I have an array of 24 KC-120's with a manufacture date of 10/2000, installed 1/2001. Last summer two panels failed (each panel has two sections internally connected in series, one section in each panel failed, so the panel produced only half its rated voltage). This year I have four panels that failed.

Anyone have experience with KC panel failure? Am I an isolated situation? How can I get on Kyocera's case about this 25% panel failure rate?

Its funny you should ask! I haven't been to solar guppy in months, but I just discovered that one of my 24 KC120s was not producing and its
failure mode was that 1/2 of the two sections in series is open circuit.
The bypass diodes were therefore forward biased an dropped 1 volt.

I have just emailed Arizona wind and Sun to ask about how to
return the panel under warranty. The fact that it has happened
to you many times is not good news. I bought my panels approx
3 years ago (summer 2001) and did not have any problems with them until now. I hope this is not as bad as the Xantrax Sun tie fiasco, where
I had to install the sun-tie 2500 4 times!

John Vallerga
Oakland, CA

Here's what I just wrote to Henry:

After reporting to Kyocera that five of my panels were producing half-voltage this summer, and the problem is not by-pass diode failure, Kyocera contacted me and is replacing all of the panels (i.e., 17 have not appeared to fail yet). Apparently I had gotten panels from a production run in 1999/2000 where bad solder was used, and it developed a high resistance and often failed. Unfortunately, this defect could not be tracked through panel serial numbers.

I removed the old panels (and re-shingled the roof while I was at it), and installed the new ones this weekend. The difference in performance with the new panels is phenomenal…. The peak output from the old array and inverter was around 1600 W AC (with all old panels functional!), the new array and inverter put out 2150 W AC. This is about a 30% improvement in output, and much more in line with the performance that I was expecting to receive from the system when I engineered it. Clearly, even though these old panels "worked," they did not produce their rated/warranted output.

My thanks for Kyocera for replacing the whole set, but clearly I was having problems with the whole array. I am wondering if Kyocera has issued a product recall, or has a formal plan to track down registered owners and report the (possible) problem to them. I don't think that I sent in my warranty registration and panel serial number list to Kyocera, so I can't fault them for not contacting me, but I would hope that they are proactive in alerting known owners of the potential KC-120 problems.

Henry, I just thought I'd bring you up to date, and inform Lisa and Michelle of the incredible improvement the new panels are giving. I will connect the solar-guppy software to the inverter soon and get some performance data for the system. Fun!

>>>>>>>>>>>>>

In talking to the Kyocera rep, having a single panel fail is not cause for panic; that can/does happen. However, if you lose more, or if your system is somehow not producing close to what you would expect (ie., 1600 W instead of the 2100+W I had engineered it for), then it may be worth pursuing.

I would suggest contacting Kyocera directly. The key person there is Michelle Rabieh, (800) 544-6466.

john

I installed two banks of Kc-120, 20 each, with Sunnyboys and have been operating since April 2002. Last September the output powers began to drop. You can see the effects on the FSEC site - Lary Home Homestead. The voltages are getting so low on lne of the banks that the Sunnyboy is dropping out. FSEC is going to take a couple of suspect panels back with them and anlyze the problems. Has anyone heard of more issues like this besides the ones already on this site?

My underperforming (30%+) KC 120 panels finally failed and were replaced free of charge by Kyocera.

I had one complete failure and 3 half failures out of a lot of 20 panels that were manufactured and purchased in September 2001. Serial numbers for this lot was 01932B0626 through 01932B0645.

I found that the only way to measure the failures were under load conditions - the panels connect to the inverter and under sun. I found that if I measure the panels disconnected from the inverter the voltage could float up and the panels would measure OK.

The measure voltage of these failures under load conditions were less than one volt.

As posted by John_Bloom, I followed his recommendation to contact Kyocera directly - namely Michelle Rabieh (800) 544-6466 ext. 7726 - she was very helpful!

I have 72 KC 120/125 panels - now all three arrays of 24 panels are performing the same - yes, yes,yes!

I had the same failure occur on 3 KC120 modules in January 2005. I had failed module on two different dates in 2004 and after paying for shipping replacement modules I finally asked to talk to a technical rep, hoping to find out how many times I was going to have to ship the panels, two at a time. I was contacted and my remaining 30 panels were replaced, shipping pre-paid. I did send in the warranty information and Kyocera was not pro-active in solving their problem. I found the service section at Kyocera was more concerned with who was the original seller of the panels and having them handle the problem than with supporting the user. I'm not sure why they wanted the original seller to handle the replacement, except their return 'policy' seemed designed to insure Kyocera was never at risk in shipping a replacement before they had the failed unit in hand. As expected, no offer was made to compensate for repair cost. The new modules do indeed perform better; I measured a 20% improvement.

So I was performing system battery maintenance yesterday when I happened to get my electric bill for the month. Low and behold I've used 300kwh? How could that be since this is the time of year where I generate more than I use.

I also notice that my C60 is showing very low numbers...and start the whole investigation process ... I discover that I've got 7 dead panels out of a total array of 24. But, upon closer inspection, I notice that if I push on the backside of the panel I actually can get a connection of the cells. Bad solder joints I ask myself?

I call my good friends at Carlson Solar and ask Mary if there is a known problem regarding this and she replied that, yes, there is but only a certain date codes...which I happen to have for my panels (2001.4) and that they will come out and replace them with KC130GT's at no charge to me!...and they will replace ALL the panels not just the dead ones now!

I am glad Kyocera did this as I wonder for how long my system has been under performing...

System currently is 24x KC120, Trace SW4024, a C60 and C40 but will be going away to become an SMA4000U with the new panels. When I put it in it was the ONLY inverter that was allowed for grid tied applications at the time.

you're getting a bonus with the replacement of those pvs with an extra 10 watts per pv. let's see 24x10=240w. it's like you'll have the equivalent of 26 good kc120s now with the implementation of the 24 kc130s.

...and I'm going to add a a whole new inverter setup with a west/southwest array of 10x 175w's to cover all the usage...now with solar radiant floor heating from the hot water and all the other things I'm completely off grid right in the middle of San Diego. SDG&E showed up about my gas uasge a few months ago. Seems, that they thought I had a "leak" since I use so little....service person had never seen a house that was so hi-tech (Crestron automation system) and so efficent.

Like to "sneak" a wind system in just to play with it so so if I could harness the morning and evenings winds down the valley. My neighbors get a kick out of me...

in the future they may be coming to you for advice on how they can do it too.