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

[Weatherman]

I have access to "Array Builder" for my system, so whenever an inverter is replaced, I just move it out of the array and into the unassigned box.

I keep the old ones around to retain their production record.

 

[DaveEV]

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.

I don't think temperature matters. A number of people in Canada have reported high failure rates with the M190s.

I do think it would be interesting if you up your array so that two near identical arrays are are on M190s and M215s.

I am highly disappointed that Enphase now hides panel level detail by default and only lets you have access to it through Enlighten Manager if you pay them an extra $250. Especially since they still don't provide any way to get the data directly off the Envoy.

For an off-grid system like the one you are thinking of, you may not even have internet access. How are you supposed to monitor panel-level production in a situation like that?

 

[QueenBee]

I am highly disappointed that Enphase now hides panel level detail by default and only lets you have access to it through Enlighten Manager if you pay them an extra $250. Especially since they still don't provide any way to get the data directly off the Envoy.

I don't know anything about Enlighten Manager but I can get panel level detail graphs of temp/voltage/etc. through my normal interface.

 

[ranss12]

I have an issue with an M215 inverter/panel pair on my recently installed 39 panel system. One of the inverters is only putting out 2/3 of the power of the remaining 38. I'm almost certain its the inverter that's the problem - the power out of the pair matches the other 38 until it reaches about 10 watts, then it drops down to 2/3 of the others. This doesn't sound like a panel issue. My system was put on line on March 14th. Enphase says they are looking into it.

Here's the public access address to my array:

https://enlighten.enphaseenergy.com/pv/public_systems/782a299796" onclick="window.open(this.href);return false;

 

[RegGuheert]

I have an issue with an M215 inverter/panel pair on my recently installed 39 panel system. One of the inverters is only putting out 2/3 of the power of the remaining 38. I'm almost certain its the inverter that's the problem - the power out of the pair matches the other 38 until it reaches about 10 watts, then it drops down to 2/3 of the others. This doesn't sound like a panel issue. My system was put on line on March 14th. Enphase says they are looking into it.

Thanks for the heads-up!

Yes, that is likely due to an inverter issue, although it could also be due to the panel. Like you, we also experienced a failure early in the life of the array. What I did to determine if the inverter or the panel was at fault was to swap two adjacent inverters with the adjacent PV panels. This allowed me to observe that the problem followed the inverter rather than the PV panel. Further to that, Enphase was able to remotely observe anomalous behavior of that inverter.

In any case, it sounds like you are experiencing infant mortality which was not caught by the manufacturer, since the issue appeared immediately after installation. Infant mortality is generally not included in MTBF calculations since MTBF is intended to indicated the failure rate of units at the bottom of the "bathtub" failure curve, not at either end where the failure rates are higher. (In the case of my own early failure, it occurred several months after installation, so it was never clear whether it should be included in the MTBF calculations or not. I decided to included it to err on the side of pessimism.)

Can you tell us whether you have the original M215 design or the new one with the integrated ground (-IG suffix)? Also, can you please post a link to the public URL for your system? TIA!

And congratulations on your new array!

 

[RegGuheert]

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

This morning, Enphase updated the firmware for just the malfunctioning inverter to be:

HW Part Num: 800-00065-r03
Running Image: 520-00008-r01-v01.07.00
Updated: Fri Apr 04, 2014 12:07 PM EDT
Controller Part Num: 480-00002-r01-v00.0b.12

It exhibited the symptom of "DC Power Too Low: Set" once right after the reprogramming, but it recovered within one minute. So far, production has been very low today: 12Wh while its eight immediate neighbors have all produced about 150Wh. We'll see how things go since the reprogramming. So far, the power level is close to its neighbors: 64W versus 68. Time will tell

 

[Weatherman]

Although the warrantee says that Ephase can attempt to repair rather than replace a microinverter, at their discretion, I do wish they would quit fiddling around trying to fix the power drop problem.

They applied an update, yesterday, to one of the eight microinverters in my array, which are experiencing power drops. The microinverter rewarded us by dropping to zero power output at 11:00 am this morning and hasn't recovered since.

 

[RegGuheert]

They applied an update, yesterday, to one of the eight microinverters in my array, which are experiencing power drops. The microinverter rewarded us by dropping to zero power output at 11:00 am this morning and hasn't recovered since.

That may be a good thing, since it likely means they will RMA it. Was it the same update applied to my inverter?

So far, that inverter has tracked its neighbor following the update. But I suspect it will not always do so, since I think there is a broken solder joint which sometime makes contact and other times does not. We'll see...

 

[RegGuheert]

Previously, I said the following about the recently-failed inverter in our system:

It exhibited the symptom of "DC Power Too Low: Set" once right after the reprogramming, but it recovered within one minute. So far, production has been very low today: 12Wh while its eight immediate neighbors have all produced about 150Wh. We'll see how things go since the reprogramming. So far, the power level is close to its neighbors: 64W versus 68. Time will tell

So far, that inverter has tracked its neighbor following the update. But I suspect it will not always do so, since I think there is a broken solder joint which sometime makes contact and other times does not. We'll see...

I have to admit that my skepticism is starting to fade. As previously noted, Enphase applied new firmware to ONLY the failed microinverter. On that day, there was essentially no production up until the update occurred. Since that time, the "DC Power Too Low" events are now gone. This one microinverter now produces a single "weird" event each day around 2:44PM: "Grid Gone". But production has been very close to that of its neighbors:

--------------------------------------------------
Date   |   Production  |  Average of 8 neighbors
--------------------------------------------------
Mar 27 |    1220 Wh    |     1181 Wh
Mar 28 |     815 Wh    |      784 Wh
Mar 29 |     188 Wh    |      183 Wh
Mar 30 |      40 Wh    |      182 Wh   <- Failure occurred here.
Mar 31 |    1390 Wh    |     1594 Wh
Apr 1  |    1050 Wh    |     1250 Wh
Apr 2  |     862 Wh    |      855 Wh
Apr 3  |     495 Wh    |      495 Wh
Apr 4  |     338 Wh    |      483 Wh   <- Fix applied in the middle of this day.
Apr 5  |    1260 Wh    |     1278 Wh
Apr 6  |    1560 Wh    |     1591 Wh
Apr 7  |     138 Wh    |      133 Wh

Anyway, so far so good since the new firmware was applied to that one inverter. Its production has gone from typically a bit higher than its neighbors to typically a bit lower, but I'll be happy even if this trend continues since this fix is far preferable to having to replace that particular inverter. Kudos to Enphase on this clever workaround so far, but let's see how things work going forward.

I don't know what they changed in the firmware, but it is interesting to think about. My guess is that they have been RMAing a bunch of microinverters and discovering that there is a failed solder joint in a part of the inverter that is NOT in the power stage. Perhaps there is a problem with a pad design or there is spme systematic soldering problem. In any case, perhaps they have modified the operation of the unit to avoid using that portion of the inverter (say, the portion of the measurement system that measures the DC voltage) and instead use other information to determine the operating point for the power stage. If so, it's a pretty clever fix that will save them a lot of money and will also get these inverters up-and-running much faster and easier than an RMA otherwise would.

FWIW, the following may be purely coincidental, but one thing that is interesting about the failed inverter is that it had previously outperformed ALL of the other 41 inverters in the array over the life of the system. It produced about 0.6% more electricity than the next-highest producer and about 1.4% more than the average of the 11 closest neighbors. So I am wondering if the next one to give us trouble will be its neighbor directly above, which has the second-highest total. Time will tell...

 

[RegGuheert]

I'm going to respond to the M215 failure posts from the "Head-to-Head" thread in this thread instead.

I am the owner of an Enphase M215 microinverters (17) system that had one "failure" during the first year of operation. I've posted about this on the Enphase Community forum in detail.

It was NOT an outright FAILURE. It was more accurately a partial failure due to accelerated degradation of that one M215's internal DC to AC conversion efficiency. Using Enlighten Manager I was able to quantify the amount of efficiency loss. Within 6 months of initial operation, the efficiency was down to 85%; a full 10% below spec. Enphase spec's say the M215 has a CEC Weighted Efficiency of 96%.

After nine months I was successful in getting a RMA warranty replacement from Enphase for that particular M215. It took a lot of "convincing" to get them to issue the RMA, and accept my findings. At first, they insisted on doing needless panel/inverter swapping as a more traditional troubleshooting measure. This resulted in my Installer having to make a total of "4" truck roll to my array. If only the suspect inverter could have been swapped out with a spare on the FIRST service call, a whole lot of extra expense could have been avoided. However, I found that carrying a "spare" M215 on service calls is not authorized. Say what?

Enlighten Manager has a lot of troubleshooting usefulness in this regard, but it seems not many people in the PV community are all that knowledgeable in using Ohm's Law. :mrgreen:

I wish Enlighten included an "at-a-glance" feature to assess internal efficiencies of each Microinverter in an array. This would be extremely beneficial in the periodic monitoring of an array's ongoing "health". The reason this feature - although very easy to implement - does not exist, is intuitively obvious; many more warranty claims would result and cost Enphase millions in lost revenue!

Nevertheless, I believe there may be countless other Enphase based arrays with similar microinverter internal conversion efficiency deficiencies, but without this feature in the monitoring platform, the system owners are none the wiser. I have posted numerous suggestion on the Community forum to add this feature all to no avail (again, not surprising). But, anyone that can do a simple Ohm's Law calculation can "see" what is truly going on behind the scenes.

Thanks for the head's up on this! I have been thinking about going back to Enphase with precisely the CEC weighted efficiency argument that you used, but I had been planning on pointing out that the output of my inverter had dropped 6% relative to its neighbors after Enphase "fixed" it with a firmware update. Your efficiency approach is much more direct! Unfortunately, my electrical efficiency is calculating out to be over 93% and the energy harvest lost has more to do with MPPT tracking efficiencies and drop-outs that are resulting in reduced energy harvest. (Although I admit that I don't follow what "Ohm's law" has to do with efficiency! In this case, you have output power, input voltage and input current. No resistance to be found anywhere. You could calculate it, but it is irrelevant.)

Anyway, please have a look at my Enphase Microinverter Field MTBF Estimate spreadsheet. That effort ended last September when Enphase turned off their old website, but the information is still rather interesting!

 

[ecoobsessive]

Just reporting that we were part of your original spreadsheet and the 24M190s were fine at the time, but we had 3 go out this spring. Waited to see if more would go out, so far that's it. The replacements are on their way from Enphase without hassle with firmware updates, etc. Original install June 2011.

 

[Weatherman]

I had three go out this spring (and one, other one, died last fall), and eight more continue to act up. This is 12 defective M190s out of 30. This has been going on since January.

I think I've called Enphase a couple dozen times on microinverter issues so far this year, and opened up a half dozen cases. The last time I talked to them, they had reached a point where they refused to replace any more microinverters unless they were completely dead. So, I'm stuck with eight microinverters, which continue to drop power during the day. The person I talked to claimed the behavior was "normal".

Needless to say, I'm very disappointed in Enphase and their very unreliable M190s.

 

[RegGuheert]

Just reporting that we were part of your original spreadsheet and the 24M190s were fine at the time, but we had 3 go out this spring. Waited to see if more would go out, so far that's it. The replacements are on their way from Enphase without hassle with firmware updates, etc. Original install June 2011.

I'm sorry to hear about your failures, Martha. Thanks for the update! I have updated the entry for your array in my spreadsheet.

Do you have any advice on getting a replacement from Enphase for Weatherman and me? Were there particular words that you spoke, or did you ask very nicely in a sweet Texas accent?

Needless to say, I'm very disappointed in Enphase and their very unreliable M190s.

Sorry to hear about your troubles. I think I would be furious at this point. (It sounds like you are there.) I have updated the spreadsheet to show four failures, but I mentioned the eight malfunctions in the notes.

My microinverter that malfunctioned has been working fine since the firmware update it received, but it still gets "GridGone" events daily and produces about 6% less electricity than it used to. Electrical efficiency is not down, however, but rather it is right around 95%. I suspect the issue is that the MPPT algorithm has been bypassed and set to a specific value. That would mean the unit is not operating to specification, but how can I prove that? With only one misbehaving so far, this is not nearly as big an issue for me as what Weatherman is dealing with.

 

[RegGuheert]

FWIW, the following may be purely coincidental, but one thing that is interesting about the failed inverter is that it had previously outperformed ALL of the other 41 inverters in the array over the life of the system.

Based on this little factoid, I am going to make the following predictions:

- The next inverter to fail in our PV array will be 121045414066. This inverter is in the garage array, top row, second from the left.
- The second inverter to fail in the future in our PV array will be 121045415714. This inverter is in the house array, middle row, third from the left.
- The third inverter to fail in the future in our PV array will be 121045414823. This inverter is in the garage array, top row, fifth from the left.
- The fourth inverter to fail in the future in our PV array will be 121045415782. This inverter is in the house array, bottom row, second from the left.

So, on what do I make these predictions? Just a hypothesis that the previous failure was brought on first because of higher stress on the components due to consistently operating at a higher power level than the other inverters. I cannot tell whether this operation at higher power levels is due to being connected to a more powerful PV module or due to some variation within the inverters which causes them to operate at higher power levels than their neighbors or some combination of the two things. But I do not think it should matter: higher power levels should mean more stress and a shorter life.

This is not an overly-important prediction, but I wanted to record it as a way to see if any predictions can be made about which M190s are at the highest risk of early failure.

On a related note, I am wondering if PV module power level is important in determining which systems are likely to experience the most failures. As I have said previously, temperature does not seem to be the primary determining factor in these M190 failures. But perhaps PV module power impacts the magnitude of thermal cycling that happens each day and therefore the stresses on the solder joints.

For reference, the PV modules in our PV system on the roof are 235Wp. The PV modules in the field array are pairs of 120Wp modules, so 240Wp total, but with 72 cells, so the voltage stresses are higher and the current stresses are lower than the 235Wp modules on the roof.

 

[RegGuheert]

Based on this little factoid, I am going to make the following predictions:

- The next inverter to fail in our PV array will be 121045414066. This inverter is in the garage array, top row, second from the left.
- The second inverter to fail in the future in our PV array will be 121045415714. This inverter is in the house array, middle row, third from the left.
- The third inverter to fail in the future in our PV array will be 121045414823. This inverter is in the garage array, top row, fifth from the left.
- The fourth inverter to fail in the future in our PV array will be 121045415782. This inverter is in the house array, bottom row, second from the left.

The inverter that is bolded above failed this morning at about 8:00AM. So I didn't get the order correct, but the failure did occur in the 10% of the original array that I had recently identified as suspect due to their high power generation.

Time to contact Enphase and see how they dispose of this issue...

 

[RegGuheert]

Time to contact Enphase and see how they dispose of this issue...

Enphase contacted me today and notified me that they are going to replace this inverter. (And I didn't even write the email with a Texas accent! )

Unfortunately, this one will not be as easy to replace as the one that was RMA'ed in 2012, but I've had a bit of practice with my recent upgrade work.

 

[ecoobsessive]

(And I didn't even write the email with a Texas accent! )

Almost three weeks after "we'll send those inverters right out" I still have not received them. "Time to contact Enphase for me, too

 

[RegGuheert]

Time to contact Enphase and see how they dispose of this issue...

Enphase contacted me today and notified me that they are going to replace this inverter. (And I didn't even write the email with a Texas accent! )

Unfortunately, this one will not be as easy to replace as the one that was RMA'ed in 2012, but I've had a bit of practice with my recent upgrade work.

Hmm. Now I'm a bit confused about what is happening. Yesterday I received an email at about 2:15PM saying "I have requested a RMA, your RMA # is blahblah." But today I noticed that ALL of the inverters are currently working. So I checked my inventory and I see that the firmware was updated to the "limp-along" version on that inverter at 3:40PM yesterday, about 1:30 later.

So we'll see if the RMA actually happens.

Almost three weeks after "we'll send those inverters right out" I still have not received them. "Time to contact Enphase for me, too

I guess I'm wondering if an RMA request triggers this firmware update *first* and if things "work", then they deny the RMA. Time will tell.

You may want to log into your Envoy locally and look at the inventory. For me, the good M190s have this firmware:

520-00008-r01-v01.04.04

The two "failed" inverters which are currently producing have this:

520-00008-r01-v01.08.00

BTW, in spite of the higher revision number the reason I think this is simply "limp along" firmware is because these two units were failed and now report "GridGone" events multiple times each day. If this were production firmware for ALL M190s, then it seems they would upload it to the entire array, which they have not done. They also harvest less energy than they did with their old firmware.

 

[DaveEV]

FWIW, my M190s are all running 520-00008-r01-v01.07.00 and were updated in late May this year.

I really wish Enphase would let us know when they update the firmware of either inverters or the Envoy.

 

[RegGuheert]

FWIW, my M190s are all running 520-00008-r01-v01.07.00 and were updated in late May this year.

I really wish Enphase would let us know when they update the firmware of either inverters or the Envoy.

O.K. That's interesting as it is lower than the "limp-along" firmware that I have, but higher than the firmware in all of the other inverters.

Do you notice any differences before and after the update in May?