LG 300, gooey stuff on panels?

2»

Comments

  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Re: LG 300, gooey stuff on panels?
    mike95490 wrote: »
    uh, not really. you can buy up to about 5% ammonia, nearly all ammonia is way diluted before it reaches the consumer (like hydrogen peroxide, very hard to get more than 3%)

    Thank you for that completely useless bit of nit-picking that contributes absolutely nothing to the thread.

    Ammonia in its natural state is a gas. Anyone reading my post would understand it was a reference to an ammonia solution rather than the window cleaner with a small percentage of ammonia as part of its ingredients.

    Unless of course they were an idiot.
  • solardan
    solardan Solar Expert Posts: 36
    Re: LG 300, gooey stuff on panels?
    mike95490 wrote: »
    maybe defects in AR coating ?

    Yup, something like that. Here are more pictures that may help generate some ideas. LG is still thinking there is some sort of substance on the surface that can be removed, but I'm becoming skeptical of that.

    The worst panel, with 20+ streaks, when dusty after a brief rain - note the streaks on this panel are in groups of 4 that repeat about every 2.2 cells up the panel (like something rolled over the panel 4 times, producing 4 sets of streaks):

    Attachment not found.

    From same panel, after cleaning off dust:

    Attachment not found.

    From same panel, after spraying with water - note water accumulation at streaks:

    Attachment not found.

    From same panel, after vigorous Windex treatment of upper middle 2 cells shown - note less water accumulation, but streak still faintly visible:

    Attachment not found.

    And finally, here's a plot showing that the streak number per panel decreases with serial number (trending toward zero), perhaps indicating a bad batch of panels:

    Attachment not found.

    So it's as if something rolled over the panel during production, producing a defect every 2.2 cells. That would produce 4 or 5 such horizontal defects per panel (perhaps not all readily visible), or some multiple of that if more than one set of defects/streaks was being added to a panel (such as the worst case above, where there are 4 sets of defects, each producing 5 streaks, or 20 streaks total).
  • niel
    niel Solar Expert Posts: 10,300 ✭✭✭✭
    Re: LG 300, gooey stuff on panels?

    am i understanding you right that you clean it free of streaks and after a rain and then a bit of dust blows that all of the streaks return? if so then let the manufacturer know this as this is unacceptable.
  • solardan
    solardan Solar Expert Posts: 36
    Re: LG 300, gooey stuff on panels?
    niel wrote: »
    am i understanding you right that you clean it free of streaks and after a rain and then a bit of dust blows that all of the streaks return? if so then let the manufacturer know this as this is unacceptable.

    Yes. The underlying defect creating the streaks can't be cleaned/removed, and the defect attracts water, so any bit of hardness or dust/dirt in the water attracted to the defect will end up exposing the defects as multiple horizontal streaks on the panels. The disagreement I'm having with LG now is that they still think the streaks can be (permanently) removed with a cleaning agent, but the evidence suggests otherwise.
  • CDN_VT
    CDN_VT Solar Expert Posts: 492 ✭✭✭
    Re: LG 300, gooey stuff on panels?
    solardan wrote: »
    No trees, and panels have been up there less than 2 weeks. Can't tell if panels came with the streaks, since only showed up when dirt/dust got on the panels once on the roof.

    Do this cleaning in the shade or make a blind , I wound use citral orange cleaner first to get the pine pollen off. If you use MEK.Actone and less , do only a small edge corner to see how well you can destroy the surface .
  • solardan
    solardan Solar Expert Posts: 36
    Re: LG 300, gooey stuff on panels?
    CDN_VT wrote: »
    Do this cleaning in the shade or make a blind , I wound use citral orange cleaner first to get the pine pollen off. If you use MEK.Actone and less , do only a small edge corner to see how well you can destroy the surface .

    Read my latest posts - can't fix the streaks by cleaning them, they just reappear at same location with least bit of water + dust/pollen/dirt. The title of the thread reflects original theory about nature of streaks (gooey stuff), but more likely they are due to a panel surface defect that attracts water.

    (And the reason that Windex has some effect on reducing water accumulation at streak lines is probably because it contains anti-streak agents that are temporarily coating the surface,
    http://www.whatsinsidescjohnson.com/en-us/products-by-brand/windex/windex-original-glass-cleaner.aspx
    So any such glass cleaning product with anti-streaking agents will probably temporarily reduce streaking on the panels, but not address the underlying panel defects - anyway, that theory would explain why Windex appears helpful, but its component cleaning agents like ammonia are not.)
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Re: LG 300, gooey stuff on panels?

    solardan I think you have analyzed the situation well; it does appear to be defect in the coating not something stuck to it. You may have trouble convincing LG of this but explain that you've done exactly what they suggested with no improvement. I'm seeing warranty claim. If it does not affect production they may offer you a rebate instead of panel replacement.
  • solardan
    solardan Solar Expert Posts: 36
    Re: LG 300, gooey stuff on panels?
    solardan I think you have analyzed the situation well; it does appear to be defect in the coating not something stuck to it. You may have trouble convincing LG of this but explain that you've done exactly what they suggested with no improvement. I'm seeing warranty claim. If it does not affect production they may offer you a rebate instead of panel replacement.

    Thanks cariboocoot for the support. I'm trying to give LG the chance to do the right thing, but it's hard/foolish to deal with this without the support of experienced folks on a forum like this (as with the recent question about applying lots of xylene to panels).

    My main concern is life expectancy of the panels, although I doubt if anyone would accept panels with these streak lines to begin with, either for aesthetic reasons, or because it seems indicative of inadequate attention to manufacturing details. LG has also made a point about the quality of these panels compared to others (one of the reasons I paid a premium for them), as seen in these slides from 2013 promotional material (note in particular the NYT reference to coating defects affecting panel life):

    Attachment not found.

    Attachment not found.
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: LG 300, gooey stuff on panels?

    Almost sounds like an embedded electrostatic charge in the glass... Which is attracting dust to the charged areas. I think they can do this with glass.

    In some products, embedding static charge is used to collect dust (i.e., electrostatic dust filters in furnace systems--They "charge" plastic fibers during manufacture so they will collect dust without the need for electrical wiring/high voltage charging system). This is done by exposing a liquid polar material to a strong electrostatic field and cooling the material until solid. Not something I would expect to happen by "accident" during processing.

    No amount of cleaning will remove the charge. This is a wild guess--But sort of fits the description of what is happening (assuming any "goo" on the surface is simply accumulated "stuff").

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • solardan
    solardan Solar Expert Posts: 36
    Re: LG 300, gooey stuff on panels?
    BB. wrote: »
    Almost sounds like an embedded electrostatic charge in the glass... Which is attracting dust to the charged areas. I think they can do this with glass.

    In some products, embedding static charge is used to collect dust (i.e., electrostatic dust filters in furnace systems--They "charge" plastic fibers during manufacture so they will collect dust without the need for electrical wiring/high voltage charging system). This is done by exposing a liquid polar material to a strong electrostatic field and cooling the material until solid. Not something I would expect to happen by "accident" during processing.

    No amount of cleaning will remove the charge. This is a wild guess--But sort of fits the description of what is happening (assuming any "goo" on the surface is simply accumulated "stuff").

    Yup, something like that. Unless they already have a good idea of what the problem is, perhaps they should be asking me to ship the worst panel (with 20+ streaks) back to them for inspection, or have someone visit the site, to nail down what is going on.

    The pictures from an earlier post showing the effect of WD40 vs. Goo-Gone vs. Windex are interesting to reconsider in light of all the evidence,

    Attachment not found.

    So what we are looking at is likely precipitate from hard water (+ dust/dirt) used to rinse the panels (in an earlier attempt to remove the streaks w/ soap and water), and the WD40 and Goo-Gone didn't touch that, but the Windex, with ammonia, etc., got to the precipitate, and also left a layer of anti-streaking agent that reduced future water accumulation over the surface defect (i.e., reducing reformation of the streaks, temporarily), but w/o fixing the defect.

    We'd also predict from that that Xylene wouldn't do a thing to the streaks, and that ammonia might clean the precipitate, but have no effect on water accumulation above the defect (allowing the streak to reform), which is what we found.
  • mike95490
    mike95490 Solar Expert Posts: 9,583 ✭✭✭✭✭
    Re: LG 300, gooey stuff on panels?

    have you tried Rain-X ? Don't let it get to the seal edges. It's solvent is some version of alcohol, and may work differently. And the hydrophobic coating left behind, may solve the issue for 6 months.

    http://newsoffice.mit.edu/2013/hydrophobic-and-hydrophilic-explained-0716
    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 ,

  • solardan
    solardan Solar Expert Posts: 36
    Re: LG 300, gooey stuff on panels?
    mike95490 wrote: »
    have you tried Rain-X ? Don't let it get to the seal edges. It's solvent is some version of alcohol, and may work differently. And the hydrophobic coating left behind, may solve the issue for 6 months.

    http://newsoffice.mit.edu/2013/hydrophobic-and-hydrophilic-explained-0716

    Hi Mike. Thanks for the link. My background includes chemistry, so am familiar with the terms and concepts.

    The issue is mainly long-term life expectancy of the panels (are the streaks an indication of a more fundamental problem?), not current performance, so a temporary solution is not needed at this time. Plus, I don't want to set the precedent that I'm willing to address the problem by applying a coating to all of the panels every few months. :roll: And you'd have to be certain that Rain-X itself wasn't a problem for the panels. But you're probably right in thinking that this is the sort of thing you would need to do to fix only the cosmetic aspect of the problem (vs. the gamut of cleaning agents we've tried).
  • solardan
    solardan Solar Expert Posts: 36
    Re: LG 300, gooey stuff on panels?

    Update: Latest theory from LG is that streaks may have been caused by EVA (ethylene vinyl acetate) contamination during production (i.e., EVA residuals left on a machine - mainly an aesthetic issue, and only indirectly affecting output/longevity via deposits in the streaks). Will be investigating this further to see if that makes sense, but thought I'd get this posted to see if anyone had any immediate insights to add.

    Given LG promotional material,

    Attachment not found.

    and classic crystalline solar panel construction,

    Attachment not found.

    and more detailed view of where the EVA encapsulant and other features reside,

    Attachment not found.

    I guess first conclusion would be that EVA belongs inside the panel, not on top of it. :roll:

    The Dupont folks in first half of this video give a nice little history of EVA and its current use:
    http://www.youtube.com/watch?v=diJ1FC-6mnc

    And here is an original JPL report on EVA:
    http://www2.jpl.nasa.gov/adv_tech/photovol/ppr_81-85/Appl%20of%20EVA%20as%20Encap%20_5101-220_JPL1983.pdf
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Re: LG 300, gooey stuff on panels?

    I only notice the irony in their claim that their process "minimizes defects". :roll:
  • niel
    niel Solar Expert Posts: 10,300 ✭✭✭✭
    Re: LG 300, gooey stuff on panels?

    solardan,
    so they have a theory, but did they say what it is to be done about it? if you can't take it off then there will be the degradation of power production and falls under the warranty due to not delivering rated power due to obstruction from accumulated minerals, dirt, and dust that is sticking to those streaks.
  • solardan
    solardan Solar Expert Posts: 36
    Re: LG 300, gooey stuff on panels?
    niel wrote: »
    solardan,
    so they have a theory, but did they say what it is to be done about it? if you can't take it off then there will be the degradation of power production and falls under the warranty due to not delivering rated power due to obstruction from accumulated minerals, dirt, and dust that is sticking to those streaks.

    I think they are doing their best to try to figure out a way to remove the stuff (assuming it's EVA, which is just a theory, although they may be able to see from the patterns that I documented that this makes the most sense). The best news is that we may be zeroing in on a likely point where the problem arose (i.e., not the glass subcontractor, not shipping, not handling, not installation, etc.). The warranty has 2 parts: defects in material or workmanship, and power production. imho, this is clearly a defect, but they might argue that it's not a defect unless it significantly affects power production and/or life expectancy. Murky stuff, but I don't think they want to leave this unresolved.
  • niel
    niel Solar Expert Posts: 10,300 ✭✭✭✭
    Re: LG 300, gooey stuff on panels?

    i guess technically it was a manufacturing defect too so imho you got it on both accounts. i believe they should replace them as removing the eva if possible is one thing, but it most likely will adversely affect the anti reflection coating. you didn't pay for blems and if they are honest and don't want a black mark against them here they will come through.
  • solardan
    solardan Solar Expert Posts: 36
    Re: LG 300, gooey stuff on panels?

    I edited today's earlier EVA post to give some background info about EVA and module construction. Next question is whether EVA contaminant on a panel's surface would behave like the streaks, and whether there are any reports of such contamination on other panels.
    Another conclusion: Any cleaning agent/solvent that readily removes EVA from surface would put panel at risk of being delaminated (i.e., could also attack EVA laminant inside the panel) if that material reaches edges (as others here pointed out). When applying Xylene, for example, it ran quickly down the panel surface and I had to catch it right away with the cover cloth. And since the streaks run to edges of panel, it will be tricky getting EVA streaks completely removed.
  • solardan
    solardan Solar Expert Posts: 36
    Re: LG 300, gooey stuff on panels?

    Update: As described in earlier post, EVA is the laminant (encapsulant) used to encapsulate cells, and lies below the glass, meaning any EVA on the surface of the glass is a contaminant. EVA has been around since the dark ages of solar panel development, and you would think that there would be well-publicized methods of cleaning it off glass surfaces. Obvious sources of such info might be the DIY folks building their own solar panels, or those involved in recycling panels, but this was not the case (recycling usually involves mechanical destruction, or use of heat to delaminate). The best lead that I found was from Peter Lin's site, http://www.evafilms.com, who recommends using isoamyl acetate (also called isopentyl acetate or banana oil), or isoamyl acetate + acetone, when trying to remove EVA from glass after failed lamination attempts. Isoamyl acetate is available as reagent grade (expensive, used in respirator fit testing), for food prep, and as a component in some cleaning agents. I'm still looking for a reasonably priced source of the stuff to experiment with to see if it removes EVA (or whatever) on the panel surface to eliminate the streaking.

    The exact physical properties of EVA depend on its vinyl acetate content, and whatever else has been added to the EVA to stabilize it (such as UV protection). So the "EVA" that may be in your shoes, for example, is somewhat different stuff than the EVA in your solar panels. Unlike glass, EVA absorbs a certain amount of water, which, along with UV, will accelerate its degradation (which is why it does best encased below glass, and with added UV protection). EVA resists ammonia, alcohols, and chlorinated solvents, but is degraded by acetone or butyl acetate, and is slightly soluble in hydrocarbon solvents.

    So it seems plausible that acetone + isopentyl acetate, or isopentyl acetate alone, might be a good cleaning agent for EVA. It's also possible that, being exposed on the surface of the panels, that the EVA (if that's what the contaminant is) will degrade with time due to the combination of water absorption and UV exposure. However, as you can see from the warranty, there is a 30 day window to make a claim once a defect has been identified, and since no one really knows for sure what the stuff is on the panels, I'll probably have to make a claim if the streaking of the panels can't be stopped.
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Re: LG 300, gooey stuff on panels?

    Given the window of opportunity I'd make the claim first and worry about how it's going to be resolved later. The defect is there, no matter what it is or what caused it. It's LG's problem.
  • AuricTech
    AuricTech Solar Expert Posts: 140 ✭✭
    Re: LG 300, gooey stuff on panels?

    Have you tried vinegar on the area where the streaking recurs?
  • solardan
    solardan Solar Expert Posts: 36
    Re: LG 300, gooey stuff on panels?
    AuricTech wrote: »
    Have you tried vinegar on the area where the streaking recurs?

    Yes, and a long list of other cleaning agents (as described in the thread).

    Update: LG engineers are still working on a solution. Last I heard, acetone was not having much effect, but that may be due to its high volatility. The previous reference I gave to using acetone and isoamyl acetate to delaminate panels (i.e., to separate EVA from glass) involved soaking the panel edge in the acetone and isoamyl acetate, but, even if that works, it may be physically impractical to match that condition on the surface of the panels to remove streaks of EVA.

    Anyway, a representative from LG is coming to the house next week and will get a closer look, so I'm happy that they are taking it this seriously. btw, I have not found any other reports of this problem on-line, and it may be an isolated set of panels that are affected, although it's also something easily missed (requires washing or weathering for streaks to show up, and a single streak might go unnoticed, or get attributed to some other plausible cause), so I'm unable to say anything about how widespread the problem may be (i.e., how widespread it is for EVA to contaminate the outer surface of solar panels).
  • niel
    niel Solar Expert Posts: 10,300 ✭✭✭✭
    Re: LG 300, gooey stuff on panels?

    glad to hear they are coming out to see what's going on. do keep us informed as to any further news or developments as they come.
  • OceanArcher
    OceanArcher Registered Users Posts: 3
    Re: LG 300, gooey stuff on panels?

    Hopefully, it's simply an isolated manufacturing defect ....
  • solardan
    solardan Solar Expert Posts: 36
    Re: LG 300, gooey stuff on panels?

    Update: The LG representative delayed coming out to the house, since the LG engineers, with help from the glass manufacturer, came up with a cleaning agent that removed EVA from the panel surface, and they wanted to send me a sample to test first. It then turned out that this cleaning agent wasn't available in US, so a similar product called Hellmanex III ("powerful alkaline concentrate for cleaning cuvettes and optical parts") was also found to work, and a test sample of that was finally sent this week for me to experiment with.

    However, I hadn't been on the roof in over a month, during which we experienced a lot of rain and high temperatures, the result of which appears to be that most of the EVA has degraded and been removed (recall that EVA is known to be degraded by light + heat + water, which is why it needs to be well-sealed within the panels). This was verified by spraying the panels with hard tap water, and then watching for the appearance of streak lines and mineral deposits to form along the lines of EVA contamination. After drying, only very faint lines could now be seen, and there was no significant accumulation of water near these lines. So the simplest explanation seems to be that the EVA has largely degraded over the course of about 2 months.

    Given this experience, I would speculate that EVA contamination of solar panels may be more widespread than anyone realizes since, (1) it is not readily apparent on new panels w/o exposure to environmental conditions that highlight the contamination, (2) installed panels are often not readily accessible for close inspection by homeowners, and (3) the EVA may simply degrade within a few months after sufficient exposure to light, heat, and water. In fact, even in my case where I became aware of the problem, it is only due to EVA’s peculiar resistance to treatment by all of the common cleaning agents that got us worried that a more serious defect was involved.

    So, this was a happy ending to a puzzling surface defect, and I'm happy to be able to again recommend use of the LG 300+ panels. All 20 of the panels are continuing to perform as expected (as tracked by Enphase microinverters).

    Thanks to everyone who has contributed to the thread. It was certainly interesting learning about all the ways that panels can be cleaned, and the kinds of problems others have encountered with panel surface defects/contamination.
  • niel
    niel Solar Expert Posts: 10,300 ✭✭✭✭
    Re: LG 300, gooey stuff on panels?

    was there any trace of the streaks left enough to test the cleaner with? others out there may not wish to wait months for it to go away as they want their pvs to give all of the power they can give now. overall, i'm glad to hear it wasn't serious, but you'd think they would have better control of where the eva goes.
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Re: LG 300, gooey stuff on panels?

    Hopefully they'll step up their QC of manufacturing too and stop panels leaving the factory with the EVA on the surface.
  • solardan
    solardan Solar Expert Posts: 36
    Re: LG 300, gooey stuff on panels?
    niel wrote: »
    was there any trace of the streaks left enough to test the cleaner with? others out there may not wish to wait months for it to go away as they want their pvs to give all of the power they can give now. overall, i'm glad to hear it wasn't serious, but you'd think they would have better control of where the eva goes.

    Hi Neil. The way I was measuring presence of the EVA was by water accumulation and subsequent deposition of minerals/dirt at the streak lines. Since this was no longer occurring, and only very faint discoloration of the panels was present at the worst of the original streak lines, I felt it was prudent to leave well enough alone, and get off the roof, preserving life and limb and panels.

    Also, if the Hellmanex had either succeeded or failed to remove the very faint lines, it wouldn't have been clear what was going on (vs. the testing that the LG engineers did on panels with known EVA contamination). Also, either way, I'd be stuck doing further experimentation: if successful, would some other agent have also removed the remaining faint lines? if not successful, did I use too weak a concentration, or not rub long enough? etc. So it could have meant a lot more testing (on panels that looked pretty clean). Also, the one thing that previously did seem to partially work, Windex, is a similar type of product (alkaline with other complexing agents?) to the Hellmanex. (wrt Windex: Ammonia alone did not work, the Windex may have been simply hiding the streaks with an anti-streaking agent, and it was a lot of work applying the Windex to get partial results, so Windex itself wasn't a good solution.)
    Hopefully they'll step up their QC of manufacturing too and stop panels leaving the factory with the EVA on the surface.

    Yup, they seemed pretty concerned about this, so it's likely something they'll fix (or have already fixed). Given the innovativeness of the panels, etc., I'm happy that the problem turned out to be relatively harmless.

    The 20 300W panels (6 kW) on a nearly South-facing 4:12 roof are performing as expected. The following charts show an interesting week that included both the maximum (clear and cool) and minimum (heavily overcast) production days (as tracked by the Enphase 250 microinverters). Note the inverter clipping (to 250W per panel) on the "perfect" day, which is small compared to the added output you get from the 300W panels at all other times. Also note that we consume about 20+ kWh per day, so are now banking lots of excess power (via net metering) for use during winter months when panels are sometimes covered in snow and ice.

    Attachment not found.
    Attachment not found.

    And here's their current output (about 8:30AM). Note that historically the 3rd panel from top, left is lowest producer, and top, right panel is highest, where this range in production is about what one would expect if the "300W" panels can be anywhere from 295 to 305W rated output (+/- 1.5%).

    Attachment not found.

    And here's what was formerly the worst of the EVA contaminated panels, several minutes after rinsing with tap water. Note how water has remained a thin film, without accumulating at streak lines:

    Attachment not found.
  • niel
    niel Solar Expert Posts: 10,300 ✭✭✭✭
    Re: LG 300, gooey stuff on panels?

    solardan,
    of the panels shown in pic/video #3 which had the streaks. would it be shown that the lower output pvs had the streaks or not?
  • solardan
    solardan Solar Expert Posts: 36
    Re: LG 300, gooey stuff on panels?
    niel wrote: »
    solardan,
    of the panels shown in pic/video #3 which had the streaks. would it be shown that the lower output pvs had the streaks or not?

    Yup, that was something I checked early on, but there was/is no correlation between streak number and panel performance (the best performing panel at top, right of array, for example, had the second largest number of streaks, as I recall). In fact, that's how I became aware of the historical output differences between the panels, and how this matched the expected range of performance (the "300W" panels are actually selected based on being in the range of 295W to 305W; in other words, they don't build "300W" panels, but rather group whatever they get into 10W ranges, and sell them as "290W", "300W", etc.). Given that agreement, and no obvious correlation with streak number, made it pretty clear that the streaks were not affecting performance.

    LG pointed out that the EVA should not affect performance (after all, the panel is filled with it). But I in turn responded that the EVA was causing deposition of material that could block light, and that I was concerned about whether it may also indicate poor workmanship and decreased panel life. However, there was never any indication of significant reduction in power output due to the EVA during the recent months of operation. On the other hand, if delivered a panel that looked like the one below (and resisted all cleaning attempts), everyone, including LG, seems to agree that it should be replaced, whether or not one believed that EVA was the culprit, and would later magically disappear. But it wasn't until it got installed on the roof that the streaks appeared, and then the focus changed to getting the streaks removed via a cleaning agent.