installation producing 3x less power and half the voltage

[littleharbor]

If you are noticing lower current than expected I would look at this panel, or any others like it first.

Sunpower cell.jpg

[MichKrom]

Nothing that strong but there are some that have different tint.

Here are the pictures:

https://photos.app.goo.gl/BgcbwVtjwuQmejRf6

Plot thickens:
- at ~2PM 190V/7A and 250/9A - note that panels are rated at 5A so with two strings in parallel it should be possible to get to 9A
- at ~4PM both arrays started to produce the same amount of power (according to inverters) still with different voltages but similar amps (190V/250V and both at ~4A and reported power about 1kW)

Something is not adding up.

I will be doing some more experiments:
- covering pannel by pannel and observing the changes
- turning the house ooff and using the SmartMeter to measure power sent to the grid

[littleharbor]

[QUOTE=MichKrom;n400835]No discoloration of panels I can see. They are bit dirty but again, uniformly dirty. Will clean them up,




The discoloring I'm referring to would be quite obvious.

Sunpower brown cells (3).jpgSunpower brown cells 2 (3).jpg

[ButchDeal]

Could be a combiner breaker tripped...

[MichKrom]

No discoloration of panels I can see. They are bit dirty but again, uniformly dirty. Will clean them up,

I was wrong on warranty, it is 10 years for product (I am out of it) and 12 years for Power no less than 90% and 25 years for no less than 80%. The warranty's language suggest they do not cover replacement cost: "SunPower will replace such loss [in power] by either providing to the Customer additional PV modules to make up such loss in power or by providing monetary compensation equvalent to the cost of additional modules".

[littleharbor]

Have you had a look at the panels, up close? Any discolored cells?

[J.P.M.]

Yea, the docs state 12 years for mfg defects and then 20y 90% and 25y 80% (afair) performance guarantee. I suppose a broken panel has peformance of 0 so it would fall into this anyway? And that covers the panels not service and reinstall. I found refurb panels for $80 on eBay so that's not a big issue.

But first I need to know what's wrong and that's gonna take me some time.

Finding out what's wrong is part of the warranty.

[MichKrom]

Yea, the docs state 12 years for mfg defects and then 20y 90% and 25y 80% (afair) performance guarantee. I suppose a broken panel has peformance of 0 so it would fall into this anyway? And that covers the panels not service and reinstall. I found refurb panels for $80 on eBay so that's not a big issue.

But first I need to know what's wrong and that's gonna take me some time.

[J.P.M.]

Ha ha, that will be fun to learn the results of.

Well, it would be interesting to me, at least, and for several reasons, to see if they are as good as their word hyped up word.

As for the OP looking for comments/suggestions: Call the vendor or Sunpower after you read their warranty that was in effect in 2008 (which BTW, does not appear to have changed too much).

Sounds like the OP may possibly not be aware that the installation may well still be under warranty, at least in theory, or that messing with the system may possibly void any warranty.

Where's the harm in asking ?

[Mike90250]

The panels ought to be under Sunpower's super-duper bumper to bumper warranty they scream to the heavens about. Have you called them ?

Ha ha, that will be fun to learn the results of.

[J.P.M.]

The panels ought to be under Sunpower's super-duper bumper to bumper warranty they scream to the heavens about. Have you called them ?

[bcroe]

Point it at panels and cells of panels, best results after some running in good sun
has stabilized temps. You will see that the good strings are pretty uniform. I am
pretty sure some big variations will be seen in the bad ones, my failures ran hot,
but colder is possible. Bruce Roe

[MichKrom]

Thanks Bruce!

With thermal gun measuring line temperatures what am I looking for? Active lines being hotter? Or for panels that are hotter/colder?

[bcroe]

Something else that may work, use a thermal gun to find out of line temps after
it is running best power. This found faults for me. Bruce Roe

[bcroe]

I would start by disconnecting one of the strings on the bad side and see if the inverter
will run on the other string, observe V and I. Then try the same on the other, this
should give clues to the problem. If both read exactly the same, the inverter does
become suspect, connect a non functioning string by itself to the other inverter. For
safety all wiring to be done well after sunset, measurements the next day.

If you find a string putting out much reduced voltage, you could take it all apart and
measure the voltage of each panel. If they read normal, you will need to make a
more dynamic, under load test. More later. Bruce Roe