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

[wwhitney]

I experienced a couple of incidents in he past month or so of arc faults occurring between the power cord and the power supply of a couple of our laptops. It was pretty scary and I felt fortunate that I was there to unplug them before they started a fire. This prompted me to install an AFCI outlet on this branch circuit to prevent a fire in case this happens again.

While AFCIs are marketed as preventing some arc faults, actual evidence of their effectiveness is limited. If you still have the faulty power supplies, you could try recreating the arc fault to see if they actually trip the AFCI receptacle. '' If you find they are useless, then you won't need to install filters for them.

Cheers, Wayne

 

[RegGuheert]

While AFCIs are marketed as preventing some arc faults, actual evidence of their effectiveness is limited. If you still have the faulty power supplies, you could try recreating the arc fault to see if they actually trip the AFCI receptacle. '' If you find they are useless, then you won't need to install filters for them.

That's a great idea! I do still have the power cord which apparently was the culprit. If I do that test again, I will try to make sure that other load are removed to try to prevent damage. It is certainly possible that the SlingBox power supply failed because of the voltages that occurred during the arc fault. Both loads were on the same branch circuit. The two instances of arc faults that I saw (or rather: heard and smelled) were so pronounced and lasted so long that I cannot imagine the AFCI not catching them.

Update: The communications between the new Envoy and the microinverters steadily improved over the couple of days following the installation of the AFCI outlet. Signal quality gradually improved from one bar to two bars, which is what it was before I installed the AFCI. Still, the new Envoy occasionally would drop out an inverter here and there.

On Wednesday evening, I decided to swap the old Envoy back into operation and move the new one back to its role as a cold spare. Thursday and Friday's performance were quite typical of what we have always seen with the old Envoy: good communications with the occasional drop-out of a few inverters when the LEAF is charging. In fact, on Friday not a single microinverter dropped communications once, even though the LEAF charged for a few hours in the morning. This is true even though the old Envoy still registers zero bars of signal quality during its communications tests. So I have to wonder: Is it JUST the result of the test that Enphase has observed degrading over this past year and NOT actual communications quality that has degraded? That sure seems to be the case, since the old Envoy works as well as it ever did AFAICT.

It is nice that I can now swap the two Envoys at will and they simply pick up where the other one left off. If I do it at night after the results have already been uploaded to Enlighten, there is absolutely no disruption in the online data. This means I should be able to have minimal loss of data when the old Envoy does eventually die.

 

[ltbighorn]

For what it's worth, the microinverter communication strength on my new install from this summer has always displayed very low. It's a short run to the 12 panels on my roof, but has always diplayed 1-2 bars, sometimes zero. Nonetheless, it always says all 12 are communicating, and I've never noticed any gaps or delays in my data. This is with the newer Envoy-S.

How do you view what firmware versions your microinverters are running? I don't see any of that info in Enlighten Manager. My Envoy-S itself is running R4.2.33 (b68db1) from Aug 18 2016.

The only issue I have had is very occasional issues with consumption metering going all over the place, including reporting negative values. However it's happened rarely enough (3-5 times in ~6 months, for a few hours at a time) that I haven't narrowed it down or been able to get Enphase to even comment on the issue. I adjusted some of the CT wires to try and keep them away from potential sources of interference, but I'm guessing the issue might be due to some bug occasionally causing the production and consumption CTs to be read non-concurrently, resulting in inaccurate readings.

 

[RegGuheert]

For what it's worth, the microinverter communication strength on my new install from this summer has always displayed very low. It's a short run to the 12 panels on my roof, but has always diplayed 1-2 bars, sometimes zero. Nonetheless, it always says all 12 are communicating, and I've never noticed any gaps or delays in my data. This is with the newer Envoy-S.

Thanks!

Yes, in my case it seems the bars do not communicate much: the old Envoy seems to communicate better with zero bars than the new(er) one does with two bars.

How do you view what firmware versions your microinverters are running? I don't see any of that info in Enlighten Manager. My Envoy-S itself is running R4.2.33 (b68db1) from Aug 18 2016.

You need to get that information directly from the Envoy itself. The Envoy serves a few webpages that you can access directly from your web browser. The old Envoys have a display which shows its IP number. But I think the Envoy-S has no display, so you may need to to get that number from your home router. Once you have the IP number, you can get to the Envoy by typing something like this into your browser:

http://192.168.X.Y

Once you get into your Envoy's web page, click on the "Inventory" link to see the inverter information.

The only issue I have had is very occasional issues with consumption metering going all over the place, including reporting negative values. However it's happened rarely enough (3-5 times in ~6 months, for a few hours at a time) that I haven't narrowed it down or been able to get Enphase to even comment on the issue. I adjusted some of the CT wires to try and keep them away from potential sources of interference, but I'm guessing the issue might be due to some bug occasionally causing the production and consumption CTs to be read non-concurrently, resulting in inaccurate readings.

Interesting. Hopefully they will get that resolved soon.

BTW, can I put you down for no failures as of today?

 

[ltbighorn]

How do you view what firmware versions your microinverters are running? I don't see any of that info in Enlighten Manager. My Envoy-S itself is running R4.2.33 (b68db1) from Aug 18 2016.

You need to get that information directly from the Envoy itself. The Envoy serves a few webpages that you can access directly from your web browser. The old Envoys have a display which shows its IP number. But I think the Envoy-S has no display, so you may need to to get that number from your home router. Once you have the IP number, you can get to the Envoy by typing something like this into your browser:

http://192.168.X.Y

Once you get into your Envoy's web page, click on the "Inventory" link to see the inverter information.

I'm familiar with the Envoy-S's direct web pages (they're fairly nice). There's no Inventory link though, unless it's hidden behind the Installer Login link, which I don't have access to. I do have access to the wifi settings etc. but no Inventory link visible.

I was able to use the production.json as a data source to feed data into pvoutput though! And archive all the data for my own use, and real-time displays around my home. It's been a rainy December week, so the production this week has been terrible, but here's an example for anyone interested in Envoy-S output:

{"production":[{"type":"inverters","wNow":603,"whLifetime":2296488.8572222223,"readingTime":1481666911,"activeCount":12},{"type":"eim","activeCount":1,"whLifetime":2287483.585,"whLastSevenDays":32469.585,"whToday":3976.585,"wNow":162.951,"rmsCurrent":2.089,"rmsVoltage":247.529,"reactPwr":186.872,"apprntPwr":258.597,"pwrFactor":0.63,"readingTime":1481666911}],"consumption":[{"type":"eim","activeCount":1,"whLifetime":1564224.148,"whLastSevenDays":82232.148,"whToday":7148.148,"wNow":198.922,"varhLeadToday":2740.963,"varhLagToday":2814.618,"vahToday":9441.243,"varhLeadLifetime":663182.963,"varhLagLifetime":819208.618,"vahLifetime":2913738.243,"rmsCurrent":4.482,"rmsVoltage":247.534,"reactPwr":-48.796,"apprntPwr":1109.38,"pwrFactor":0.18,"readingTime":1481666911}]}

BTW, can I put you down for no failures as of today?

Yup, no failures.

 

[SurfHawk]

Are you looking for more systems for your sheet? On 11/28/2016 we activated our system consisting of 12 LG 320W panels with Enphase S280 microinverters. We are in Santa Barbara, CA. Here is the link to the system: https://enlighten.enphaseenergy.com/pv/public_systems/Qn2V1099549/overview?preview=1

 

[RegGuheert]

I'm familiar with the Envoy-S's direct web pages (they're fairly nice). There's no Inventory link though, unless it's hidden behind the Installer Login link, which I don't have access to. I do have access to the wifi settings etc. but no Inventory link visible.

O.K. That makes me glad I bought a "traditional" Envoy as my spare instead of an Envoy-S.

Yup, no failures.

Updated. Thanks!

 

[RegGuheert]

Are you looking for more systems for your sheet?

Yes! Especially S280-based systems. As it stands, I am only accumulating 43 device-years per year for the S280s, so it may take well over a decade to see any failure. It would be much better to track a couple hundred inverters so that we might find a failure within the next few years.

On 11/28/2016 we activated our system consisting of 12 LG 320W panels with Enphase S280 microinverters. We are in Santa Barbara, CA. Here is the link to the system: https://enlighten.enphaseenergy.com/pv/public_systems/Qn2V1099549/overview?preview=1

Added. Thanks! You are the last system in the spreadsheet. Please have a look and let me know if I made any mistakes (link in my signature).

The measured MTBF for your S280s is currently above 0.5 years! ;-)

 

[RegGuheert]

I'm familiar with the Envoy-S's direct web pages (they're fairly nice). There's no Inventory link though, unless it's hidden behind the Installer Login link, which I don't have access to. I do have access to the wifi settings etc. but no Inventory link visible.

O.K. That makes me glad I bought a "traditional" Envoy as my spare instead of an Envoy-S.

Just one more thought on this:

Can you simply browse directly to the inventory page (change the IP number to yours)?

http://192.168.1.248/inventory?locale=en

 

[ltbighorn]

Can you simply browse directly to the inventory page (change the IP number to yours)?

http://192.168.1.248/inventory?locale=en

Funny, I had the same thought on the way home, though via thejson. Indeed, 403 Unauthorized for /inventory, but success for /inventory.json.

Includes:

* part_num: 800-00350-r02
* img_pnum_running: 520-00044-r01-v01.22.00
* ptpn: 540-00060-r01-v01.22.00
* img_load_date: Fri, 27 Feb 2015 00:44:35 GMT
* device_status (currently during night: envoy.cond_flags_pcu_ctrl.dc-pwr-low, envoy.cond.flags.obs_strs.failure)
* as well as as: installed, serial_num, last_rpt_date, admin_state, created_dated, producing, communicating, chaneid, device_control (gficlearset)

img_load_date is about 5 months before the installation of my system. created_date looks to be the install date.
Seems my firmware version is the same as your RegGuheert, though newer then philip's despite having a much earlier 'load' date. I have the metal-cased M250s.

 

[RegGuheert]

Funny, I had the same thought on the way home, though via thejson. Indeed, 403 Unauthorized for /inventory, but success for /inventory.json.

Clever! I wouldn't have thought of that! (Even if I had, my old Envoy doesn't speak JSON. I didn't try the new(er) one, yet.)

Includes:

* part_num: 800-00350-r02
* img_pnum_running: 520-00044-r01-v01.22.00
* ptpn: 540-00060-r01-v01.22.00
* img_load_date: Fri, 27 Feb 2015 00:44:35 GMT
* device_status (currently during night: envoy.cond_flags_pcu_ctrl.dc-pwr-low, envoy.cond.flags.obs_strs.failure)
* as well as as: installed, serial_num, last_rpt_date, admin_state, created_dated, producing, communicating, chaneid, device_control (gficlearset)

img_load_date is about 5 months before the installation of my system. created_date looks to be the install date.
Seems my firmware version is the same as your RegGuheert, though newer then philip's despite having a much earlier 'load' date. I have the metal-cased M250s.

Yes, only the metal-can M215IGs (part_num: 800-00351-r02) in my system have that firmware version. Interestingly, the images were loaded into my inverters at *exactly* the same time as yours: "Thu Feb 26, 2015 07:44 PM EST". Maybe when the build is installed at the factory, it uses the image creation date rather than the load date. (Perhaps the new metal-can M190IGs I got this year also have this firmware version, but I have never installed them into the system so I do not know.)

 

[RegGuheert]

Please be aware the MTBF spreadsheet (found in my signature below) will no longer load in Firefox on my laptop now that it has upgraded itself to version 50.1.0. It still loads in Safari on my iPad, but Google Sheets are not full-featured in that browser. Fortunately, if I need to edit the sheet, it seems to work fine on Safari on the Mac.

 

[RegGuheert]

Can the Envoy be polled for production data locally?

Yes it can - but it doesn't have inverter level detail:

Envoy URL: /api/v1/production

Response body:

{
"wattHoursToday": 3820,
"wattHoursSevenDays": 144430,
"wattHoursLifetime": 4988977,
"wattsNow": 1609
}

Thanks!

How new is that feature? Here is what I get:

{
"status": 404,
"error": "",
"info": "Resource /api/v1/production not found",
"moreInfo": ""
}

Of course I can still see the information in a browser directly at: Envoy URL://production

Since the new one has the latest firmware, it will get you the local system production API - for whatever that is worth to you.

Thanks again, Philip!

I finally got around to testing that feature with my new(er) Envoy and it works just as you describe:

{
"wattHoursToday": 0,
"wattHoursSevenDays": 0,
"wattHoursLifetime": 291985,
"wattsNow": 0
}

I'll be keeping the old one in place until it dies, but it is good to see that this one has local polling capabilities.

Funny, I had the same thought on the way home, though via thejson. Indeed, 403 Unauthorized for /inventory, but success for /inventory.json.

Clever! I wouldn't have thought of that! (Even if I had, my old Envoy doesn't speak JSON. I didn't try the new(er) one, yet.)

Even though my new(er) Envoy includes the API that philip told me about, it does not seem to have the JSON feature that you mentioned. I tried:

http://192.168.X.Y/inventory.json

Is that correct?

Perhaps JSON is new in the Envoy-S.

 

[ltbighorn]

Funny, I had the same thought on the way home, though via thejson. Indeed, 403 Unauthorized for /inventory, but success for /inventory.json.

Clever! I wouldn't have thought of that! (Even if I had, my old Envoy doesn't speak JSON. I didn't try the new(er) one, yet.)

Even though my new(er) Envoy includes the API that philip told me about, it does not seem to have the JSON feature that you mentioned. I tried:

http://192.168.X.Y/inventory.json

Is that correct?

Perhaps JSON is new in the Envoy-S.

Yes, that's correct. Seems like the extended /inventory.json and /production.json may indeed be limited to the Envoy-S, which generally seems much more modern and has support for metering via CTs, and supposedly managing the Enphase batteries.

I hadn't tried it before, but /api/v1/production does work on my Envoy-S. Technically this older view is also json, but certainly much less detailed than /production.json.

 

[RegGuheert]

Please be aware the MTBF spreadsheet (found in my signature below) will no longer load in Firefox on my laptop now that it has upgraded itself to version 50.1.0. It still loads in Safari on my iPad, but Google Sheets are not full-featured in that browser. Fortunately, if I need to edit the sheet, it seems to work fine on Safari on the Mac.

Grrr!! It's not just the spreadsheet that doesn't work: the Enlighten Manager page also does not render properly.

Why would Mozilla push out such a crippled release? I guess it's time for a bunch of Firefox users to abandon ship...

 

[philip]

Please be aware the MTBF spreadsheet (found in my signature below) will no longer load in Firefox on my laptop now that it has upgraded itself to version 50.1.0. It still loads in Safari on my iPad, but Google Sheets are not full-featured in that browser. Fortunately, if I need to edit the sheet, it seems to work fine on Safari on the Mac.

Grrr!! It's not just the spreadsheet that doesn't work: the Enlighten Manager page also does not render properly.

Why would Mozilla push out such a crippled release? I guess it's time for a bunch of Firefox users to abandon ship...

Firefox 50.1.0 is working fine for sheets and enlighten on my windows 8.1 machine.


Some other random "API" data from the envoy - this is all from XHR calls the envoy makes in the web interface.

Events:
http://{envoy ip}/datatab/event_dt.rb
Inventory:
http://{envoy ip}/datatab/inventory_dt.rb
Device Conditions:
http://{envoy ip}/datatab/admin_dcc_display_dt.rb?locale=en&name=adcdActiveTable


The content type header is text/xml, but it looks like it returns JSON on all of these.

 

[ltbighorn]

Some other random "API" data from the envoy - this is all from XHR calls the envoy makes in the web interface.

Events:
http://{envoy ip}/datatab/event_dt.rb
Inventory:
http://{envoy ip}/datatab/inventory_dt.rb
Device Conditions:
http://{envoy ip}/datatab/admin_dcc_display_dt.rb?locale=en&name=adcdActiveTable


The content type header is text/xml, but it looks like it returns JSON on all of these.

Interesting, thanks philip! I hadn't noticed those calls on my Envoy-S, though there are calls to:

http://envoy/home.json
Standard dashboard info like software build date, db_size / db_pacific_full, timezone, datetime, network interface details, comm strength, alerts, update_status, etc.

http://envoy/admin/lib/dba.json
{"tunnel_open":false} - guessing it's the status of the tunnel Enphase can use to troubleshoot you device.

The datatab endpoints you listed also worked and in my case, they do return content type application/json.

 

[philip]

Interesting, thanks philip! I hadn't noticed those calls on my Envoy-S, though there are calls to:

http://envoy/home.json
Standard dashboard info like software build date, db_size / db_pacific_full, timezone, datetime, network interface details, comm strength, alerts, update_status, etc.

http://envoy/admin/lib/dba.json
{"tunnel_open":false} - guessing it's the status of the tunnel Enphase can use to troubleshoot you device.

The datatab endpoints you listed also worked and in my case, they do return content type application/json.

The .json links you posted are not available on the "pill" envoy. http://envoy/admin/lib/dba does open up though, looks like it allows a force vpn connection back to enphase. The page is titled Enphase Debug Access. This isn't a page that is available in the navigation menu that I can see.

 

[RegGuheert]

Please be aware the MTBF spreadsheet (found in my signature below) will no longer load in Firefox on my laptop now that it has upgraded itself to version 50.1.0. It still loads in Safari on my iPad, but Google Sheets are not full-featured in that browser. Fortunately, if I need to edit the sheet, it seems to work fine on Safari on the Mac.

Grrr!! It's not just the spreadsheet that doesn't work: the Enlighten Manager page also does not render properly.

Why would Mozilla push out such a crippled release? I guess it's time for a bunch of Firefox users to abandon ship...

Firefox 50.1.0 is working fine for sheets and enlighten on my windows 8.1 machine.

Thanks! Mine is now working again with Firefox version 50.1.0.

I'm really not sure what happened. I had uninstalled Firefox and then reinstalled 50.0.2 to try to get it going, but I had the same problem. Then I saw your post and tried my wife's computer, which is identical to mine but was still running 50.0.2 and it worked fine. Then I updated hers to 50.1.0 and it still worked. So I decided to update mine to 50.1.0 again and now it works.

The only thing I can figure is something went awry with the update the first time around. Anyway, nothing to see here. Move along. :|

Some other random "API" data from the envoy - this is all from XHR calls the envoy makes in the web interface.

Events:
http://{envoy ip}/datatab/event_dt.rb
Inventory:
http://{envoy ip}/datatab/inventory_dt.rb
Device Conditions:
http://{envoy ip}/datatab/admin_dcc_display_dt.rb?locale=en&name=adcdActiveTable

The content type header is text/xml, but it looks like it returns JSON on all of these.

Thanks to you and ltbighorn for details about more calls which can be made to the Envoy.

For the record, the calls you listed above work fine on my old Envoy.

I'm wondering: Is there a form of the 'datatab/event_dt.rb' call which will retrieve more than 500 events. In my system with 54 microinverters operating and with some of those throwing 'Grid Gone' events multiple times per day, 500 events covers just over one day. Often I have wanted to look back further in time, but I don't know how to do that. There must be a way, since it appears that the data is stored on the Envoy. Any ideas on how to do this would be appreciated!

 

[pclifton]

Often I have wanted to look back further in time, but I don't know how to do that.

I shared my Python 3 script to archive events with RegGuheert a few days ago. We tweaked the script to add some delays for timeouts with slower Envoys. The script works with the first and second generation Envoys (apparently not the Envoy-S).

The commented script can be read about and downloaded at: http://www.kenclifton.com/wordpress/2017/01/saving-archiving-the-enphase-envoy-event-log/
And a short link to it is: https://goo.gl/kWHcBJ

I have the script scheduled to run daily.

Ken Clifton