Sunpower positively grounded modules

newenergy
newenergy Solar Expert Posts: 291 ✭✭✭✭✭✭
I have wondered why, if Sunpower puts the wires on the back of the modules, no other manufacturers do. I have a theory, but don't know enough about the technology to know if it's a good one or not.

Sunpower apparently came up with the idea to put the wires on the back while they still made negatively grounded modules. There was a problem where the frames got a positive static charge and this somehow reduced the power output of the modules significantly. Changing to positive grounding apparently solved this problem. That much, I've read and it seemed official, but I can't swear to it.

My idea for an explanation is that the top layer of silicon normally takes on a negative static charge at the P-N junction, but the junction between the top layer of silicon and the glass and frame is also a semiconductive junction (slightly) and allows a negative static charge to accumulate on the top of the top layer of silicon which induces a positive static charge in the frame.

When the modules were negatively grounded this static separation of charge caused a current to circulate from the frame of the module through the grounding system, through the negative terminals of the module(s), through the bottom layer of the silicon, across the P-N junction, to the top of the top layer of silicon and back to the frame. This current would be in opposition to the intended current and diminishes the power.

When the modules switch to positive grounding, this wouldn't cure the problem with the static charge on the top of the module, but it would alter the path of the current to: frame->grounding system->positive terminals->top layer of silicon and would thus bypass the P-N junction and would not cause the problem with loss of power.

But, IF this is the case, is the intentional use of the grounding system as a current carrier a problem with the NEC?

My theory of why this isn't a problem when the wires are on top is that the accumulation of charge on the top surface of the silicon is carried away by the negative conductor to the grounding system at the point of connection in the service panel where it's supposed to go - or it interacts with other loads.

Anyway, this is just a guess - I'm not an electrical engineer or physist.

Comments

  • boB
    boB Solar Expert Posts: 1,030 ✭✭✭✭
    Re: Sunpower positively grounded modules

    Last summer at Solar Power 2009 in San Diego, I tried to find out more about these positive grounded modules. Nobody in the booth could tell me anything. The problem with grounding the positive lead of course is that most controllers switch the positive lead so you cannot ground that AND the battery positive or negative lead.

    These are not very useful in a battery backup system because of this. You can't even float the modules and keep the high efficiency. So, my question to them is, does this all mean that you couldn't use them is an airplane or a satellite or boat ? That is basically a rhetorical question, but is a question of what they mean by "ground". I thought that maybe you could make some kind of Faraday shield behind it to ground to, but nobody at their booth could answer that question.

    You can't run operating currents through the ground leg. I guess that would be a "ground fault"

    boB
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: Sunpower positively grounded modules

    It is not the metal frames that affect the "charging process"... It is the positive or negative grounding of the solar panel power leads themselves. So--there is not direct operational effect by the Frames so NEC would not care about that.

    I found a new link to one of the old papers here (PDF Download).

    The original thread thread with a bunch of quotes (from other papers that may no longer be available on the web) about the issue is in this thread:

    http://forum.solar-electric.com/showthread.php?t=1458

    It has been a long time since I read the details--but a simple way of understand is to think of the solar panel as having a floating capacitor element. When negatively charged--the solar PV energy capture charges this capacitor--and the charge, more or less, interferes with the movement of holes and electrons in the PV substrates.

    When positively grounded, the charge will bleed off and the charge carriers no longer have an "E Field" from that "capacitive element" in the front of the panel that keeps them from moving as well.

    For Grid Tied systems, this is not a problem, at least one vendor (Xantrex--and I am sure there are others) have a Positive Ground stuffing option for their inverters so there is no issue (GT inverter isolate the PV panel wiring from the house wiring/grounding--so they don't care how, or even if, the PV panels are grounded--just changing a couple component stuffing locations changes which way the ground is connected).

    For Off-Grid systems--the problem is more difficult. Most (all?) solar charge controllers (PWM and MPPT) are not isolated between their PV input and the battery output... So however the battery system is grounded is how the solar panels will be grounded.

    In the last 40+ years, for what ever reason, negative grounding for consumer electronics became the standard (1950's and earlier cars, and cars from Europe used to, commonly, have positive grounding).

    So--you are kind of stuck with either making your own home system positive ground (and double checking everything you connect does not use the metal case as negative ground)--or staying away from, otherwise very nice, Sunpower solar panels (telecom/phone network systems and galvanic cathode pipe protection systems use positive ground to help prevent pipes from being corroded by stray electric currents).

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: Sunpower positively grounded modules

    boB,

    My guess is there are a couple of solutions...

    One is one that Sunpower was supposed to have implemented (but, I guess, never did):
    SunPower is currently modifying the A-300 to include a front surface conductive layer. This layer shunts leakage current and completely eliminates the polarization effect. Production is slated for 2006.

    The other would be to create a local ground plane--probably by using static discharge principles (such as used on jet aircraft--or to measure local electric fields by using a "needle" tipped with a radio isotope to increase conductivity in the local area).

    It seems that the local ground plane would work in air--and it might work in space--but creating a local conductive surface layer would probably be more reliable when dealing with solar winds and radiation in space (hard/cosmic radiation usually being the killer in space born electronics--and even responsible for the occasional bit flip in a computer's RAM memory--and not very healthy for humans either when outside the Van Allen Radiation Belt).

    My 2 cents worth of ideas anyway...

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • newenergy
    newenergy Solar Expert Posts: 291 ✭✭✭✭✭✭
    Re: Sunpower positively grounded modules

    Thanks. I'll read the links when I get the chance. I guess no NEC violation. I was partly concerned because Sunpower has been making a big deal about how they want things grounded lately - no weeb clips for example.

    SMA and Fronius allow for positive grounding as well. Sunpower just rebrands them as well as Xantrex.

    I understand the problems with batteries and charge controllers, but it seems like the vast majority of the market is grid-tied and if putting wires on the back and positive grounding helps, then other manufactorers would offer that unless it's patent protected.

    The other thing about Sunpower is it's hard to find out what they are made out of. It seems like they are basically monocrystaline, but do they have an amorphous portion the way the Sanyo HIT modules do?

    I'm far from in love with the company, but from an installers point of view, more powerful modules that are 6" narrower and weigh 10lbs less have their advantages.
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: Sunpower positively grounded modules

    I don't know anything about WEEB Clips... But, it is possible, that their UL Listing (or equivalent) requirements only allow specified grounding techniques... And the Weeb is not specified.

    Not that the Weeb is bad or unsafe--just that, for whatever reason, their NRTL testing did not include or had some specific prohibition on certain grounding methods.

    If you like the Weeb--try contacting Sunpower and Wiley Electronics directly and see why Sunpower does not allow the Weeb.

    Also, it is possible that you could take the information from Wiley to your local building inspector and show they the Weeb is superior and get them to allow you to use them.

    If I understand your use of the Weeb--it is to both mount and ground the solar panel frame (no need for mounting, then running a second grounding clip and wire panel to panel).

    Normally, the NRTL (UL and others) requirements are that mounting hardware shall never be used as a safety ground. The reasoning is that UL wants the person servicing an electrical system will not put accidentally lift the safety ground while unbolting the hardware. With the separate safety ground, the servicing person realizes that they are lifting the safety ground and the metal object may become energized (in an unsafe manner).

    I am not saying I agree with the reasoning (typically, the mechanical mounts provide a much better current path than a simple 6 awg wire bolted between to frames)--it is just how it was explained to me several decades ago.

    You might be able to get around the limitation by double nutting one of the bolts per frame--That was how we got around the "safety" issue when needing to share digital ground and safety ground... The safety ground had to go down first--then a nut, then the digital ground and a second nut.

    You might also be able to dab green paint (safety ground color) on the bolts after they are made up--to convince the inspector that you have identified the safety grounds (you could also get or have custom made weather proof safety ground symbols or rubber stamps with UL approved ink to place next to each mount--again, the rule is about identifying safety grounds--not exactly how they are made up).

    That is how I would attempt to resolve the problem if it were in my lap.

    -Bill

    PS: As one of my arguments for discussion with the inspectors--use electrical conduit as an example... The conduit is used as both a safety ground and a container for wire runs. There are conduit fittings at the box and in the middle of the runs that can be disconnected (gland fittings, screw lock, fitting nuts, etc.) and there is no "jumper wire" across the fitting... These fittings with a single nut/screw are both mechanical mounts and safety ground connections.

    On second thought--it has been many decades since I have been involved in wiring conduit--does a dedicated safety ground need to be pulled through a conduit with the power runs or not?
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • newenergy
    newenergy Solar Expert Posts: 291 ✭✭✭✭✭✭
    Re: Sunpower positively grounded modules

    In my experience inspectors have had no problem with WEEBs or other clips (unirac makes clips too). It's just sunpower. The weeb and unirac clips are not UL listed though, they are ETL listed, which is accepted pretty much everywhere. The 2008 NEC in 690.43 states:

    Devices listed and identified for grounding the metallic frames of PV modules shall be permitted to bond the exposed metallic frames of PV modules to grounded mounting structures.

    This seems to have been written with the grounding clips in mind. Sunpower's excuse is that they are not UL listed.

    In my opinion they make a good bond, but it's easy to install them incorrectly, especially with Prosolar rails. I've seen people installing them turned in the wrong direction and not really doing anything to ground the module frames to the rails.

    If the problem is that there is a charge build-up on the module frames, and they are acting like a capacitor, and the positive grounding discharges that capacitor, or prevents it from charging in the first place - doesn't that necessarily mean that there is an unwanted current in the equipment ground? Maybe it's too small to trigger the ground fault interrupt in the inverter?

    Someone else suggested that positive grounding would lead to corrosion problems and that in the early days of cars there were positively grounded batteries and they caused corrosion.
  • newenergy
    newenergy Solar Expert Posts: 291 ✭✭✭✭✭✭
    Re: Sunpower positively grounded modules

    Regarding the dedicated safety ground:

    We always have one.

    690.43 says one way to ground equipment is to comply with 250.134, which says that you must use and equipment grounding conductor permitted by 250.118, which lists EMT as one of the equipment grounding conductors.

    On the other hand 690.45 says the equipment grounding conductor shall be sized in accordance with table 250.122 and that lists the AWG size and I don't know how to size EMT in AWG, so it seems like I have to use wire.

    The inverter installation manuals all call for an equipment grounding wire as well.

    690.48 is the section on Continuity of Equipment Grounding Systems and whether or not a jumper would be required for a conduit fitting would depend on whether or not conduit counts as equipment.

    Anyway, I'm not totally sure NEC-wise, but I'm sure you will fail a lot of inspections without a ground wire. That's one of the few things that inspectors almost always look at.
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: Sunpower positively grounded modules

    ETL is, I believe, a NRTL (Nationally Recongized Testing Laboratory).

    In the "olden days"--Each country pretty much had its own approved agency--and these were used as trade barriers (if you don't pay $10,000 to my testing lab for a listing, I won't let your equipment into my country)...

    Since then (some 20+ years ago), the NRTL's have approval to pretty much test and list anything locally (in the country of manufacture) to requirements as needed (local in-country requirements can still exist--but they can be tested by NRTL's out side the destination country). Some places (like New York City) had building codes that would call out UL specifically--and they were forced (as I recall) to change to NRTL approvals instead.

    So--if ETL is recognized as competent for doing the clip testing/listing (I don't know if they are or not)--then there is no way that UL can be a limiting requirement. ETL may be able to help you in this discussion (NRTL's make money testing and doing ongoing inspections--it is to their benefit that their approvals be "respected").

    Sometimes, there are limitations listed in the NRTL's private files (data available to the NRTL and the NRTL test engineer) that is not available to the "outside world". I ran into one of these limitations when I went to get UL approval of a computer system I was designing--it was a pain to find out in the private file that the part was not approved for my use. It is possible that in SunPower's or Wiley's NRTL files that may prohibit the Weeb clip's use--and nobody will tell you about it (not likely--but possible).

    Regarding ground current through the frames--re-reading the article, they are saying the minor surface (and through glass?) conduction it what bias the MOS Transistor effect on or off--depending on electrical polarity and voltage value between the solar cell and the metal frame.

    I would have to guess that the amount of leakage current is on the order of millionth's of an amp (or less).

    Since a safety ground needs to be able to pass (continuously) 10's or more amps of currents--the minor amount of current (1/1,000,000 of an amp or less) associated with charging/discharging the is too small to have any physical or measurable effects (that I can see). It takes a 1,000 volt differential for ~1 hour to charge or discharge the EVA "capacitor" MOS Gate. Not very much conduction at all.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • boB
    boB Solar Expert Posts: 1,030 ✭✭✭✭
    Re: Sunpower positively grounded modules

    Thanks for all that extra info, Bill ! I had seen Sunpower's "white paper", but that's about it.

    boB
  • icarus
    icarus Solar Expert Posts: 5,436 ✭✭✭✭
    Re: Sunpower positively grounded modules

    I just added one Sun 90 to my string,, even after reading all I could on the positive grounding issue. I wired this new panel in parallel with the existing string,, all negative ground.

    Now my metering capability is limited, but I found virtually no difference between expected out put of the entire string relative to what I really got. The Sun should be putting out ~5 amps and indeed it did increase my yield (shown on the Tri-metric) ~5 amps in full sun.

    This all gets a bit hazier when measuring through the MPPT controller. My system is set up with one meter and shunt BEFORE the controller, and the Tri-metric shunt AFTER the controller,, so that I can see the MPPT advantage,,, at least until the controller begins to taper to absorb.

    My non scientific conclusion is that Sun has resolved the issue. Sometime when I have the time and inclination, I will consider trying to figure out a way to monitor the system a bit better. In point of fact, this time of year, with cold panel temps, much reflection off of snow and ice, my battery is fully charged by~ 10 am! The sun starts charging ~7:15,, full out put by ~8:00. I also only draw the batteries down ~5% each day so the voltage never gets low enough to full take advantage of MPPT. The true test will come in October/November when the sun is way less friendly.

    Tony
  • niel
    niel Solar Expert Posts: 10,300 ✭✭✭✭
    Re: Sunpower positively grounded modules

    the phenomenon did not occur right away so check it periodically.
  • icarus
    icarus Solar Expert Posts: 5,436 ✭✭✭✭
    Re: Sunpower positively grounded modules

    I understand Neil,

    I have been watching it for a couple of weeks. I will watch it over time,, but as I said,, my monitoring capability is limited,,,AND, I probably won't notice until sometime in the fall,,, since I don't need the power now. I'll keep everyone informed though,


    T
  • jimmy
    jimmy Registered Users Posts: 1
    Do you guys know if the Solaredge  with optimizer it’s OK with those panel positive ground Required
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    You probably will not find much about the Sunpower positive grounding issues... One because of link rot, and two because they redesigned the panels something like a decade ago--And "new"(er) protection is no longer grounding sensitive... An old thread:

    https://forum.solar-electric.com/discussion/14165/positive-ground-vs-negative-ground-solar-panels

    I cannot find the post(s) that said exactly when/what models were "fixed"... But if you are dealing with panels that are something like 6 years old or newer, positive grounding should not be an issue. If you are working with panels that are 10 years or older, then positive grounding may be an issue.

    Check with Solaredge--Many times, these manufacturers have a list of "compatible" panels (for UL/NRTL Listing).

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • Dave Angelini
    Dave Angelini Solar Expert Posts: 6,730 ✭✭✭✭✭✭
    I think it was more than a decade, it always seems like it is more than we think huh?
    The cell was the A300 and that would be hard to determine as the module has a different model number.
    If your panels were before 2008 you might have the issue. attached an old tech paper.
    "we go where power lines don't" Sierra Nevada mountain area
       htps://offgridsolar1.com/
    E-mail offgridsolar@sti.net