On Tue, 11 Aug 2015 11:13:39 +0100, Dave Liquorice wrote:

On Tue, 11 Aug 2015 02:33:14 GMT, Windmill wrote:

But the charging current you'd get from a north-facing window
would be
tiny. Did you measure it?

ISTR (badly) a few mA at 0.5 Hz or slower.

I don't follow. Where does a frequency of 0.5 Hz come into it?

The LED flashes, when on it takes most of the power from the panel,
off the battery gets it. The LED flashes as when on it pulls the voltage
below that required for the LED conduct so it stops this means the
voltage rises to the conduction point, repeat.

In other words, a faulty panel, one that's gone very high resistance in
the photo-cell array cct to the extent that the couple of milliamps drawn
by the self flashing blue LED indicator lamp is sufficient to drop the
voltage below the 12v mark during its on period.

It's been a while since I last bothered to test my own panels but istr a
short cct current in full direct sunlight of just over 100mA and around
80 to 90mA when charging a 12v SLA at around the 13.5 volt mark. At this
current level, it was just possible to see the 2mA modulation effect from
the self blinking blue led on an analogue multimeter.

For something *this* buggered, it will be worth prising the rubber/
plastic bungs out to undo the screws that retain the case halves together
and take a closer look.

If it's just a single cell out of the 30 or so that's gone HR, it's
probably worth strapping it out and accept the slightly reduced charging
current that results compared to its original spec. If you're lucky, it
might simply be a corroded connection you might be able to make good one
way or another. It's certainly worth taking it apart and taking a closer
look before chucking it out (whether it's whole or 'in bits' makes no
difference to the bin).

--
Johnny B Good