Criteria for defect rate of Junction boxes of PV modules 3

[LJthien]

I have checked some related standards for this matter but still not get to know.

Our Solar Farm has operated for 6 months and we recently found that some Junction Boxes of PV modules (JB) had been bulged. We have investigated and pointed out the root cause that the connection of 1 of 2 JB cords was overheated at 110degC while the acceptable temperature is 85degC maximum as per IEC 62790(4.1).

In the result of the finding, there are 195 bulged JBs out of total 107.625 JBs. As the defect rate is 0.18%, I feel that the rate is not so high.
However, It must be convinced by a standard s' criteria to wrap up the non-conformity. Could you advise me which document I should refer to assess the defect rate of the defective JBs.

Thanks a lot.

 

[zappedagain]

No takers, eh? Maybe a post in the 'Engineering Codes, Standards & Certifications' section might get answered.

Should the OP link to this message from that board, or is a week without a response long enough to not be considered double posting?

Z

 

[Sparweb]

I would not consider this double-posting if LJthien tried asking in another area, since a sincere effort was made to pose the question in the right place the first time.
I do not frequently read this discussion section because it's not my in discipline, but it does interest me.

Unfortunately, I do not have a specific answer, either. I can pose a few follow-up questions because they may put the failure rate into context.

What is the cost to find and replace the failed JB's?
Presumably the panels are inspected on a regular basis, so this would be the time expected to discover the hot spot. Are they inspected thermally? (Infrared camera?)
If there really are 100,000 panels in the field, how many people are required to maintain the entire system? How often are they inspected?
It would be an important difference if you said that 2 people will inspect the entire system once per month, versus saying that 1 person inspected them all for the very first time at 6 months. The implication is that your system could have had failed JB's since the early days of operation, but you didn't know about it until now. What revenue could have been lost in that time?

Failure rates are expressed not only as a percentage, but also in terms of time. Your experience is that after 6 months the JB defect rate is 0.18%. Does that mean 0.36% annually? This may actually be a factory defect rate that shows up quickly (first months) and much fewer failures are expected in the next years of life. Depending on how often you inspect the system (my questions above) you could show that the rate of failures if dropping, or increasing. That would point you to the rate of factory failures VS. the rate of service failures.

I once spoke to a residential PV installer who told me that about 2-5% of the panels they receive are defective, and they normally bring 10% extra panels to any job site to be sure they can complete the commissioning on schedule. If the defect is found before installation, the manufacturer they deal with will cover the cost of replacing the panel. They will not cover the cost of field repair or shipping the defective panels to the factory for repair. This leads to many PV panels in North America being sent to the waste disposal or recycling every year.

http://www.sparweb.ca

 

[LJthien]

Thank you very much SparWeb,

In my case, panels are of series connection type therefore an overheated JB finally damaged its diodes which decreased the voltage of the whole string. That means the overheat had been formed recently.

Yes the JBs were scanned by infrared cameras as a measure for investigation of the trip of the inverter. It took us almost half month with 15 cameras for scanning all.

On the other hand, we can not link it to production defect rate as such kind of massive production is tested on production Lot/batch basis therefore the production defect rate always is at acceptable rate that Manufacturers obviously will not disclose.

Decreasing performance is an other issue since RTI is defined by IEC standard and acceptable for increased operation temperature of JBs after a certain service life.

I still hope somebody who ever suffered from it will help me.
Regards.

 

[argotier]

Could you advise me which document I should refer to assess the defect rate of the defective JBs.

Perhaps this can help you. The standards being referenced are ISO 2859-1 and ANSI-ASQ Z.14.

 

[IRstuff]

Unless specified in a contract, there likely isn't a standard, since it's a risk/benefit trade-off. A manufacturer looks to have the lowest failure rate that doesn't break the warranty cost bank and doesn't break the perception of reliability bank. Other than that's it's pretty much up to the manufacturer to do the right thing.

You say they're "bulged," but you've apparently not had outright failures, per se.

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[LJthien]

You say they're "bulged," but you've apparently not had outright failures, per se.

Thanks. Some bulged JBs had been found with broken diodes after further investigation on those. We acknowledge that a heat source had heated both JBs housing and its diodes and then the diode was broken as a result of continuously standing over temperature.

 

[LJthien]

Perhaps this can help you. The standards being referenced are ISO 2859-1 and ANSI-ASQ Z.14.

Thank you. Yes, massive production manufacturers always apply such standards for Batch / Lot inspection during manufacturing. We purchaser unfortunately never specify it in the contract with both manufacturer and end-user (Owner of Plant) for operation stage.

 

[Sparweb]

I have experimented with repairing PV panels when the diodes fail in the junction box.
Here is an Eng-Tips thread where I discussed my project:

https://www.eng-tips.com/viewthread.cfm?qid=453688

First off: there is no way you will get repair instructions from the PV panel manufacturers. They would rather you purchase a new panel.
However, the skills, tools, and materials needed to repair many types of PV panel failures are not out of reach, if you're willing to do it.
You have enough failed panels (taking out a whole string it seems) for this kind of repair to be worthwhile. The effort you'll need for this to work is for an engineer to design a suitable repair process, document the procedure rigorously, and validate the repair with tests. Once the engineering's done, you'll have a repair method that you can use on a routine basis. Maybe even on any panel with hot-spots before it kills the string.

http://www.sparweb.ca

 

[LJthien]

I have experimented with repairing PV panels when the diodes fail in the junction box.
Here is an Eng-Tips thread where I discussed my project:

https://www.eng-tips.com/viewthread.cfm?qid=453688

Thanks for sharing and sorry for abruptly turning to your issue herein since "

https://www.eng-tips.com/viewthread.cfm?qid=453688"

has halted. What was the reason for replacement of the diode? The clarity will benefit me as well.

 

[IRstuff]

Was the overtemp caused by the PV equipment or something else? If it's something else, the the defect rate you quote is inapplicable to the manufacturer, since it is not liable for damage incurred through misuse or abuse of the product.

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[Sparweb]

LJthien,
Since you're interested, there's much more to the story here:

https://www.fieldlines.com/index.php/topic,149794

This friend of mine has been bedeviled by some long-term issues with his large PV system, the electric utility, and the reliability of the panels he used. The problems tend to be "annoyances" not full-stop problems, so Pierre lives with them. During the PV repair seminar we heard many sides of the problem of PV panel reliability and repairability, from installers, users, and electrical engineers. What we learned was not encouraging because it informed us of the limited freedom Pierre would have in repairing his own PV panels unless he was prepared to have a repair engineered for him, to show somehow that the UL/CSA label is still valid after the repair. The electric utility may have to sign off on it, too. The panel manufacturer would be unlikely to participate or approve. By our experimental efforts, we know that repair is possible, now. Unfortunately, the burden of proof and approval is beyond the pocketbook, even for a system of his size.

Pierre is also unable to determine the exact cause of the failure of any given panel's diodes, but he has a theory, and some evidence to prove it. From time to time, the electric supply that is measured at his house (in the microinverters under his PV panels) can exceed 130VAC. The utility company says the problem is from his PV system, but he believes that the wires at the step-down transformer are tapped too high to begin with. The whole issue with the transformer hook-up is a comic farce so I won't hijack your thread any more than I already have with it here.

The point is that tracking down the cause of PV panel faults is probably very hard. It can be overheating but there are other causes. They are exposed to the natural environment with fluctuating temperature and humidity and insolation for a long time before they fail. They are also subject to electrical loads which fluctuate. And then there are variations in the manufacture of every semiconductor part, and the method of assembly when the PV panel is made.

It makes sense to me that repair of PV panels should be possible, but they seem to be cheap enough to be handled like a commodity. Lots of defective units are discarded every year.

How seriously would YOU take the subject of repairing PV panels, LJthien?

http://www.sparweb.ca

 

[LJthien]

How seriously would YOU take the subject of repairing PV panels, LJthien?

Sincerely thank you for your ultimate sharing and supports.

In my case, we are handling the issue under the obligation of an EPC contractor. We have to do it with no excuse.
The PV modules having defective JBs are being replaced by spare ones. The disassembled PV modules will then be transported to a PV module factory for repairing after successfully passing some commonly agreed tests therein.

 

[Sparweb]

It's interesting to learn that the PV manufacturer is accepting to repair the panels rather than force you to discard them. Was this a provision made in the EPC's contract that made this obligation? Perhaps you feel it's not proper to answer that; I am curious. Also, may I ask roughly how large the total installed PV power or area is?
This will be valuable information to me if I find myself discussing the subject of repairability of PV panels in the future. Knowing that, at some scale, repair is possible instead of discard would put to rest some (not all) of the concern about wasted material that I encountered when discussing this with local agencies concerned with the promotion of renewable energy here.

http://www.sparweb.ca

 

[LJthien]

It's interesting to learn that the PV manufacturer is accepting to repair the panels rather than force you to discard them. Was this a provision made in the EPC's contract that made this obligation?
Also, may I ask roughly how large the total installed PV power or area is?

It is not complicated like you think. Discarding defective modules results in replacing them by new ones, which is more costly than repairing them (exactly they just replace broken JBs).
In term of EPC contract, it is so called warranty provision for which EPC contractor & Manufacturer can repair the equipment within warranty period, subject to approval by Client (Plant Owner). The plant I am talking about is of 25MWac, for your reference.

 

[Sparweb]

Thank you LJthien,
This discussion as been very interesting for me, like it's been for you.

http://www.sparweb.ca