Weird PV Data - mis-wired strings??

63skippy

Sorry if a similar issue has been addressed elsewhere on this forum, but I couldn't find anything during a brief search.

I've just had a solar+battery system "professionally" installed, but the system monitoring app is throwing up weird data, that I cannot explain, and to date I haven't got an answer from the installer either (they are not being very helpful now they've been paid). So I hoped there might a solar guru on here that might be able to give me a answer.

The system consists of 20 identical 'JA Solar' 325w panels, arranged into 2 strings. One 13 panel string on a south east facing roof, and a 7 panel string on an adjacent south west facing roof. Both strings are supposed to be wired in series - so 13x 40v = 520v and 7x 40v = 280v. The strings are connected to a Luxtec 5k Hybrid inverter with 7.2Kw of lithium battery storage.

The issue I'm having is that according to the data the 7 panel string is generating nearly twice the amount of power to the 13 panel string. The obvious answer would be shading, but is not the issue, there is no shading on either string. When its bright sunlight and the sun is square onto the 13 panels (and therefore at 90 degs to the 7 panels), it does produce about the same power as the 7 panels, however in diffuse light its lower. When the sun is at 45 degs to both strings, the 7 panels are always producing double the power of 13 panels. With the sun at 90 degs to 13 panels, the string produces minimal power.

At times in brilliant sunlight the 7 panels are reported to be producing 3.5kw, which would be 500w/panel, which is well above what the 325w panels could possibly be producing. However the voltage seems about right for 7 panels at 233v. The 13 panels have never shown a power production above 2.8w, but again the voltage seems about right at 500v.

The only explanation I could think of, is that some of the south-east panels have been wired incorrectly into the south-west string, so its not actually a 13/7 setup. But the the strings voltages don't back up this idea. And a faulty panel in 13 panel string wouldn't account for the 500w/panel production in 7 panel string

Any theories would be much appreciated!!

BB.

Welcome to the forum 63skippy,

I know nothing about the AC Hybrid-Inverter brand...

https://www.enfsolar.com/pv/inverter-datasheet/10603

It looks like the 13 panel string is not operating correctly. The unit runs 120-500 VDC panel input in MPPT range. At 500 Volts, that is right on the edge... And if your 7 panel string is logging/operating correctly, the Vmp per panel is:

• 233 volts / 7 = 33.29 volts Vmp-panel "operational"

And that looks correct. The solar panel has Vmp-std = 37.39 volts... And a "hot panel" would expect a 10-20% depression @ Vmp-operational temperature.

For 13 panels in series, Vmp-operational

• 37.39 Volts Vmp-std * 13 panels in series = 486 Volts Vmp-std (paper specification standard)
• 33.29 Volts Vmp-operational * 13 panels in series = 433 Volts Vmp-operational ("real life")
• 32.29 Vmp-op / 37.39 Vmp-std = 0.86 = 86% of Vmp-std

The 7 panel string appears to be operating correctly, and the 13 panel string does not.

The report of 500 Watts from 325 Watt panels--That does not make sense. In subfreezing weather, you can see >325 Watts per panel--But guessing you are in the Birmingham area of England, you are in the mid 50s to low 60s (F) today...

You are "overpaneled" at ~6,500 Watts--And a typical derating of 77% for "real life" operations is:

• 6,500 Watt array * 0.77 derate for panels+controller = 5,005 Watts "real life" best harvest on cool/clear spring/fall days (typically).
• Max "over panel" is rated for 8,000 Watts (again, OK here).

Overpaneling for quality GT inverters is normal... And yours seems fine (panels are "rated" for higher Vmp-std than is seen in normal day to day operation).

Can you give us the current for each array? At solar noon is should be around 8.69 Amps on a perfectly clear day (or probably more reasonably closer to 7 amps on a typical clear day).

AC Power is (first approximation) Power = Voltage * Current ... So knowing the voltage and current we can figure out the Wattage per string.

Things you can do--Get/borrow an AC+DC Current Clamp DMM (digital multi-meter). A current clamp meter just "clips" around one wire and you can measure the current in the wire (AC mode measures AC current, DC mode measures the solar string current).

https://www.amazon.com/gp/product/B019CY4FB4 (mid-priced AC+DC current clamp meter)
https://www.amazon.com/gp/product/B00O1Q2HOQ (low-cost AC+DC current clamp meter)

Note: To measure DC current, you need a "DC Current Clamp meter"... Most electricians are going to have AC only clamp meters. They are perfectly good meters--They just cannot measure DC Current... Note most AC Current Clamp meters can measure AC+DC voltages with their test leads--Make sure it is an AC+DC clamp meter, not an AC Clamp with AC+DC volt meter.

If I was testing the system.... First measure the DC Current and voltage for each string (see if inverter-charger is accurate). Operating correctly, the current in each string equals the amount of sunlight/solar energy hitting the string. If both strings are the "same angle" to the sun, then both should have the same current.

Next, swap the array inputs to the inverter-charger... See if the problem "follows" the arrays or the inverter MPPT input (bad array or bad inverter). This is something your installer should do--Unless you are real comfortable working with relatively high voltage electricity and occasional sparks (you could change wires at night--If there is no solar power cutoff switches).

My first guesses regarding problems (in random order):

• Inverter solar panel inputs are not designed for widely differing Array input voltages (433 vs 233 VDC) and the different Power harvest per string
• Inverter Assembly or design error--The reported voltage/harvest differences do not make sense if all is OK
• Software Issues--Check to see if inverter-charger is "up to date" with software/firmware.
• 13 string array is "just too high of voltage" for the inverter-charger to use... And the inverter is either getting "confused" or damaged by the excessive voltage (Voc and Vmp "rise" as temperature fall below ~25C). The "max voltage" of your array could be:
• 13 panels * 46.38 Volts Voc (open circuit voltage) = 603 Voc-array-std (that is at 75F/25C) which would be "installer error" >550 VDC max input voltage

One could argue the question of Vmp-array-cold (operational max power voltage) vs Voc-array-cold (array voltage cold, inverter not drawing power)... I don't know the details of how this inverter-charger works--But we usually assume worst case design rules... And 603 Volts Voc-std is > 550 Max inverter input... If the inverter ever operates in a mode where Voc-cold is an "operational" voltage--Then the inverter's specifications has been exceeded--And in most cases, the warranty is voided (many companies will "log" max array voltages just for warranty validation issues).

The manual is here:

https://www.luxpowertek.com/uploads/LXP-Hybrid-usermanual1.pdf

Does not really say much about designing the solar array for the inverter-charger.... Just adds a 4,000 Watt limit per MPPT (solar) input.

• 13 panels * 325 Watt panels = 4,225 Watt on 13 panel string

Possible that this is an issue--But not sure this is "the issue":

• 13 panels * 325 Watts * 0.77 panel+controller deratings = 3,253 Watt "typical" best case harvest

Good luck,
-Bill

63skippy

Thanks Bill for that extensive analysis! Got the installer coming out next week to investigate, so your insight will be valuable in not getting fobbed off if they claim all is well.
Today is the first day since installation that it hasn't been sunny, and the 13 panels are producing next to nothing, only 34w, but the voltage is 508v surely this cant be right??

BB.

It really does look like (as a starting point), there are too many panels in the 13 panel string. Since Vmp and Voc rise as the array gets colder (next winter), if there are excess voltage problems, it will only get worse.

PV1 of 316 Watts is certainly reasonable for heavy overcast/think clouds.

Also, watch the Li Ion battery voltage/state of charge... You do not want the battery to be taken below the (per cell) minimum voltage. There should be a BMS (battery management system) that gives "warnings" and possibly cell voltage balancing to prevent cells from too low (and too high) of per cell voltages (most BMS can shutdown the battery bank if the voltages exceed specifications). Li Ion batteries are very sensitive to over/under voltage events which can immediately trash cells/battery bank.

-Bill

63skippy

Installers cancelled appointment, so system is still not working properly. Supposedly coming next week now.
Bill, I think you are right in saying there are too many panels in the 13 panel string, not sure where you got 46.38v open circuit voltage from ? The spec for the panels states open circuit voltage is 40.87v? Which would mean max voltage of 531v, which less than the 550v max quoted for the inverter input range, but more than quoted 500v for the MPPT voltage range - I don't understand why these figures are different ? 13 panel string was generating only 88w with a voltage of 522v this morning?

Could someone please explain how the power generation should vary with the array voltage? My understanding was that higher voltage causes higher current flow hence great power output? But the 13 panel string has the highest voltages when its producing virtually no power. Does this mean its the inverter 'throttling' the current from flowing and preventing power generation?

And although I'd like to think its just a problem with the 13 panel string, the 7 panel cannot be right either, it was supposedly generating 3535w the other day which is 505w per panel - just not possible from 325w panels! Also its summer here, this figure was produced with the panels being very hot (well very hot for the UK!)

BB.

The panel Vmp/Voc and such... I had just did a quick search on the vendor name and wattage and the first link was:

http://www.solardesigntool.com/components/module-panel-solar/JA-Solar/4160/JAP72S01-325-SC/specification-data-sheet.html

It may be the "wrong" panel specifications. There could be different models/designs out there. Details matter in solar/electrical systems.

Here is a random list of JA panels and their part numbers:

http://www.solardesigntool.com/components/f/man-130/solar-panels-modules/130-solar-panels.html

You are correct... It is very rare for a solar panel to produce more than it rated wattage (sub freezing weather, reflections from sand/water/snow/"edge of cloud" effects)...

With a GT inverter--Yes, the inverter is taking measurements of the voltage and current from the array, and adjusting its operating parameters accordingly. Basically figuring out the Vmp-array (voltage maximum power) of the array, and adjusting its current draw to keep Vmp-array-estimated at the correct value (kind of like you using the gas pedal in a car to keep it going at 65 MPH on the freeway).

But there is no way the GT inverter can draw more power than the panels are capable of. And for "hot panels" (panels in full sun), the Vmp of the panel falls by 10-20% as they get hot (panel temperature over 25C test conditions).

On a warm, cool (not near or below freezing), clean panels, clear day around solar noon--I would expect around 77% (at best) of the array's output rating for a GT inverter system--And that is usually not a daily event. "Typical" output for a GT inverter is around 50%-77% of rated array output (haze/dust/humidity in air, dust on panels, warm day, etc.).

The only way a string can produce more power than rated on a sunny/warm day... The array is not wired as thought (two strings in parallel, going to the wrong terminals in inverter, inverter "errors" (software, hardware, etc.).

There are other reasons a GT inverter may draw less than maximum available wattage from an array--But that is usually in "special" conditions (the power company wants to limit GT power because too much solar is causing them problems with their generator/distribution system... Or other utility problems and issues).

Your issues are way out of the range of "normal". This is an installer issue--Not something to ignore.

Make the installer fix the problem(s). You don't need to tell him what is wrong. The discussions here are, more or less, to educate you so you can understand the issues as explained by the installer/engineer--And make sure that he is not "B.S.ing" you (such as somehow turning this into "your fault").

-Bill

BB.

One other question... Are the two arrays more or less "identical" (same mounting, on same type roof, facing same direction/angle, etc.)?

You have to input to your GT Inverter. If the two arrays are identically mounted and "see" the same sun--They they should have near identical (anything within ~10% of each other is "close enough" for solar) outputs (output per panel). If the arrays are "different" (one facing south west, the other facing south east, chimney shadow on one array, different angles, etc.), then comparison of A to B is not as clear cut/easy.

If they do not--Then either a solar panel issue (there could be a "bad panel" somewhere), an array wiring issue, a GT Solar Inverter software issue, or a GT inverter hardware issue (could be an inverter design issue too).

Don't let them try and convince you this is "how solar works" or your expectations are "wrong".

-Bill

63skippy

Thanks again Bill,
To answer your question about orientation above, the 13 panel string faces south east, the 7 panel string faces south west. Even when the sun is square onto the 13 panels, the 7 panels are producing more power

Dave Angelini

I did not read most of this, but is the inverter a dual mppt?

BB.

You are very welcome.

Tell them "it broke, they fix". 

Honestly, you are getting close to needing to look into your legal options. Your laws are quite different from ours in the US (I would guess)... You might be getting to the point of getting your money back and having them uninstall/repair any damage to your roof/etc. At they very least, you got a Lemon of a System at this point. And unless they can convince you that they found the fault(s) and have permanently fixed the problems--You don't want the system (at least as it is now) and you need to hold their feet to the fire.

-Bill