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

[Valdemar]

I'm not processing. Adding solar would cause the utility to deliver less power to you so I'm not seeing how that would preclude you from having an EV.

I'd have to derate the main breaker to 100A in order to be able to install the size of array I need, and that would not be enough for my loads, EV including.

 

[QueenBee]

I'm not processing. Adding solar would cause the utility to deliver less power to you so I'm not seeing how that would preclude you from having an EV.

I'd have to derate the main breaker to 100A in order to be able to install the size of array I need, and that would not be enough for my loads, EV including.

Ah, so it's not that you can't do solar + EV, just not as much you'd like. Some PV is better than none

What kind of loads do you have that make 100 amps not enough? Assuming the loads were spread equally between phases that gives 19,200 watts of continuous capacity and another 4,800 watts of non continuous capacity before hitting 100 amps.

I realize this isn't how you do a load calc (local AHJ has never required one to be done) but if I turned on everything in my entire house I probably would not be able to break 80 amps. My 1400 sqft house is all electric except for a gas range and a gas fireplace that only is used when guests are over.

Base usage of about 200 watts
1,500 heat pump peak
3,800 EVSE
4,500 hybrid heatpump water heater when it gets too cold.
5,800 clothes dryer
950 dishwasher
1,500 microwave
150 fridge
1,000 for misc other loads
200+1500+3800+4500+5800+950+1500+150+1000=19,400

 

[wwhitney]

I'd have to derate the main breaker to 100A in order to be able to install the size of array I need, and that would not be enough for my loads, EV including.

A couple comments:

110A is a "standard" size breaker according to the NEC, so you might be able to find find a 110A main breaker for your panel. That would work if you can get the NEC load calculation down to 110A and you only need 40A for your solar.

The NEC load calculation is extremely conservative, so even if your load calculation comes to 125A, you may never draw more than 100A. That is why the POCO service conductors are often much smaller than the NEC sized conductors after the main breaker.

A 100A main breaker might hold indefinitely at 120A.

Of course the last two points don't help you code-wise, and since the solar install will require a permit and POCO inspection, they probably don't help.

Cheers, Wayne

 

[QueenBee]

I'd have to derate the main breaker to 100A in order to be able to install the size of array I need, and that would not be enough for my loads, EV including.

A couple comments:

110A is a "standard" size breaker according to the NEC, so you might be able to find find a 110A main breaker for your panel. That would work if you can get the NEC load calculation down to 110A and you only need 40A for your solar.

The NEC load calculation is extremely conservative, so even if your load calculation comes to 125A, you may never draw more than 100A. That is why the POCO service conductors are often much smaller than the NEC sized conductors after the main breaker.

A 100A main breaker might hold indefinitely at 120A.

Of course the last two points don't help you code-wise, and since the solar install will require a permit and POCO inspection, they probably don't help.

Cheers, Wayne

Good idea, looks to be reasonably availible for GE panels.

Also always have the load side tap after the meter and before the existing service panel.

 

[Valdemar]

I'm not processing. Adding solar would cause the utility to deliver less power to you so I'm not seeing how that would preclude you from having an EV.

I'd have to derate the main breaker to 100A in order to be able to install the size of array I need, and that would not be enough for my loads, EV including.

Ah, so it's not that you can't do solar + EV, just not as much you'd like. Some PV is better than none

What kind of loads do you have that make 100 amps not enough? Assuming the loads were spread equally between phases that gives 19,200 watts of continuous capacity and another 4,800 watts of non continuous capacity before hitting 100 amps.

I realize this isn't how you do a load calc (local AHJ has never required one to be done) but if I turned on everything in my entire house I probably would not be able to break 80 amps. My 1400 sqft house is all electric except for a gas range and a gas fireplace that only is used when guests are over.

Base usage of about 200 watts
1,500 heat pump peak
3,800 EVSE
4,500 hybrid heatpump water heater when it gets too cold.
5,800 clothes dryer
950 dishwasher
1,500 microwave
150 fridge
1,000 for misc other loads
200+1500+3800+4500+5800+950+1500+150+1000=19,400

Well, here is my list, main loads only, 2400sqft house:

3,000 pool/spa motors
7,200 EVSE (not with the 2011 Leaf but potentially)
4,800 oven
1,500 microwave
1,200 dishwasher
5,000 5-ton AC/furnace (guestimate)

So even without computers, irons, lights, etc. I'm already at 22700W. I guess I could make it work with a 100-110A breaker but I wouldn't be able to add say a Tesla HPWC in the future.

I'd have to derate the main breaker to 100A in order to be able to install the size of array I need, and that would not be enough for my loads, EV including.

A couple comments:

110A is a "standard" size breaker according to the NEC, so you might be able to find find a 110A main breaker for your panel. That would work if you can get the NEC load calculation down to 110A and you only need 40A for your solar.

The NEC load calculation is extremely conservative, so even if your load calculation comes to 125A, you may never draw more than 100A. That is why the POCO service conductors are often much smaller than the NEC sized conductors after the main breaker.

A 100A main breaker might hold indefinitely at 120A.

Of course the last two points don't help you code-wise, and since the solar install will require a permit and POCO inspection, they probably don't help.

Cheers, Wayne

Thanks, the ship has sailed already. New panel and service upgrade was done last year. As of this week I'm up and running with 9kW solar, so in the end I don't think it was a complete waste.

 

[FairwoodRed]

RegGuheert,

Are you still tracking my Enphase system? I had a D380 failure in April.

https://enlighten.enphaseenergy.com/pv/public_systems/EPbL26487" onclick="window.open(this.href);return false;

 

[billg]

RegGuheert,

Are you still tracking my Enphase system? I had a D380 failure in April.

https://enlighten.enphaseenergy.com/pv/public_systems/EPbL26487" onclick="window.open(this.href);return false;

And mine? I have lost 3 more 190's, for a total of 8.

 

[RegGuheert]

RegGuheert,

Are you still tracking my Enphase system? I had a D380 failure in April.

https://enlighten.enphaseenergy.com/pv/public_systems/EPbL26487" onclick="window.open(this.href);return false;

Yes. I have updated your entry in the spreadsheet to show two failures (I count one D380 failure as two failures.)

And mine? I have lost 3 more 190's, for a total of 8.

Updated.

Bill, your case in quite interesting, and not in a good way. You have one of the oldest systems in my spreadsheet. At 200Wp, you also have some of the lowest-power PV modules connected to your M190s. Only drees and one other owner have lower-power modules connected, at 180Wp and 175Wp, respectively.

It appears you went five years with NO failures in your system containing 23 microinverters. But you have had EIGHT failures in your sixth year. My conclusion is that SOMETHING in your microinverters is wearing out. It could be a solder joint, but it is something.

I wonder what your seventh year of operation will bring.

 

[billg]

Bill, your case in quite interesting, and not in a good way.
I wonder what your seventh year of operation will bring.

Same here. Every time I call to report another failure, I ask if they wouldn't prefer to be proactive and just go ahead and replace all of the remaining units. They decline.

 

[Weatherman]

Every time I call to report another failure, I ask if they wouldn't prefer to be proactive and just go ahead and replace all of the remaining units. They decline.

Same with me. And it's starting to surprise me a bit, since I'm now well past the 50% failure rate.


Here's a summary of where my array stands, today:

Out of the thirty, original micros, installed June 18, 2012, only six remain working with no problems. Serial numbers all start with 1211, hardware part number 800-00065-r03, image 520-00008-r01-v01.08.00, updated Sept 20, 2014.

One, additional one has the original hardware and 1211 serial number, but its 520-00008-r01-v01.08.00 image was updated April 3, 2015. That, occasionally, produces grid-gone events and cuts power. It doesn't do it very often, so I still consider it to be "ok", but not in perfect working order.


Now comes the fun part:

Eleven have the original hardware and 1211 serial number, but the image has been updated to 520-00008-r01-v01.08.01. Every one of these produces grid-gone events and reduces power output. Four of these have power reductions so severe Enphase has agreed to replace them (after nearly a year of my complaining about them).

So, that's the status eighteen out of the original thirty. Seven doing "ok", seven working in a degraded mode and four due for replacement.


The remaining twelve out of the original thirty have been replaced:

Three were replaced last spring with serial numbers starting 1213 and 1214. They have the same hardware part number and image as the original ones, which are working ok. Same software update date as well, Sept 20, 2014. All of these, three, seem to be doing ok.

Another nine of the original thirty were replaced over the past year with serial numbers starting with 1214. They are all the ones with part numbers of 800-00225-r11 and software image 520-00044-r01-v01.18.00 updated March 6, 2014. All of them are less than a year old, but seem to be doing ok.

One of these twelve was a replacement of a replacement. My first micro died in October 2013, and its replacement only lasted six months before it died.


Quite a saga, considering the micros were advertised to me as being so reliable and the array so maintenance free.

 

[RegGuheert]

Just to confirm your supposition, my Envoy that failed was an old "square corner" unit, shaped much like the brick it now resembles.

I just ran across the manual for the original Enphase EMU "brick". I thought perhaps others would like to see what it looks like.

That file is being served by Sams Club's website. Bill, did you happen to purchase your system from there?

 

[RegGuheert]

Serial numbers all start with 1211,...

Just a quick note on serial numbers: I have mostly convinced myself that the serial numbers include the year and week of manufacture. Here is a breakdown of what I believe the serial number represents:

12 YY WW Z XXXXX
-- -- -- - -----
|  |  |  |   |______The actual serial number from that week for the unit.
|  |  |  |__________Perhaps a code of some sorts.
|  |  |_____________Week of manufacure.
|  |________________Year of manufacture.
|___________________This is always 12.  Perhaps that means it is a production unit, with 10 and 11 meaning prototype or brassboard.

I'm pretty sure about the 12 and the YY. Part of the reason that I think WW might be a week number is that I do not have any WWs higher than 52. But I'm really unsure about the Z number, since all serial numbers are unique, even between product lines: Envoys, M190s, M215s, M250s, S275s and AC Batteries all need to have non-overlapping serial numbers. So perhaps the Z indicates a batch within a week or perhaps WWZ combine to make a batch number within a year.

For now, I'm going to assume we know the year and week of manufacture for each of our Enphase products by looking at the serial numbers. If you know of anything which contradicts this idea, please let us know.

 

[Weatherman]

Makes sense. Of my four latest replacements, one of the serial numbers starts with a 121450, then it goes 121507 for the other two. All three are the models with the IG.

Not sure what I'm going to get for the final one. Still waiting on the shipping notification on that one, since I only called it in last Friday. Hopefully another 1215 with IG.

 

[billg]

Just to confirm your supposition, my Envoy that failed was an old "square corner" unit, shaped much like the brick it now resembles.

I just ran across the manual for the original Enphase EMU "brick". I thought perhaps others would like to see what it looks like.

That file is being served by Sams Club's website. Bill, did you happen to purchase your system from there?

No, my system was purchased from Meridian Solar, a large commercial installer out of Austin. They have since exited the residential market and the DFW market, so I have been using a local company for the replacements.

 

[DaveEV]

Here is a breakdown of what I believe the serial number represents:

12 YY WW Z XXXXX
-- -- -- - -----
|  |  |  |   |______The actual serial number from that week for the unit.
|  |  |  |__________Perhaps a code of some sorts.
|  |  |_____________Week of manufacure.
|  |________________Year of manufacture.
|___________________This is always 12.  Perhaps that means it is a production unit, with 10 and 11 meaning prototype or brassboard.

I'm pretty sure about the 12 and the YY. Part of the reason that I think WW might be a week number is that I do not have any WWs higher than 52. But I'm really unsure about the Z number, since all serial numbers are unique, even between product lines: Envoys, M190s, M215s, M250s, S275s and AC Batteries all need to have non-overlapping serial numbers. So perhaps the Z indicates a batch within a week or perhaps WWZ combine to make a batch number within a year.

For now, I'm going to assume we know the year and week of manufacture for each of our Enphase products by looking at the serial numbers. If you know of anything which contradicts this idea, please let us know.

All my M190s are of this format:
11 09 43 1 1XXXX

Pretty sure they're not prototypes.

My Envoy has the following values for them:
HW Part Num: 800-00037-r08
Running Image - Updated: 520-00008-r01-v01.08.00 - Wed Dec 31, 2014
Controller Part Num: 480-00002-r01-v00.0b.12

 

[billg]

Here is a breakdown of what I believe the serial number represents:

12 YY WW Z XXXXX
-- -- -- - -----
|  |  |  |   |______The actual serial number from that week for the unit.
|  |  |  |__________Perhaps a code of some sorts.
|  |  |_____________Week of manufacure.
|  |________________Year of manufacture.
|___________________This is always 12.  Perhaps that means it is a production unit, with 10 and 11 meaning prototype or brassboard.

I'm pretty sure about the 12 and the YY. Part of the reason that I think WW might be a week number is that I do not have any WWs higher than 52. But I'm really unsure about the Z number, since all serial numbers are unique, even between product lines: Envoys, M190s, M215s, M250s, S275s and AC Batteries all need to have non-overlapping serial numbers. So perhaps the Z indicates a batch within a week or perhaps WWZ combine to make a batch number within a year.

For now, I'm going to assume we know the year and week of manufacture for each of our Enphase products by looking at the serial numbers. If you know of anything which contradicts this idea, please let us know.

All my M190s are of this format:
11 09 43 1 1XXXX

Pretty sure they're not prototypes.

All of my remaining originals are this format:

03 09 1x 0 25xxx

 

[RegGuheert]

11 09 43 1 1XXXX

Pretty sure they're not prototypes.

All of my remaining originals are this format:

03 09 1x 0 25xxx

Hmm. Perhaps the first two numbers represent the country (or perhaps factory) of origin...

Mine all came from China, so perhaps 12=China.

Do you guys know where your microinverters were manufactured?

Date codes seem correct, though.

My Envoy has the following values for them:
HW Part Num: 800-00037-r08
Running Image - Updated: 520-00008-r01-v01.08.00 - Wed Dec 31, 2014
Controller Part Num: 480-00002-r01-v00.0b.12

Here's what I have:

42 Original M190s:
Serial number: 12 10 45 4 1XXXX
HW Part Num: 800-00065-r03
Running Image - 520-00008-r01-v01.08.00 - Mon Jan 19, 2015 03:41 PM EST
Controller Part Num: 480-00002-r01-v00.0b.12
Assembly Part Num: fff-fffff-rff-vff.ff.ff (In other words, not reported.)

First M190 Replacement (Feb 2012):
Serial number: 12 11 44 0 3XXXX
HW Part Num: 800-00065-r03
Running Image - 520-00008-r01-v01.08.00 - Mon Jan 19, 2015 03:41 PM EST
Controller Part Num: 480-00002-r01-v00.0b.12
Assembly Part Num: fff-fffff-rff-vff.ff.ff (In other words, not reported.)

Second M190 Replacement (Jul 2014):
Serial number: 12 14 24 0 0XXXX
HW Part Num: 800-00065-r03
Running Image - 520-00008-r01-v01.08.00 - Mon Jan 19, 2015 03:25 PM EST
Controller Part Num: 480-00002-r01-v00.0b.12
Assembly Part Num: 880-72412-r16

Third M190 Replacement (Sep 2014 - Replaced with M190IG):
Serial number: 12 14 21 0 5XXXX
HW Part Num: 800-00225-r11
Running Image - 520-00044-r01-v01.19.01 - Thu Apr 09, 2015 06:43 PM EDT
Controller Part Num: 480-00010-r01-v03.05.10
Assembly Part Num: 880-00060-r12

Fourth M190 Replacement (May 2015):
Serial number: 12 14 38 0 2XXXX
HW Part Num: 800-00065-r03
Running Image - 520-00008-r01-v01.04.04 - Fri Dec 20, 2013 07:25 PM EST
Controller Part Num: 480-00002-r01-v00.0b.12
Assembly Part Num: 880-72412-r16

12 Added M215IGs (Jun 2014):
Serial number: 12 14 02 0 1XXXX
HW Part Num: 800-00221-r11
Running Image - 520-00044-r01-v01.19.01 - Thu Apr 09, 2015 05:16 PM EDT
Controller Part Num: 480-00010-r01-v03.05.10
Assembly Part Num: 880-00056-r06

Summary:

- All the M190s, new and old except the M190IG, in both my system and drees' use the same controller.

- drees' M190s have an older HW Part Number than all the M190s in my system, which all use the same number.

- All the M190s have been updated to the same firmware with the exception of the one I installed last week. That one will likely be updated to the same firmware soon.

- None of the M190s before 2012 report an Assembly Part Number.

- The M215IGs share a controller with the M190IG, but everything else is different from the M190s and the M190IG.

 

[RegGuheert]

From the thread on the new RMA system:

...Now 2 more panels are out - 1 more 380 - so I'm at 50% failure rate and it's been 4 years since installation. ...

I've updated the spreadsheet to reflect your current condition. MTBF for your system is down to 8 years now.

Here's a summary of where my array stands, today:

Out of the thirty, original micros, installed June 18, 2012, only six remain working with no problems. Serial numbers all start with 1211, hardware part number 800-00065-r03, image 520-00008-r01-v01.08.00, updated Sept 20, 2014.

One, additional one has the original hardware and 1211 serial number, but its 520-00008-r01-v01.08.00 image was updated April 3, 2015. That, occasionally, produces grid-gone events and cuts power. It doesn't do it very often, so I still consider it to be "ok", but not in perfect working order.


Now comes the fun part:

Eleven have the original hardware and 1211 serial number, but the image has been updated to 520-00008-r01-v01.08.01. Every one of these produces grid-gone events and reduces power output. Four of these have power reductions so severe Enphase has agreed to replace them (after nearly a year of my complaining about them).

So, that's the status eighteen out of the original thirty. Seven doing "ok", seven working in a degraded mode and four due for replacement.


The remaining twelve out of the original thirty have been replaced:

Three were replaced last spring with serial numbers starting 1213 and 1214. They have the same hardware part number and image as the original ones, which are working ok. Same software update date as well, Sept 20, 2014. All of these, three, seem to be doing ok.

Another nine of the original thirty were replaced over the past year with serial numbers starting with 1214. They are all the ones with part numbers of 800-00225-r11 and software image 520-00044-r01-v01.18.00 updated March 6, 2014. All of them are less than a year old, but seem to be doing ok.

One of these twelve was a replacement of a replacement. My first micro died in October 2013, and its replacement only lasted six months before it died.

I've updated the spreadsheet to show 16 failures in your system. Is that correct?

 

[Weatherman]

I've updated the spreadsheet to show 16 failures in your system. Is that correct?

18 replacements, so far, with seven others operating in degraded mode with grid-gone events. Only six out of the thirty are still functioning ok.

 

[RegGuheert]

18 replacements, so far, with seven others operating in degraded mode with grid-gone events. Only six out of the thirty are still functioning ok.

O.K. Thanks! Updated.

Wow! I just noticed that your system doesn't even have its third birthday until tomorrow and you have had 60% replaced already.

My system is about 4.4 years old and I have had 10% of the originals replaced to date.