Charge Light On at Night, Only

[deltasig]

Recently I had to motor during the night and I noticed the red Charge light on the Yanmar panel was on. No alarm sound just the light was on and I noticed it when it became really dark.
I checked the voltage at the starting battery and it was a solid 13.8 V. I figured the battery was charging and postponed further investigation for the next morning.

I double checked the R L and all other connections at both ends (alternator - instrument panel), verified the output of alternator and I could not see the red Charge light no matter how much I blocked the daylight. The starting sequence totally normal: turn ignition key -> all lights on and alarm sound, crank, engine starts, lights go out, rev to 800-900 to make sure after a few moments and all look fine. I figured something was not making good contact and as I moved the wires to check them reseated itself. So at night I try to see the result. Same starting sequence but the Charge light goes initially off for a few seconds and then starts getting red gradually and staying red.

The set up of the boat is as follows: House 2 x 200 Ah Gel wired in parallel as one bank, Starting 1 x 60 Ah Gel, Alternator stock Yanmar 55 A, Charger TrueCharge 40, 2 x Kyocera KD135GX panels regulated by Morningstar ProStar 30.
Alternator feeds to isolator to charge both House and Starting. ProStar 30 feeds solar current to the output of the TrueCharge 40 which feeds the House bank only. The Solar satisfy comfortably the day loads and keep the 2 x 200 Ah House fully charged even in the morning before dawn the House shows 12.7 V.

I suspected something may be happening with the fact that there's no current at night from the solar panels. I disconnected all the solar parts, started the engine and the Charge light did not come on.

Any thoughts?

[ShipShape]

Since this only happens at night and disconnecting the solar panels "solves" the problem it may be a blocking diode issue - without a blocking diode the solar panels act in reverse at night and suck energy from the batteries. I don't have any idea if it is possible for a diode to quit working - a little research should answer that. Do you know where the diodes are in your system? If they are part of your charge controller then maybe it is a corrosion problem inside inside of that.

Cover the panels with a heavy blanket during the day and see if that makes the charge light come on, if it does then it is most likely a diode issue. I got diodes for my Kyocera panels from Seelye Equipment Specialists in Charlevoix, MI 49720. If you do diagnose it as a diode problem, the company owner, Don Seelye, is very informative and can help you proceed. Sorry I don't have his number.

[GordMay]

Originally Posted by ShipShape

Since this only happens at night and disconnecting the solar panels "solves" the problem it may be a blocking diode issue - without a blocking diode the solar panels act in reverse at night and suck energy from the batteries ...

Indeed.

[boatman61]

If you have no regulator this tends to happen... if you have one try replacing it... if not then fit one... stops feed back.

[Krakato]

Originally Posted by ShipShape

...I don't have any idea if it is possible for a diode to quit working...

It is quite possible. Working with electronics over the years, I've replaced quite a few. Normally they quit working in a puff of smoke, but sometimes the internal substrate breaks down from over usage or age and they either quit passing current all together, or allow current to pass in both directions.

I agree with the other posts here, a diode on a solar panel is your most likely problem. Most solar panels have a diode on the back side of them (at least all of the ones I've worked on). You can identify the diode as it is normally a small black cylindrical part with a silver line around one side (sometimes with an arrow pointing into the silver line).

To check it, remove it from the circuit and use a continuity tester. It should show open in one direction and closed in the other. If it doesn't, it has failed.

Hope this helps!
Mike

Hope this helps!

[cuthbert]

As described it really does point towards a solar panel blocking diode failing short circuit. The easiest test would be to take a DVM on the ammeter setting and measure that no current flows into the solar panels at night (or with the panels covered) and the engine is running. If you have never used one or owned one Harbor Freight sell a cheap DVM with leads and a battery installed for $1.99,

you measure current like this.....

How to measure current and power volt meter DMM

If is it a diode you can just replace it for a few dollars.

[Dockhead]

One other thing to check is belt tension. Don't ask me how I know, that correcting belt tension may extiniguish an inexplicably illuminated Yanmar warning light.

[GordMay]

Greetings and welcome aboard the CF, cuthbert.

Thanks for the informative link.

(I used to work for a “Cuthbertson”)

[GordMay]

Originally Posted by mhward1337

It is quite possible (for a diode to quit working). Working with electronics over the years, I've replaced quite a few. Normally they quit working in a puff of smoke, but sometimes the internal substrate breaks down from over usage or age and they either quit passing current all together, or allow current to pass in both directions...

Quoting Zach (elsewhere on CF):
“... one of the grand things about electronics... when they start getting hot, their resistance goes up. When the resistance goes up, they get hotter... until they let out the magic smoke.”

Magic smoke (also called factory smoke, mysterious blue smoke, or blue smoke) is smoke produced by malfunctioning electronic circuits.

The magic smoke is observed to come out of electronic components when overheated, often through exposure to an extreme electrical current usually caused by the application of excess voltage through some failure of the circuit. Manufacturers put a little bit of magic smoke into every electronic component (at the factory) and it is this smoke which makes the device work. In support of this joke, once the magic smoke has been released, the device lacks its key component and no longer works. The smoke thus can be thought of as an essential part in the device's function. It is also noted that once let out, the magic smoke cannot be put back in.

* The actual origin is the black plastic epoxy material that is used to package most common semiconductor devices such as transistors and integrated circuits. When it burns, it produces smoke that is blue in color.

More ➥ Magic smoke - Uncyclopedia, the content-free encyclopedia

[ShipShape]

Mike, thank you very much for the 411 on diodes! My Kyocera's do not have diodes, and when I soldered them to the wire I didn't know enough to use a heat sink - and now I know how to check to see if they are toasted.