Solar Voltage Drop mystery....

[Rewilding]

Hi all - hoping someone might be able to help troubleshoot a solar charging problem. The problem is: it isn't working.

 

The set up has worked perfectly since I installed it 5 years ago, and nothing in this system or any of the electrics on the boat has been changed recently. Yesterday, I noticed the epever xtra MPPT controller was showing the panel had light, but it was not charging. Vpanel = ~12.4v and  Vbatt =12.6v at that time. This was around midday, it was working in the morning.

 

I have a second identical and fully working MPPT controller, so first thing I did was rewire into this one - same problem, so not the controller.

 

I was also able to switch out the panel with another similar one - same problem, so not the panel (I also tried the second panel on both controllers).

 

So presumably the problem lies between controller and batteries? The batteries are healthy, they charge fine with the engine running and they hold their charge. One thing I noticed is when I ran the engine and Vbatt went up to around 13.4V, the Vpanel shown on controller was ~13.1v. I also tested the wires from the panel when disconnected from the controller, and it was giving 31V.

 

So, something in my electrics is dragging it down?

 

I took out a single leisure battery and connected the controller direct to this, so nothing else on the system. Same problem - still showing not charging, and volts always about 0.2v below Vbatt.

 

This has my completely stumped, so any other thoughts very welcome!

 

Thanks,

Andrew

 

[Onewheeler]

Slightly random thoughts:

 

Has the controller been disconnected from the batteries with the panels connected? The controller should be connected to the batteries first (most controllers anyway).

 

Disconnect the panels and connect an ammeter straight across the terminals to see what open circuit current they're producing. It sounds a bit like a high impedance connection in the panel circuit (possibly in the panel, I had a vaguely similar problem with a damaged panel once).

 

err... out of ideas for the moment.

[WotEver]

22 minutes ago, Onewheeler said:

Disconnect the panels and connect an ammeter straight across the terminals to see what open circuit current they're producing.

^^^^ this.

 

At 12.3V input, the controller is never going to charge the batteries.  You're not moored under trees, are you?

32 minutes ago, Rewilding said:

So presumably the problem lies between controller and batteries?

I can't see how, myself.

[Rewilding]

13 minutes ago, WotEver said:

At 12.3V input, the controller is never going to charge the batteries

Full bright sun all day yesterday and today, clean panel. Open circuit is 31V, drops to 0.2v below battery voltage as soon as it's connected.

 

I have been disconnecting panel first from controller whilst trying to fix the issue, but the connections hadn't been touched before this problem.

 

Thanks for the thoughts, it is quite puzzling.

 

My only thought for it being between controller and batteries was because both panel and controller appear fine. I even tested the diodes in the panel and all working fine, and as I said, 31v open circuit.

 

Hmmm...

[WotEver]

And the short circuit current?

[Onewheeler]

My dodgy panel was producing almost full voltage output but not much current. Check the s/c current.

[David Mack]

21 minutes ago, WotEver said:

And the short circuit current?

Sounds time like something has gone high resistance in the panel, but the connection isn't completely broken. So full voltage with the panel open circuit, and next to nothing under load.

[WotEver]

3 minutes ago, David Mack said:

Sounds time like something has gone high resistance in the panel, but the connection isn't completely broken. So full voltage with the panel open circuit, and next to nothing under load.

Yup, hence the importance of the SC current reading. 

[Rewilding]

I don't know how to measure the short circuit current, presumably disconnect panel and measure across them as if voltage, but with multimeter set to 10a?

 

 

[Tony Brooks]

12 minutes ago, Rewilding said:

I don't know how to measure the short circuit current, presumably disconnect panel and measure across them as if voltage, but with multimeter set to 10a?

 

 

 

Yes but only if the output is less than 10 amps. Otherwise its joint the two ends and use a clamp type DC ammeter.

[Rewilding]

I wasn't able to measure that short circuit current - but given that I have tried it with two panels with the same problem, it must be in the wiring from the connectors to the controller, right?

 

One of those connectors was very (despite not being in the sun) - could this be the source of the resistance?

 

I've now actually just dismantled that connection and put it back together - all looked good, no signs of corrosion of poor connection. Having put this back together it is now....working.

 

Problem solved but I feel like i've not satisfactorily solved the mystery...?

[BEngo]

If the connector was hot, but not in the sun or resting on a hot roof then there was a fairly high resistance in the connection.  In the absence of corrosion this was probably caused by poor contact between the metal bits.  Had it been pulled,  kicked or otherwise accidentally strained/displaced recently?  Has a bird had a go at it even?

 

You may never know, but it is one to keep an eye on.

N

[WotEver]

37 minutes ago, Rewilding said:

One of those connectors was very (despite not being in the sun) - could this be the source of the resistance?

Yep

[Sea Dog]

2 hours ago, WotEver said:

Yep

Beautifully put!  I've read all the way through this thread thinking: "dodgy connection, dodgy connection".