Voltage but no amperage

[Sunking]

Shorted cell on my bank also crossed my mind 2 weeks ago as well. Not knowing how to diagnose a shorted cell, I switched batteries. Good cell put out 4-6amps at the same voltage as the battery (mid 12v). Switched batteries again and same result...charging rate was the same, voltage was different but the backup battery was a different state of charge.

Shorted or open cells are easy to find. A Hydrometer will tell you what is going on. Shorted cells will measure low voltage, 2 volts for every cell, and never reach charged voltage causing your charger to never shut off. A hydrometer would show as flat SG.

Open cell, the most common problem and cause of 95% failures, will not take a charge. Battery voltage will be low, SG will be low in all cells, and as soon as you put it on a charger, the charger trips off because it thinks the battery is fully charged.

Again hydrometer will instantly point out either problem, or nay problem.

[Switchback]

Shorted or open cells are easy to find. A Hydrometer will tell you what is going on. Shorted cells will measure low voltage, 2 volts for every cell, and never reach charged voltage causing your charger to never shut off. A hydrometer would show as flat SG.

Open cell, the most common problem and cause of 95% failures, will not take a charge. Battery voltage will be low, SG will be low in all cells, and as soon as you put it on a charger, the charger trips off because it thinks the battery is fully charged.

Again hydrometer will instantly point out either problem, or nay problem.

I have agm batteries so a hydrometer is a no go.

I've started to second guess myself at every turn, thinking everything is operator error. Here is my most current test method

Decently sunny day voc is 21.3 and isc is 4-6a depending how I hold the panel.

I have completely removed my meter panel and controller, hooking to a battery bank with 12.44v in it but I'm not getting any current to the batteries. Is that even possible with good voc and isc? Not my first time measuring current with a DMM so I don't think it's operator error.

My next step is to try and find a solar guy in the area and have him look over my system in case I'm missing something obvious

Thanks for all your help.

[sensij]

If you have any loose light fixtures floating around, you could try hooking it up to an appropriate light bulb instead of the battery and see how that goes. You should get a current a bit less than the Isc, and voltage that follows the resistance of the bulb.

[PNjunction]

Looks like you are missing a diode(s) between terminals 2 and 3 from the photo.

In a pinch, you could parallel some 5A common germanium rectifier types in just to prove a point, and then get some "solar schottky diodes" from the usual online or nearby parts emporium to do it right.

I had to do much the same when I blew my matched bypass diodes in my Sanyo panel. How? Too much coffee and I hooked the battery directly to the panel in the wrong polarity.

[sensij]

Looks like you are missing a diode(s) between terminals 2 and 3 from the photo.

In a pinch, you could parallel some 5A common germanium rectifier types in just to prove a point, and then get some "solar schottky diodes" from the usual online or nearby parts emporium to do it right.

No, if you look at the panel documentation (linked a couple posts ago), there isn't supposed to be anything between those terminals. The good Voc tells you that all cells are contributing to the panel voltage. There could be a high resistance connection that gets bypassed through a diode during the Isc test, but in that case, you should see power output equal to half the expected value.

[Switchback]

If you have any loose light fixtures floating around, you could try hooking it up to a 100 W incandescent bulb instead of the battery and see how that goes. You should get a current a bit less than the Isc, and voltage that follows the resistance of the bulb.

Good idea, isolate everything but the panel itself. I don't have any high wattage 12v bulbs. I have 13w 12v CFL...I don't know of the innards of the CFL will work well, so I'll see if a friend on the lake has any 12v incandescents

[Switchback]

No, if you look at the panel documentation (linked a couple posts ago), there isn't supposed to be anything between those terminals. The good Voc tells you that all cells are contributing to the panel voltage. There could be a high resistance connection that gets bypassed through a diode during the Isc test, but in that case, you should see power output equal to half the expected value.

I agree sensji, manual shows no diodes between 2-3. Diodes all ohmed out properly. One way with .450 resistance if I remember. I'll try the light bulb test and see if that gets me anywhere. Thanks again for the ideas and help

[sensij]

Good idea, isolate everything but the panel itself. I don't have any high wattage 12v bulbs. I have 13w 12v CFL...I don't know of the innards of the CFL will work well, so I'll see if a friend on the lake has any 12v incandescents

MR16's are pretty common. If you can only find one with lower wattage (20 W or 50 W are typical), just keep your panel tilted so your Isc is less than 4A (for 50 W), or less than 1.6 A (for 20 W).

[sensij]

Thinking through it some more, having a high resistance connection somewhere in the panel would explain all of the results so far, and is very consistent with the known defects of this panel model.

Voc shows OK because no current flows, and all cells can contribute to voltage.
Isc shows OK because one of the 18 cell strings is intact and producing current, while the 18 cell group with the high resistance is bypassed through the diode.
Load test on the battery shows zero current, because with only one 18 cell string fully intact, not enough voltage is generated to drive any current into the battery.

Since you have access to the junction box, a simple test for this that would not require a light bulb or other test load would be to do the Isc test, except remove the bypass diodes first. That would force the current through the high resistance connection, and should meaningfully drop the measured Isc value from what was measured with the diodes in place. With any luck, enough power gets dissipated to heat up the bad connection and discolor material around it, so you can see it (assuming you don't have an IR camera so you could just image it)

[Switchback]

Thinking through it some more, having a high resistance connection somewhere in the panel would explain all of the results so far, and is very consistent with the known defects of this panel model.

Voc shows OK because no current flows, and all cells can contribute to voltage.
Isc shows OK because one of the 18 cell strings is intact and producing current, while the 18 cell group with the high resistance is bypassed through the diode.
Load test on the battery shows zero current, because with only one 18 cell string fully intact, not enough voltage is generated to drive any current into the battery.

Since you have access to the junction box, a simple test for this that would not require a light bulb or other test load would be to do the Isc test, except remove the bypass diodes first. That would force the current through the high resistance connection, and should meaningfully drop the measured Isc value from what was measured with the diodes in place. With any luck, enough power gets dissipated to heat up the bad connection and discolor material around it, so you can see it (assuming you don't have an IR camera so you could just image it)

I really like your thinking on this. From the minute I had problems I figured it was a bad diode or connection that was strong enough to pass voltage but too weak to transfer current.

Instead of spinning my wheels and wasting forum members time on another kc120 problem I called the company and they are going to send out a replacement.

In the meantime I'm looking for a good charger for my battery bank but I'll open a new thread for that topic

Thank you guys very much for your help. Even though I never fixed this problem I've learned a ton from you guys.

[PNjunction]

Ok, my brain went into rewind, and sure enough we are going through an issue that hits about all solar boards.

Early models of the KC-120 went defective, even though the diodes test good. Back then, Kyocera would replace the panels with fixed models of the 120 or upgrade to the 130 - but that was about 2009 or earlier.

Members on the NAWS site went through this (replicating the same thing we are doing here pulling our hair out), as well as those on Cruisersforum. Op sgtPluck there in 2011 describes what you are going through right now.

Unfortunately, you got ripped off like so many others.

So do us a favor and destroy them so they don't end up back into the used / questionable seller market to fool another newbie again. Or use the good side for smaller non-critical educational purposes.

There is an honest seller on the UK version of ebay, which states up front that these are not working, although "tested" to see if they put out *something*.

Kyocera made good to the original owners, so I'm not slighting them.

Brought back some memories - in an article in the 2003 edition of Homepower magazine, these babies cost about $520 each.

[littleharbor]

Thing is, from my experience, Kyocera will ask for the panels back in the packaging that the refurbished replacements came in. They are taking the failed modules and completely rebuilding a them. They will be relabeling them as KC-120-r. They will be rated at 102 watts at STC. Kyocera. will pay for all shipping.

[littleharbor]

I've seen sellers on eBay and craigslist offering these defective modules (sometimes by the container load) though not mentioning or, maybe not knowing these are bad. I have called some of
them on this and have been told things like "They are metering ok" or haven't gotten a reply. Thing is, Kyocera knows these are bad modules and has been really straight up about replacing them. The 3 batches I have had replaced over the years as long as the serial numbers check out within the particular timeframe (1999-thru 2002) they were replaced at no cost. This says a lot about the integrity of Kyocera. 16 years down the road the majority of todays manufacturers likely wont be around to honor their warranties.

[Switchback]

Yea. In my lack of experience I picked Kyocera because they had a good warranty. The company I hit the panels from (ML solar) is going to replace or exchange them according to the guy who called yesterday. He's not the one who handles the warranty work so he didn't know the process for exchanges. I will not be getting more kc120's though.

If they let me keep the panels I'll play around, see if I can find and solder a bad connection so I can use them somewhere else, otherwise they're going into the garbage.

[PNjunction]

Hi LH, glad to see you made it. Yes, that is quite amazingly great of Kyocera to do. That really speaks VOLUMES about integrity and kind of sets the bar that others need to match if honor, and not just profit is involved. Good work!