Ground Fault Problem

newenergy
newenergy Solar Expert Posts: 291 ✭✭✭✭✭✭
I did a service call on this system, but I didn't install it. I didn't have time to pull up panel or to disconnect and reconnect too many things on the this first visit.

4 stings of 8 Sanyo HIT modules and a SMA 7000 inverter. Combiner on the roof.
It's reported that the inverter threw a ground fault error - this is a new system that has never operated. The Ground Fault fuse (1A) is blown. The owner reports having combed through the array thoroughly looking for pinched or torn wires.

With the positives (ungrounded) disconnected in the combiner (fuses) and the negatives still landed, and the array equipment ground detached (owner cut it) so that it is only connected to the rails and module frames, there is 503V (or so) between each individual string positive and the negative bus bar, but....

When you test the voltage between the positive from any string and the ground wire that is only connected to the array you get a voltage around 300v which then decreases over about 5 seconds to about 140V and stays around there. If you do the same thing against the ground wire that continues where the cut was and goes down to the system ground you get a similar result but it rests at about 23V (this is grid-tie - no batteries). If you repeat the test immediately it won't start at such a high voltage, but if you wait about 10 or 15 seconds it will.



What do you think the problem is?

(my thoughts below - giving you a second to think about it first)





























It seems there is something holding a static charge and the only think I think could do that on the roof there is actual modules. I think there is still a pinched wire that is not making enough contact to cause a fire or anything (yet) but is charging up the modules somehow (junction between the silicon and the glass surface?) to around 300V, but the meter partially discharges. I'm not sure why the voltage would come to rest where it does. Maybe the fault current opposes the modules voltage?

If he's right and he really found any source of pinched wires, could it be a defect in one module? Maybe it is shorting to it's frame?

Comments

  • mike95490
    mike95490 Solar Expert Posts: 9,583 ✭✭✭✭✭
    Re: Ground Fault Problem

    Sunpower positive ground modules ?

    Before daybreak, short the array at the combiner box and look for smoke as sun lights the array up?
    Powerfab top of pole PV mount | Listeroid 6/1 w/st5 gen head | XW6048 inverter/chgr | Iota 48V/15A charger | Morningstar 60A MPPT | 48V, 800A NiFe Battery (in series)| 15, Evergreen 205w "12V" PV array on pole | Midnight ePanel | Grundfos 10 SO5-9 with 3 wire Franklin Electric motor (1/2hp 240V 1ph ) on a timer for 3 hr noontime run - Runs off PV ||
    || Midnight Classic 200 | 10, Evergreen 200w in a 160VOC array ||
    || VEC1093 12V Charger | Maha C401 aa/aaa Charger | SureSine | Sunsaver MPPT 15A

    solar: http://tinyurl.com/LMR-Solar
    gen: http://tinyurl.com/LMR-Lister ,

  • Dave Angelini
    Dave Angelini Solar Expert Posts: 6,731 ✭✭✭✭✭✭
    Re: Ground Fault Problem

    OK, you have to emiminate the easy stuff. Many time owners make a mistake of connecting MC's under load or allowing the (+) or (-) end to touch the frame. Pretty tuff to do with the (-) end but where there is a will or laziness in not covering the array bad things can happen. The first is a fatal or worse near fatal blow to the 600vdc GCFI fuse. Make sure you have a good fuse.

    You wrote,

    Wen you test the voltage between the positive from any string and the ground wire that is only connected to the array you get a voltage around 300v

    If there are multiple strings why can't you isolate which string is bad?

    Good Luck!
    "we go where power lines don't" Sierra Nevada mountain area
       htps://offgridsolar1.com/
    E-mail offgridsolar@sti.net

  • newenergy
    newenergy Solar Expert Posts: 291 ✭✭✭✭✭✭
    Re: Ground Fault Problem

    Sanyo, not Sunpower.

    Each string behaves the same - voltage from positive to ground goes from about 300V to about 140V over about 5 seconds. I didn't have time to detach each negative in the combiner, but the voltage from each positive to the negative bus bar is 502 or 503.

    Whatever the problem on one string is, I think it's infecting the whole array.

    If I go back, I'll definitely undo each string from it's homeruns and then test, but I'm thinking nothing will stop this until the problem module/wire is detached from the rails.
  • Dave Angelini
    Dave Angelini Solar Expert Posts: 6,731 ✭✭✭✭✭✭
    Re: Ground Fault Problem

    Yes! you have to isolate by disconnecting. Start with one string. Make sure that fuse is good. I have seen them do funny things under load when they have been abused.
    "we go where power lines don't" Sierra Nevada mountain area
       htps://offgridsolar1.com/
    E-mail offgridsolar@sti.net

  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: Ground Fault Problem

    What kind of meter are you using? A DVM may have 1,000,000 to 10,000,000 Ohms (very little--I should clarify, very little current taken by the meter). While a meter may have 10,000 Ohms across the leads.

    In either case, it sounds like you do not have a hard ground short (voltage slowly changing as the meter "discharges" the string capacitance.

    V=I*R

    R=V/I= 500 volts / 1 amp = 500 Ohms

    Presumably, your are looking at a "soft" short. Or, more likely, an intermittent short.

    The slow changing of "ground referenced" voltage (capacitor discharging) indicates to me that you do not have any "ground fault" shorts at the time you did the measurements with the meter... A 500 Ohm (or less) short would typically look like a "solid/stable" voltage when measuring with a typical high impedance volt meter.

    The possible issues I see:
    1. Connecting panels while energized to GT Inverter/Charge controller when sun is out. May be enough capacitance to blow fuse when bringing Ground voltage to zero (not sure it is possible there is enough capacitance to blow a 1 amp fuse).
    2. Bad fuse (or fuse damaged before installation)
    3. Wrong fuse (0.1 amp instead of 1.0 amp)
    4. Grounds to array are not configured as required by Solar RE device manufacturer (two or more safety ground connections in solar array/DC output ground paths).
    5. Somebody momentarily connected one lead to earth ground during the install and popped the fuse.
    6. Really is an intermittent short. Pinched wire, loose connection or free strand that is shorted when covers/doors are closed as installation is buttoned up, etc...
    What is the system? Grid Tied or Off-Grid. What is the GFI device (Xantrex GT inverter or Xantrex XW charge controller, or what)?

    -Bill

    PS: I would doubt the modules are at fault unless one is broken or, for example, a nail is sticking up and the module was installed down on the nail and it penetrated the membrane on the back of the panel. Each panel should be tested with 1,800 Volts AC (or higher) as part of the final "UL" inspection (less than a few milliamps of leakage current).
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • Dave Angelini
    Dave Angelini Solar Expert Posts: 6,731 ✭✭✭✭✭✭
    Re: Ground Fault Problem

    "What kind of meter are you using? A DVM may have 1,000,000 to 10,000,000 Ohms (very little)."


    Bill,
    I have an Agilent 34401 on the bench here at work with 10,000,000,000 input impedence in DC. It is amazing how good the gear is these days.
    "we go where power lines don't" Sierra Nevada mountain area
       htps://offgridsolar1.com/
    E-mail offgridsolar@sti.net

  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: Ground Fault Problem

    The problem with high impedance meters (and OP Amps--operational amplifiers with nano-amp 10^-9 amps--scale leakage currents) is that they can measure what is not really there (static electricity, stray EM fields, stray currents from probes/clips, etc.

    In some cases (like measuring a potential ground fault)--you need to put some sort of resistance on the circuit (like 500 Ohm / 500+ watt resistor) on the circuit to give you some load so you can find the short (if there is one) or if it is just some sort of capacitance issue.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • Dave Angelini
    Dave Angelini Solar Expert Posts: 6,731 ✭✭✭✭✭✭
    Re: Ground Fault Problem

    It is failures like this that sound complex but are often very simple in origin.
    "we go where power lines don't" Sierra Nevada mountain area
       htps://offgridsolar1.com/
    E-mail offgridsolar@sti.net

  • newenergy
    newenergy Solar Expert Posts: 291 ✭✭✭✭✭✭
    Re: Ground Fault Problem

    It is a fluke 36 meter.

    It's a grid tied installation with an SMA 7000. The blown ground fault fuse in the inverter is 1A.

    If you disregard the capacitance, what about the 140V with respect to the module frames/rails and the 23V with respect to the house's grounding system. Depending on what you measure that voltage stays steady after having decreased from a higher voltage.

    I see why a build-up of charge on the modules and frames would be a negative indication of a ground fault, but in this case the owner has cut the ground wire so the modules and frames are not currently grounded. He did this after the inverter showed a fault error and the fuse blew.
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: Ground Fault Problem
    newenergy wrote: »
    If you disregard the capacitance, what about the 140V with respect to the module frames/rails and the 23V with respect to the house's grounding system. Depending on what you measure that voltage stays steady after having decreased from a higher voltage.

    There is always some sort of leakage current. The fact that you are "adjusting" the voltage with the Fluke meter's own resistance would indicate to me that there is probably only (at that point) a very high resistance connection (why over 500 ohms)... Unless below is true:
    I see why a build-up of charge on the modules and frames would be a negative indication of a ground fault, but in this case the owner has cut the ground wire so the modules and frames are not currently grounded. He did this after the inverter showed a fault error and the fuse blew.
    There should be several grounds here--and they may not even be connected to the same ground reference/rod.

    1. The frame grounds. Have you measured the frames (now floating?) and see how they behave. I would expect them to be fairly close to earth ground (all of the metal railings, mounted to wood/roof/earth frame work.

    2. Was there a separate ground wire connected to the panel lead? (+ or -?). If there was, then that was probably incorrect. Normally, the green wire connection in the Inverter would not be accessible (inside the service area?).

    Just taking a quick look at the SMA manual--there appears to be a jumper that needs to match the position of the Ground Fault Fuse (positive or negative panel ground).

    The grounding of the panels is done (I am guessing) by where the fuse is positioned (the jumper setting just tells the inverter where the fuse is installed for sensing failure?).

    Where was the green wire to the inverter/panels "cut"?

    If there was a separate ground connection to the panel lead--that could create a ground loop which could pop the fuse.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • newenergy
    newenergy Solar Expert Posts: 291 ✭✭✭✭✭✭
    Re: Ground Fault Problem

    Ok, so maybe along the lines that the simplest explanation is the most likely:

    He had a ground fault because of a pinched wire. He picked up all the modules and put them down, and cleared the ground fault. The inverter wouldn't go on and thinking that there was still a problem with the ground, he cut the ground wire. But, the inverter simply wouldn't go on because the ground fault fuse was blown.

    There should be 500 V between the ground and the ungrounded conductors, but there's only 140 since there's a lousy ground. The negative should be connected to the ground, but either because of the cut wire and/or the blown fuse it really isn't.
  • Dave Angelini
    Dave Angelini Solar Expert Posts: 6,731 ✭✭✭✭✭✭
    Re: Ground Fault Problem

    Did you or the owner ever replace the fuse and test after the fault was cleared?
    "we go where power lines don't" Sierra Nevada mountain area
       htps://offgridsolar1.com/
    E-mail offgridsolar@sti.net

  • newenergy
    newenergy Solar Expert Posts: 291 ✭✭✭✭✭✭
    Re: Ground Fault Problem
    Did you or the owner ever replace the fuse and test after the fault was cleared?

    I'm pretty sure he didn't do that. I didn't have the right fuse, but I ordered some.