphase will need to take in M190s and D380s at the same rate as they replace failed ones. How do they do this? I don't think they manufacture them. They would take the worst sites and offer to upgrade all microinverters to the 215s and put the "good" M190s (or D380s) that they get back into stock to replace those who are experiencing normal failure rates and only need one or two. I think this is a relatively low-cost way to deal with the M190 issue. So I would push them and suggest that you have a bunch of "hardly used" D380s that you would be willing to trade for M215s if they will only pay for the swap!

Since time is running out on this experiment, I have added a bunch of new installations to the Enphase Microinverter Field MTBF Estimate spreadsheet. All of the new installations are in Phoenix, AZ. Interestingly, the M190 MTBF in Phoenix is not worse than here, with only one really bad small one at 5 years MTBF (versus two like that here!), but with several larger, older sites with no failures. In fact, the overall MTBF for the 190s went UP from 55 years to 59 years just by adding Phoenix locations. Again, I conclude that heat is not the big issue for these inverters (yet, anyway). Something else is killing them, methinks.

Also, I still have found no failures of M215s. MTBF is over 900 years for the M215s at this point. So far, so good. I will update the thread title...

 

[madbrain]

Thinking about this a bit, it makes sense that Enphase will need to take in M190s and D380s at the same rate as they replace failed ones. How do they do this? I don't think they manufacture them. They would take the worst sites and offer to upgrade all microinverters to the 215s and put the "good" M190s (or D380s) that they get back into stock to replace those who are experiencing normal failure rates and only need one or two. I think this is a relatively low-cost way to deal with the M190 issue. So I would push them and suggest that you have a bunch of "hardly used" D380s that you would be willing to trade for M215s if they will only pay for the swap!

Yes, this might be what they are doing if they actually stopped all production of M190s/D380s.
Enphase today offered to replace my entire site with M215s, not just the failed modules. Since they are local to me, it will be their own crew doing it, not one of my 2 installers. This way there won't be any issue with labor reimbursement. I expect to hear back after labor day to have them come for a site review. I mentioned the issue with my max panel load on the phone, but was told this is something their crew would look at during site review. I'm impressed that they are willing to go to this length. I didn't expect they would want to replace more than just the failed units.

I will see how the main panel issue gets resolved. I am certainly willing to chip in a little bit to get 400A feed. It's not just for generation that I want 400A. I already have some pretty high loads. If I were to run every appliance I got already at once, I would likely already be over 200 amps peak, though this is unlikely to ever happen. Some day my EVSE will be higher than the current 15A, and there may be 2 of them. And I may add a wet sauna too. So the solution to use a different type of 200A panel and reduce the main breaker (to, say, a 175A breaker) is not particularly appealing, I would rather have some more room than less.

The highest kWh I have ever seen on my smartmeter is 17 kWh for a 1 hour period which is 70A, and the sun was shining a little too, so actual power usage was a bit more. But that's the average usage over an hour, not the peak which could have been more.
And yes, I was actively trying to generate load.

A few days ago I bought a TED 5000-C (energy detective). It's not been installed yet. The goal is to try to get rid of all the idle vampire power, which is about 1 kW when sleeping. I notice the TED 5000 has a monitoring limit of 200A. If I get my main panel upgraded, this may be an issue. I wonder if I can keep my current 200A main panel as a subpanel of a new 400A main panel which would have both the grid and solar generators hooked up.

 

[RegGuheert]

Enphase today offered to replace my entire site with M215s, not just the failed modules.

That's great news! Based on what I see, the M215s are very reliable.

Since they are local to me, it will be their own crew doing it, not one of my 2 installers. This way there won't be any issue with labor reimbursement.

It also allows them to see if there are any installation issues. I think this is a very good thing!

I expect to hear back after labor day to have them come for a site review. I mentioned the issue with my max panel load on the phone, but was told this is something their crew would look at during site review.

Personally, I wouldn't sweat the difference between M190s and M215s unless you need to pull a permit or get the utility involved. Since you are not changing the PV panels, the difference will be very small and will only occur in cold weather. Plus, the new panels on the roof are fairly flat from your description, so you may not even see a difference with them (since they should not point directly at the sun in wintertime).

I would deal with your other load issues as a separate topic.

 

[RegGuheert]

The old Enlighten sites were still up this morning, so I decided to update all the data for the sites in the spreadsheet and add more sites. I added the public sites I could find from Petaluma, CA, (where Enphase is headquartered) and Santa Rosa, CA, just up the road.

This effort yielded the following:

- Virtually no change to the M190 MTBF: Still ~60 years.
- I have found a failure of an M215 for the first time. MTBF now stands at around 1300 years.

I will update the OP and thread title.

 

[RegGuheert]

As of this morning, the old Enlighten public sites containing per-module production data are no longer available. As a result, I will no longer be able to see failures in most of the systems that I was tracking.

I may ping everyone here occasionally to see if there have been any failures with our personal systems. Please feel free to post here if you experience any failures in the future.

 

[RegGuheert]

It appears that we now have our second microinverter failure. I have contacted Enphase for a replacement. This drops the measured MTBF for our system down from 134 years to 68 years, which is pretty close to the overall average, though well below Enphase's predictions for the M190s. Unfortunately this inverter will not be as simple to replace as the previous one which failed.

If your system is recorded in the Enphase Microinverter MTBF Estimation Spreadsheet, please let me know how many failures you have had since the last date I updated the list (or just tell me the total number of inverter failures you have ever had). Since Enphase updated Enlighten on September 3, 2013, I have been unable to observe any failures except those within my own system. Therefore self-reporting by owners is the only way I can update the list of failures. TIA!

 

[Weatherman]

In less than two years, I've had two, complete failures, out of the set of 30, M190s.

I have about a half dozen others, which have problems with periodic power loss (power output drops to zero for anywhere from a few minutes to several hours during the day). Enphase has updated the software on a number of the inverters, but now they erroneously register "critical temperature" and reduce power during the day. Enphase claims to have additional software updates in the works, but no estimate on the distribution date. They refuse to provide replacements until the inverters are completely dead.

So, to summarize... two complete failures and a half dozen or more with software problems for a failure rate of close to 30%.

 

[QueenBee]

In less than two years, I've had two, complete failures, out of the set of 30, M190s.

I have about a half dozen others, which have problems with periodic power loss (power output drops to zero for anywhere from a few minutes to several hours during the day). Enphase has updated the software on a number of the inverters, but now they erroneously register "critical temperature" and reduce power during the day. Enphase claims to have additional software updates in the works, but no estimate on the distribution date. They refuse to provide replacements until the inverters are completely dead.

So, to summarize... two complete failures and a half dozen or more with software problems for a failure rate of close to 30%.

That's disappointing

Although our climate is more temperate but all of my "first gen" M215s are still going strong. The first 21 are 2.5 years old and the second 37 are a year old now. 15 more of the "second gen" M215s are going up sometime in the next couple weeks.

 

[DaveEV]

Still at zero M190 failures here after 4 years.

It seems there was a bad batch of M190s shipped in the last couple years somewhere with some sites having failure rates high enough to warrant replacing all inverters. The M215 doesn't seem to have anything close to the failure rate of the M190s.

I personally wish that Enphase made a version of the M215/250 with the same cable as the M190.

 

[RegGuheert]

Although our climate is more temperate but all of my "first gen" M215s are still going strong. The first 21 are 2.5 years old and the second 37 are a year old now. 15 more of the "second gen" M215s are going up sometime in the next couple weeks.

Thanks. I have updated the line for your system to reflect no failures to date. As of today, your system MTBF is >87 years.

Still at zero M190 failures here after 4 years.

Thanks. Updated. With my recent failure, your system's measured MTBF, at >73 years, has surpassed mine.

It seems there was a bad batch of M190s shipped in the last couple years somewhere with some sites having failure rates high enough to warrant replacing all inverters.

My spreadsheet shows three M190 sites with MTBF of 5 years and one with MTBF of 6 years. That's a far cry from the 331 years on the datasheet!

The M215 doesn't seem to have anything close to the failure rate of the M190s.

According to my calculations, the M215s have an MTBF that is over 20X higher than that of the M190s.

I personally wish that Enphase made a version of the M215/250 with the same cable as the M190.

But I wonder if the failures in the M190s could be related to the cabling system. Maybe not, but that IS one of the changes between the M190s and the M215s. More likely the failures in the M190s are due to a solder joint or component which is failing due to thermal cycling. (And I really don't think the electrolytic capacitors are the main culprits at this early stage in the game.)

 

[RegGuheert]

In less than two years, I've had two, complete failures, out of the set of 30, M190s.

Thanks for posting this. Since I do not have any systems on my spreadsheet from FL, I'm guessing yours is not in my spreadsheet. Would you mind posting the public URL for your system so that I can add it in?

I have about a half dozen others, which have problems with periodic power loss (power output drops to zero for anywhere from a few minutes to several hours during the day).

That is what I see with the inverter which recently started exhibiting problems. Here is how I described the issue to Enphase:

This inverter started exhibiting problems on March 30 when it produced 40 Wh of energy while its eight neighbors produced between 175 and 186 Wh. On March 31, it produced 1.39 kWh while its neighbors produced between 1.58 and 1.63 kWh. Today, April 1, it produced 1.05 kWh while its neighbors produced between 1.23 and 1.27 kWh. This low production is due to periods when this inverter produces at very low power levels while the others are producing substantial power.

To me, it appears to be a solder joint which has failed under repetitive thermal stress and it does not make electrical contact until the power levels get fairly high. Then it works for the remainder of the day. Sorry, but producing less than 22% of its neighbors is a defect.

Enphase has updated the software on a number of the inverters, but now they erroneously register "critical temperature" and reduce power during the day.

Thanks for the heads-up! They updated the firmware in my microinverters in January 2012 when the first inverter was replaced and the system has worked fine since that time. I have unplugged the Envoy ethernet cable from the wall to prevent any unauthorized firmware updates.

Enphase claims to have additional software updates in the works, but no estimate on the distribution date.

I seriously doubt Enphase can fix the units which are underproducing by any software modification.

They refuse to provide replacements until the inverters are completely dead.

That will not be a popular position if they take it with me. As noted earlier in this thread, not even the *internal* temperature of ANY of my inverters has ever exceeded the 65C maximum specified on the M190 datasheet. The maximum internal temperature ever seen by the newly-failed inverter was 61C and the lowest was -20C, so the unit has always been within the range of -40C to +65C specified on the datasheet. At -20C, the maximum open-circuit voltage that my panels produce should be less than 53V, which is below the 56V maximum input voltage rating specified for the M190s. Unfortunately, the MTBF of my system now stands at 68 years, which is about 1/5th the MTBF Enphase provided on the datasheet. And the MTBF of my system is better than the average which is 60 years.

I didn't purchase these inverters to massively underproduce. This one clearly has a defect which needs to be repaired or replaced under warranty. Let's see what they have to say.

So, to summarize... two complete failures and a half dozen or more with software problems for a failure rate of close to 30%.

That's not good.

 

[DaveEV]

But I wonder if the failures in the M190s could be related to the cabling system. Maybe not, but that IS one of the changes between the M190s and the M215s. More likely the failures in the M190s are due to a solder joint or component which is failing due to thermal cycling. (And I really don't think the electrolytic capacitors are the main culprits at this early stage in the game.)

Yeah, I really doubt the cabling is the issue here. Some component is aging / failing in a way that they didn't predict/expect in the M190.

I just remembered that I did have one inverter start reporting erroneous ground faults last year, but a firmware update fixed that.

FWIW, here's the details for my M190s from my envoy's inventory page:

HW Part Num: 800-00037-r08
Running Image: 520-00008-r01-v01.03.19
Updated: Sun Sep 29, 2013 07:25 AM PDT
Controller Part Num: 480-00002-r01-v00.0b.12

The M190 does currently have one large advantage over the M215/250. You can buy them for $99 and don't need a trunk cable: http://www.gogreensolar.com/products/enphase-micro-inverter-m190-72-240-s12-mc4" onclick="window.open(this.href);return false;

 

[Weatherman]

Looks like someone else is having power-drop problems:

https://community.enphaseenergy.com/enphase_energy_community/topics/last_few_months_have_been_and_still_are_a_challenge" onclick="window.open(this.href);return false;


I examined the power output graphs for my system over the past week, and I had eight inverters with partial or complete power drops during the day. I have an open ticket with Enphase on this issue (the ticket has been open since February 1st, when I asked them to replace a completely dead inverter).

It's bad that Enphase changed their public displays. It makes it very easy for them to hide all the m190 failures.

 

[RegGuheert]

Yeah, I really doubt the cabling is the issue here. Some component is aging / failing in a way that they didn't predict/expect in the M190.

I just remembered that I did have one inverter start reporting erroneous ground faults last year, but a firmware update fixed that.

Interesting. I suppose they just increased the fault threshold.

FWIW, here's the details for my M190s from my envoy's inventory page:

HW Part Num: 800-00037-r08
Running Image: 520-00008-r01-v01.03.19
Updated: Sun Sep 29, 2013 07:25 AM PDT
Controller Part Num: 480-00002-r01-v00.0b.12

Thanks! Here is what I currently have:

HW Part Num: 800-00065-r03
Running Image: 520-00008-r01-v01.04.04
Updated: Sun Oct 27, 2013 08:11 AM EDT
Controller Part Num: 480-00002-r01-v00.0b.12

So it appears that the controller is the same, but the hardware and the firmware are a bit more recent. I guess they must have updated the firmware since my last interaction with them.

The M190 does currently have one large advantage over the M215/250. You can buy them for $99 and don't need a trunk cable: http://www.gogreensolar.com/products/enphase-micro-inverter-m190-72-240-s12-mc4" onclick="window.open(this.href);return false;

Thanks, but I think I'm done with M190s.

Interestingly, I was JUST about ready to pull the trigger on 12 M215s when I noticed this failure. I have an old off-grid array which has 72-cell panels which will work with the M190s but not with the M215s. So I am considering moving 12 of the M190s from my roof down to other array and putting M215s up on the roof. That would likely zero our power bill and the expense would likely pay back within about 7 years since those panels are already in place. The M215s would be installed on the top row, where the highest temperatures occur, so perhaps that would reduced some of the thermal stress on the M190s. Also, I think it would be interesting to compare production between the two types in nearly-identical situations in my system. (And its possible the 12 on the other array would outperform all those on the roof since they are mounted in a way which allow elevation to be adjusted seasonally.) It also would extend the inverter warranty coverage on those 12 panels by 8.5 years.

I will likely still do this, as I really have no concerns about the M215s based on the study I have done. But I may hold off a bit until I see how warranty discussions go on this broken inverter.

 

[Weatherman]

In less than two years, I've had two, complete failures, out of the set of 30, M190s.

Thanks for posting this. Since I do not have any systems on my spreadsheet from FL, I'm guessing yours is not in my spreadsheet. Would you mind posting the public URL for your system so that I can add it in?

Here it is:

https://enlighten.enphaseenergy.com/pv/public_systems/smq489803" onclick="window.open(this.href);return false;

With this kind of display, the only way you would know I had inverters replaced is that I have a 32 inverter count on a 30-panel array.

 

[RegGuheert]

Looks like someone else is having power-drop problems:

https://community.enphaseenergy.com/enphase_energy_community/topics/last_few_months_have_been_and_still_are_a_challenge" onclick="window.open(this.href);return false;

That interesting, since that system is the one that prompted me to post this thread. I have since rewritten the OP for this thread, but I saved the original text at the bottom of the OP. Here is a bit of it:

Anyway, I found an array yesterday in Gilbert, AZ that appears to have a failure about 9 months ago and there is currently a second panel which is not producing. Of course this second failure is not confirmed to be a microinverter until it gets replaced, but I will assume for now that it is. If so, then I see three failures in AZ and one here in VA. But my failure was very early in the life of the array and it could have been caused by improper operation, so it is a questionable inclusion.

Anyway, I'm starting to wonder if we will see a higher failure rate for Enphase microinverters in AZ than in other areas. Or even worse, I'm wondering if these failures in AZ might be the leading edge of the back end of the bathtub curve where units reach their end-of-life.

For reference, that system is currently in Row 86 of my Enphase Microinverter Field MTBF Estimate spreadsheet.

I examined the power output graphs for my system over the past week, and I had eight inverters with partial or complete power drops during the day. I have an open ticket with Enphase on this issue (the ticket has been open since February 1st, when I asked them to replace a completely dead inverter).

It sounds like the failures are stacking up pretty quickly. If Enphase no longer acknowledges these as failures, then that is a huge problem.

It's bad that Enphase changed their public displays. It makes it very easy for them to hide all the m190 failures.

Agreed. But I think they are doing themselves a disservice at the same time, since it also hides how well the M215s are doing in the field, which I was just starting to get a real handle on.

 

[RegGuheert]

Here it is:

https://enlighten.enphaseenergy.com/pv/public_systems/smq489803" onclick="window.open(this.href);return false;

Thanks! I have added your system in Row 87 of the spreadsheet, but note that the row could change in the future. Using "2" as the number of failures, your system has an MTBF of 27 years. If I use "10" instead, the MTBF is only 5.3 years, which is one of the lowest around.

With this kind of display, the only way you would know I had inverters replaced is that I have a 32 inverter count on a 30-panel array.

Interesting... Perhaps they have a new way of updating Enlighten when they replace an inverter. When they replaced mine, the number of inverters in the system did NOT increment (it still reads 42). I wonder if I can track failures this way. I'm off to check out that possibility. Thanks!

...and good luck getting your failed inverters replaced.