Some of my Arrays are Outputting Volts but no Amps. Why?

x0054
x0054 Registered Users Posts: 8 ✭✭
Sorry, I am cross posting from the "Solar Beginners Corner" board, maybe it's more appropriate here.

We recently purchased a house with a 20 panel array installed on it. The house was built in 2001 and the panels were installed around 2002 or 2003. The house is off the grid, so all of the power is generated by the array.

Recently while doing some routine maintenance on the system we ran a diagnostic test on the array and found out that about half the panels are not working. The panels are arranged in a 2S10P configuration with the system voltage of 75V, each panel is a 35V 140W unit from BP. Out of 10 arrays 5 output normal amperage in the 1.8-2.2A while the other 5 are barely outputting 0.08-0.18A. We isolated each of the 10 arrays and tested the voltage independently, and they all were in the 76-78V range, so all appear to be working with no shorts, but still, about half are not outputting current.

We checked all of the wiring, and it all looks good. Any suggestions. Also, has anyone had to deal with the BP Solar warranty department. Any tips?

 - Bogdan
Tagged:

Comments

  • mcgivor
    mcgivor Solar Expert Posts: 3,854 ✭✭✭✭✭✭
    Here is a shot in the dark, assuming the panels are not shaded in any way, you might want to check the bypass diodes in the junction box of the affected panels to ensure they have not failed short circuit, thereby bypassing the current output. Just a thought.
    1500W, 6× Schutten 250W Poly panels , Schneider MPPT 60 150 CC, Schneider SW 2524 inverter, 400Ah LFP 24V nominal battery with Battery Bodyguard BMS 
    Second system 1890W  3 × 300W No name brand poly, 3×330 Sunsolar Poly panels, Morningstar TS 60 PWM controller, no name 2000W inverter 400Ah LFP 24V nominal battery with Daly BMS, used for water pumping and day time air conditioning.  
    5Kw Yanmar clone single cylinder air cooled diesel generator for rare emergency charging and welding.
  • animatt
    animatt Solar Expert Posts: 295 ✭✭✭
    I believe BB did a claim with his gridtied system this was several years ago.
    He responds to almost everything so if I am remembering correctly he will post his experience. 
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    That is correct... I had something like 50% or more of my solar panels fail to pass current (or rated current). Basically, there is a problem with many BP panels of the era where the junction box internals were failing over time (my array was >5 years old when it failed).

    If you are not getting Imp or so under full sun (Imp is proportional to the amount of sun hitting the panel). On my array, if you looked on the glass side of the j-box, you could see a brown haze (chocolate or coffee colored "stain") above the J-Boxes on the failing panels (you had to look carefully, it was not visible with a quick glance).

    When over ~20% to 30% or so (I don't remember how many exactly) failed panels, then BP would replace the whole

    There is also a BP Solar class action settlement for folks that want to replace their good panels (suite was the the panels will eventually fail and where unsafe--BP chose to settle rather than fight as I understand):

    http://www.bpsolarsettlement.com/

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • x0054
    x0054 Registered Users Posts: 8 ✭✭
    Hi all,

    Thank you so much for the help. So I have an update, maybe you guys could help me out figure this out. Below I attached a picture of the fuse box installed in my system.

    There are 10 arrays built from 2 SX140S panels connected in series each. Each array is connected to the fusebox in parallel with all others, and each has a fuse on the positive connection, as shown. I removed the fuse from one array at a time and did the following tests. As an example I'll list results from Array1 and Array2. Array1 is functional, Array2 outputs no voltage. Measurements were performed at noon with the sun hitting the panels mostly perpendicularly.

    1. Removed fuse and measured voltage across "Array Neg" and "Array Pos." Array1 =  81V; Array2 =  81V
    2. Removed fuse and measured current across "Array Neg" and "Array Pos" (the short circuit test). Array1 = 4.1A; Array2 = 4.1A
    3. Removed fuse and measured current across "Array Pos" and "Pos to Charger." Array1 = 3.8A; Array2 = 0.185A

    As you can see, the voltage for both Array1 and Array2 are identical, and short circuit current is identical as well. However, when I measure "contribution current" from Array 1 and 2, the current they send to the charger while connected to the system, one is feeding 3.8A while the other is basically contributing nothing at all. Array1 and Array2 are sample examples. In reality 5 out of 10 arrays are acting like Array1 and the other 5 are acting like Array2.

    Theory: I have a theory that Array1 and Array2 have different power curves. Current target voltage for the Array is 60V. I think that maybe Array1 can hit 60V and provide good current while Array2 cannot. It can provide good current, but only at lower voltage. Can this be the case?

    Any other suggestions or solutions? Thank you for the help.

     -Bogdan


  • westbranch
    westbranch Solar Expert Posts: 5,183 ✭✭✭✭
    this test:
    As you can see, the voltage for both Array1 and Array2 are identical, and short circuit current is identical as well. However, when I measure "contribution current" from Array 1 and 2, the current they send to the charger while connected to the system, one is feeding 3.8A while the other is basically contributing nothing at all.

    Was it done with only one string (of 2 panels)  done with the other string disconnected?
    You need to test the 'low' strings separately from the other 'high' strings
     
    KID #51B  4s 140W to 24V 900Ah C&D AGM
    CL#29032 FW 2126/ 2073/ 2133 175A E-Panel WBjr, 3 x 4s 140W to 24V 900Ah C&D AGM 
    Cotek ST1500W 24V Inverter,OmniCharge 3024,
    2 x Cisco WRT54GL i/c DD-WRT Rtr & Bridge,
    Eu3/2/1000i Gens, 1680W & E-Panel/WBjr to come, CL #647 asleep
    West Chilcotin, BC, Canada
  • animatt
    animatt Solar Expert Posts: 295 ✭✭✭
    I would document [maybe you have] which  5 of the 10 are not working.  Disconnect fuses to all working ones.

    Leaving just 5 substandard performers connected.

    Then see the voltage they output as well as the current.

    If they perform better it is as you think.

    If not really examine all things in common to those 5 strings that are sub performers.
  • x0054
    x0054 Registered Users Posts: 8 ✭✭
    Ok, so I figured out one thing in common to all of the sub-performing strings. They were installed in such a way that during the day closer to the evening time, one out of 2 panels in that string would get a partial shadow. The shadow would cover 1/3 to 2/3 of a panel. When I tested the panels, they were all exposed to nice bright sunlight, so the shadow shouldn't have payed a role in those readings, but I am guessing that they were damaged over time due to being exposed to a shadow while the other panel in that string was still lit.

    I reconfigured 4 sub-performing strings into 2 strings consisting of panels that are never exposed to a shadow and 2 strings consisting of panels routinely exposed to a shadow. Now I have 7 fully functional strings and 3 strings that are still dead. See diagram attached.

    Can shadows cast on the panel damage the panels? It appears to be the case, but I don't know much about solar to be honest, still learning. Thanks for all the help!

     - Bogdan

  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Any shadow that cuts across a panel will pretty much knock that panel (and others in the series string) out of production (while the shadow lasts).

    The next question--Is the array working correctly when all panels are in full sun?

    It is possible for a partially shaded panel to cause current to flow through the bypass diodes inside the J-Box. Overheated diodes can eventually fail either open or shorted.

    If a diode fails open--it is possible for that portion of the cells covered by the bypass diode to fail (too high of reverse voltage causing breakdown--The diode is supposed to bypass current around the shaded cells).

    Or the diode can fail shorted--And cut the Vmp output of that panel by 1/3rd.

    At this point--You really need to test each panel for Vmp/Imp (or Voc and Isc) in full sun--Usually, the bad panels will quickly stand out from such a test.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • x0054
    x0054 Registered Users Posts: 8 ✭✭
    Hi Bill. Based on the specs I found here http://www.posharp.com/bp-sx140s-solar-panel-from-bp-solar_p257080246d.aspx it appears that these panels do not have bypass diodes. I'll have to wait till the weekend to test them all one by one. Got to say, who ever installed this array didn't do such an amazing job of it, there is no reason they had to install the panels in the way they did, there is plenty of room to space them out so none of them get covered up.
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    I would guess that these panels have Bypass diodes, but no series/blocking diodes. Modern crystalline panels should all have bypass diodes. And probably no modern larger format panel would not have bypass diodes.

    The chances of damage from shaded cells is too high--Especially when you put multiple panels in series.

    This is a paper that has a test that was done with a simple vertical angle iron in the middle of two panels that had the bypass diodes removed. And some cells delaminated or even burned through the rear of the panel. This is a large set of papers with photographs (~65 MByte download). Don't download if this will impact your data plan:

    http://arizona.openrepository.com/arizona/bitstream/10150/556469/1/azu_etd_13837_sip1_m.pdf (starts at page ~150/151).

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset