I should mention one of the difficulties I had was the way the app updates with inverter & optmiser  information every 15 minutes. Afaik I could not see a live output in real time

Posted by: @tim441

↑

@derek-m 🤣 🤣 too late!! I should have... especially I a couple of g&t in....

Sadly not fully conclusive. Not sure if it was me... g&t...or what!

For most part covering one panel dropped all others from say 40 to 50w down to single digits. I used cardboard that covered probably 90% of panel.... so not entirely decisive with such small outputs.

One panel was always on 0w whichever panel was covered. Odd? But if all panels uncovered was similar output as others.

We know the failed optimiser is not even registering.  Can't help but wonder if I've more than one fault. 

Will try again... but main thing is to replace the known failed optimiser

 

Have you seen the following document?

@derek-m thanks for that. No, I had not seen it before. Looks like that will help guide me on things to check.

As I previously had no monitoring setup it was near impossible. But now monitoring is basically working I feel I'm in with a chance.

1. Hopefully the replacement optimiser arrives soon. I will also take the opportunity of checking the physical location of all optmisers as I've currently got them allocated randomly. All the right serial numbers... in the wrong order! Waiting for wind to drop and helper so I can remove each panel.

Plan is to test each optimiser and panel with a multimeter as well as check all connections (as you suggested earlier). Updating & correcting the physical layout. 

for good order... here is the official advice from Solaredge on the ethernet connection:
====================

Our advice will be to perform a power-cycle to both the inverter and the router first:

= Move the inverter's red toggle switch (on the bottom left) to the OFF position
= Wait 5 minutes and after that shut down the AC supply to the inverter (there should be an AC disconnect close to the inverter)
= Shut down the power supply to your house router
= Wait 5 minutes, then turn ON the power supply to the router
= Turn ON the inverter's AC supply
= Move the inverter's toggle switch to ON position

Then, if you are still not seeing all 1's, please verify the following with a call to your internet provider:

= Ensure there is no Firewall rule applied that might be blocking the inverter from communicating.
= Ensure that the following TCP ports: 22222 and 22221 have a 2-way communication open to:
http://prod.solaredge.com&source=gmail&ust=1685445391726000&usg=AOvVaw11Tb4imGeRBekq_Yb7-Po q">prod.solaredge.com
http://prod2.solaredge.com&source=gmail&ust=1685445391726000&usg=AOvVaw1uhpeY7WK11e72f-nkqkI h">prod2.solaredge.com

==================

i've tried that shutdown & restart too but so far no success so then I moved to the firewall rule.

Not sure what i'm doing!! But have gone into advanced settings for my BT hub router and added a "port forwarding rule" for external ports 22221 - 22222

not sure what internal ports should be set so just chose a couple of random ones. 

so far no change. But almost certainly I've done something wrong?

also tried internal ports as 22221 - 22222 but no success so far

various updates:

1. Solaredge pushed an inverter firmware update. Resulting in LAN connection now being seen including pings
   11111111
   I'm not sure if the firewall "issue" was just a red herring? probably ... as i think the firmware update made the difference not any other changes
2. Had a friend round to help me check all panels. Checked connections as suggested and tested any faulty looking optimisers (based on app monitoring). Took note of true physical layout of optimisers and updated app to show correct layout and serial numbers. Faulty optimiser moved to last (#8) in string. So 7 out of 8 working more or less correctly.
3. With info from app I can see 7 panels producing around 1.45kwh in full sun soon after midday. With 7 x 285w panels the max in theory would be 2kwh. Does that seem reasonable?
4. The 7 panels are operating with optimiser voltage 54 - 55v (most around 55v) and with output of 202 - 213w each (mostly around 210w)
   does that seem reasonable?

for now my plan is to continue to monitor and then update the failed optimiser once replacement received

interesting to note that it was just after noon yesterday that system was running normally and production for the day was 9kw. Recent days had typically been between 3 and 4kw. So I'm hopeful that a full day in sun will be nearer 15kw. Maybe a bit more even. I will monitor in coming days.

I'm surprised that simply moving the faulty optimiser to last in string made such a difference? Not a problem obviously - but slightly interested to understand why that would be?!

Solaredge have confirmed replacement is shipped and will be here in around 1 week so will see what the impact is then. As the 8th panel is heavily shaded it may not add a lot.

replacement optimiser arrived today. Installed and immediately significant jump in production compared to previous day at same time and same conditions. Even though the added panel is shaded. I guess the optimisers all like to "see" each other to play nicely. I expect peak production and full day production to also jump.

@tim441 This is something of a speculation, but with the faulty one in the middle, it may have been pulling the output down for all of those downstream of it. the optimisers (when working) are there to make sure that the post-optimisation output voltage of all panels is the same . you can see this if you walk around your panels in the monitoring portal  - voltage will be the same for all, amps and watts will be lower if shaded.

really glad to see you got  the solaredge monitoring working and are seeing the benefit. there are so many benefits to monitoring of all the forms of renewable energy tech that I think it should be mandatory with any and all of them.

Looking good! Yes, I will use the monitoring in coming days to get a good feel of how it should be working

Posted by: @tim441

↑

replacement optimiser arrived today. Installed and immediately significant jump in production compared to previous day at same time and same conditions. Even though the added panel is shaded. I guess the optimisers all like to "see" each other to play nicely. I expect peak production and full day production to also jump.

The optimisers are DC to DC converters, though I am not certain if it is the voltage from each panel that they are trying to balance or the current. Do you know if they are wired in parallel or series?

I don't think that optimisers were available when my panels were installed over 10 years ago, but I do believe that each of my panels contain bypass diodes, which are designed to prevent each string being affected by a shaded panel. The panels are configured as two parallel strings, of 8 panels in series, which can develop a voltage of 400v DC at the input of the inverter, when each panel produces a nominal voltage of 50v DC.

I was therefore surprised to read, when you stated previously, that shading each panel in turn affected the output from all the other panels, which is not what I would have expected.

@derek-m 

mine is the same type of system as tim's. here's snapshot of one of my panels just now (its not sunny). 

current is more or less the same for all panels.  optimised voltage is a different value to  panel voltage in pretty much all panels.

they are in series, they vary voltage to achieve nominal 400v for the string and to balance current. actual DC voltage at the inverter right now is 390v despite no sun.

some decent wiring diagrams and explanations here 

i'm wondering if I've still got a fault... currently cloudy here

#8 was the faulty one that was recently replaced

Perhaps #1 is also faulty?

Optimiser Current Optimiser voltage Power Voltage
1 1.14 36.75 36.97 32.5
2 1.83 51.12 55.89 30.62
3 1.92 54.75 59 30.75
4 2.06 55.38 62.46 30.38
5 1.92 56.12 58.28 30.38
6 2.04 53.75 61.31 30
7 1.94 53.38 59.28 30.5
8 0.59 15.62 19.52 32.88

Notes:
Optimiser 8 is shaded most of the day
A chimney & gable to the east part shades 1-7 and gradually clears so that #1 is finally unshaded by midday