How to test for a short from solar panels?

[demonboy]

Hi all,

Another 'basic' question. I've tried Calder but can't quite interpret the following whilst testing for a short:

Disconnect hot and ground size of circuit at the power source and at the equipment, and insert meter in place of the equipment

What's throwing me is the difference he is establishing between the 'power source' and the 'equipment'. In this instance the power source (solar panels) IS the equipment, isn't it?

To expand further: I have six solar panels on two banks which may be flipped from vertical to horizontal (they're mounted as drop-downs from the pushpit). One is smashed with a perspex cover and still working but with some trapped water inside. I suspect the monsoon got the better of it in my absence. I think this may be my problem but I don't want to take the perspex off unless as a last resort. When the panels are connected to the battery via the regulator, sometimes they put in the correct amount of amps, but other times they are showing minus amps. Yesterday they were showing -18! If I disconnect just the positive from the batteries, they were still showing the minus figure.

Last week we had a problem with a short manifesting itself by making the guardrail gate get white hot. I hadn't noticed this until I burnt myself on it quite badly. Thank god that was the only injury but it's a stark lesson in the dangers of boat electrics!

How do I test for the short?

 

[VicS]

If you have a fault which is passing sufficient current through the guardrail gate to make it white hot you have something very seriously wrong and potentially very dangerous.
That must have come very close to starting a fire and I would think is likely to have caused damage to cable insulation.

You would be well advised to disconnect all power sources until the cause is identified and rectified and any damaged cable also replaced.

I'd guess you have a negative connection to the guard rail at one point ( possibly by design, but more probably due to a fault) and a positive connection due to a fault at another point.

The discharge ( minus amps ) through the solar panels indicates a major problem in their circuit ... it should just not happen.

I'd disconnect the whole lot and after sorting out the above problem check each part, ie each panel and the regulator, reconnect it all a bit at a time ( observing any special instructions for connecting the regulator) checking what happens as each part is reconnected.

I don't have Calder's book. I Have no idea what one sentence taken out of context means. Maybe some one with the book can help.

 

[demonboy]

What I find curious though is that nothing has changed in the last two years. I have rewired nothing. The only thing that has changed is the fact I have been in some terrible monsoon weather and water has got into the solar panel.

The white-hot incident made me examine the mains wire cable, thinking perhaps there was no ground from the pontoon or perhaps faulty wiring in the plug, but I found nothing untoward. I rewired it anyway and since that one time the problem has not returned. None of the cable was damaged.

But the minus amps in the circuit of the solar panels still persists. Perhaps that one time I left it on minus amps long enough that it heated the guard rail, but now that I am aware of the problem I have not allowed it to get that far. Whatever, what could be the cause of this? Would trapped water in a solar panel cause this problem? And what am I checking when I inspect each panel? A continuity test of some kind, but what? I'm at a loss here.

The quote from Calder comes under the heading 'Chart 3.4 Testing for Short Circuits with an Ohmmeter'.

 

[pappaecho]

By the sounds of things the blocking diode (s) on the array are not working and so you have got the batteries shorting to the guardrail - in effect a sort of welder. The blocking diodes, stop reverse current flow from the batteries back to the solar panels
As VicS says, disconnect the lot and work you way through the damaged and undamaged panels.
Each panel should show an unconnected voltage of about 18 volts if it is a 12 volt panel and roughly double if 24 volts. That should be your starting point to establish good panels from bad. Usually each panel has a built in blocking diode, unless they are are all linked up to a controller which has its own blocking diode built in. As you have 6 panels probably the latter.
Regardless of the Perspex cover, you need urgently to sort it out before you have a major fire

 

[William_H]

Solar panels

They generally have a junction box under the panel which often contains the blocking diode. If you have a short circuit then it is most likely that there is mechanical failure of the insulation of the box. You should check each panel closely at the mounting points.
It is highly unlikely that a panel with water in it will short and pass 18 amps backwards. The usual mode of failure is corrosion leading to open circuit. You need to get at the broken panel and confirm it is still doing its job. ie producing 20v open circuit and an appropriate amps into a 12v battery. Expect .5 amp per 10watts of panel.
Presumably you have a regulator and it seems unlikely that the current could flow from batteries to panel and any short via the regulator. In any case there should be fusing or circuit breaker to protect the wiring from this possibility. I can't imagine the railing getting hot before any fuse failed or wiring burnt through.
So much so that I suspect the hot stern rail was coincidental to another problem.
Anyway close physical inspection is the first step.

Regarding Calders comments. He was obviously generalising that in any system you have a source of power and a drain of power. Short circuits are best found by disconnection and check at various stages as to when the short symptoms dissapear. good luck olewill

 

[demonboy]

Thanks, all. This is my job for today so I'll post back my findings.

Yes, the solar panels have blocking diodes, indeed they each have two blocking diodes (they are Kyocera 40w panels). All junction boxes were bone dry save for one which had some minor corrosion. See the pics below of my port side panel bank:

Pic 1: The end panel junction box closest to the regulator, which is the black cable that goes to the left. The other red cable is the other solar bank on starboard side of the boat. The cable coming in from the right are the other two panels in the bank.

Pic 2: The middle junction box on the end panel.

For the record the panels go to a regulator (Steca PR 2020) and this has only ever shown the correct amount of amps coming in from the panels and the correct amount of amps going to the batteries.

@William - I suspect you are right about these being two separate issues.

 

[demonboy]

OK, seems the findings on the solar panels are all ok, but I appear to have identified a different issue.

I've disconnected everything and I get continuity between my rigging and the negative terminal on any of my domestic batteries. Is there some basic electricity theory that I am missing here? It's pretty obvious I know little about boat electronics but I certainly don't understand whether this is supposed to happen or not. Perhaps this suggests that my through-hull fittings are grounded? If this is the case then how does this show continuity? I did also check with the volt meter to see if there were volts running from the positive to the rigging, and there aren't. I do have a through-hull ground plate that I'm going to check next.

 

[whipper_snapper]

When the panels are connected to the battery via the regulator, sometimes they put in the correct amount of amps, but other times they are showing minus amps. Yesterday they were showing -18! If I disconnect just the positive from the batteries, they were still showing the minus figure.

When you say they were showing minus amps, do you really mean that you know that current is flowing backwards through the panels?! Or that you battery system as a whole is discharging at this rate? As shown on some kind of battery management system? Because, on the face of it, what you describe is impossible. If you disconnect the solar array positive you cannot be discharging through them and certainly not at 18A. So I would be 99% confident that you are not seeing current flow through the panels but some other drain on the system.

[edit] I've disconnected everything and I get continuity between my rigging and the negative terminal on any of my domestic batteries. Is there some basic electricity theory that I am missing here? It's pretty obvious I know little about boat electronics but I certainly don't understand whether this is supposed to happen or not.

You have clearly not disconnected everything! I cannot tell what you have actually done, but, do you see 12V if you put the meter between the rigging and the pos terminal of he domestic battery? And 0V between the rigging and the neg terminal? In fact, if you put a 12V bulb between rig and battery +ve, instead of the meter, does it light? (a better test because a digital meter can give strange results because it pulls such a tiny current)

I certainly don't understand whether this is supposed to happen or not.[/I]

Many boats are wired like this, with dc neg bonded to everything and it is probably perfectly OK. Other boats (like mine) have completely 'floating' dc systems where the rig and other bits of metal are isolated from each other and from the dc neg.

But of course your guard wire getting hot cannot be right! There must be some pretty fundamental problem and it must be happening to wiring capable of carrying a heck of a current without melting, and it cannot be at 12V because your batteries would have fried and the wiring burned to a smelly soggy pile of goo.

SO. It really must be something from your shore power. Many boats have AC 'neutral' commoned with dc negative. It would not be at all surprising to find that is the situation on your boat. Is there an mains 'AC panel' as part of your main electrical panel? If so I would bet that is the case. So that is where the problem is likely to be. Reversed 'polarity' on incoming mains is common and I think it may cause such problems. Do you have an isolation transformer (or galvanic isolator) in the system? From the sounds of your system you really (really, really) should. Given your setup, I would be very wary about plugging into shore power without at the very least checking that live/neutral are correct way round.

@William - I suspect you are right about these being two separate issues.

+1

 

[demonboy]

OK, I appear to have discovered the cause of the problem. My suspicions started when I said I had disconnected everything, because I had (at least from the negative terminal), yet I was showing some kind of continuity between the negative terminal and the rigging. The only explanation was that the negative and positive terminals were shorting so I disconnected both positive and negative connections so that I had nothing connected to the battery, ran a continuity test between the positive and negative and voila, continuity.

Correct me if I am wrong but a battery shouldn't show continuity between the +ve and -ve terminal. Therefore I have a battery short. This pretty much explains everything.

@whipper_snapper - hope this answers some of your questions. Since posting I have been discussing some of these questions with my friend on 'Pure Chance', who explained that indeed many boats are wired this way.

I suppose the next issue is to establish how this happened, or is this merely a manifestation of tired, old batteries?

With the problems identified in mind, does your explanation of reverse polarity from the mains still stand? Like I said before I have rewired the mains cable and no longer have the have the white-hot guardrail problem, so it's possible I've solved that problem. When I rewired it I noticed the earth had never been connected, so could this have solved the problem? I don't have a galvanic isolator but would it be naive of me to assume that my Mastervolt mains charger has one built it?

 

[demonboy]

Ah no, wait a minute. It's only showing continuity ONE WAY, when I connect the COM to the +ve and ohm cable to the -ve. I've guessing this is merely showing the polarity movement between the two so that makes sense and is therefore not the answer to my problem, damn it.