On Mon, 16 Jun 2014 20:18:30 +0100, tony sayer
wrote:

The obvious solution to hours long autonomy by using a petrol genset
is only workable with the more expensive inverter type of genset. An
ordinary 2.8KVA generator will grossly overvolt if it sees even a
modest leading current load. A 4.7uF PF correction cap was enough to
send the nominal 230v north of 270 volts and there's nearly 9uF's
worth in the mains input circuit of the SmartUPS2000 alone, before we
even consider that due to the protected loads themselves - no wonder
the poor UPS kept cycling endlessly between mains and battery power
when it was supplied by the genset!

Yes I've seen just that happening but the generator voltage measured on a true
RMS Fluke meter was indicated at 230 volts;!..


Just out of interest, reproduced below an excerpt of an RAIB report issued today
were a railway level crossing gate system didn't work. Fortunately nothing was
going thru the crossing at the time from the road vehicle direction..

A bit of malarkey with the power it seems and the UPS system. All originally
caused be a RCD tripping which wasn't needed as there were multiple earth's, so
it sez;!....




80. Network Rail issues guidance about UPS systems. The guidance acknowledges
that the batteries of a UPS system ‘are the most critical component’ and are
‘perishable’ in that their performance will degrade over time. The guidance also
states that ‘manufacturers will provide batteries with a design life which will
often not be achieved. For example an 8-10 year battery will probably need
replacing after 7 years provided that maintenance and good temperature control
has been maintained’. The batteries of the UPS system in use at Butterswood
level crossing had not been replaced since the system was installed around 20
years before the incident.


81. The signalling department whose responsibilities included Butterswood level
crossing had no plans to replace the UPS system batteries during the life of the
system. In fact, Network Rails knowledge about the condition of the UPS system
at Butterswood, including the age of its batteries, was incomplete until after
the incident on 25 June 2013.


The data logger


82. The data logger fitted at Butterswood level crossing had its own internal
UPS system, separate from the level crossing’s main UPS system. Normally the
data logger is powered from the main incoming network power supply and uses its
own internal batteries to provide back-up power in the event of a main power
failure for a maximum of 6 hours. The data logger is tested annually by
signalling technicians. The test involves pulling the main network power lead
out of the unit, and checking that the indications are still showing that the
data logger is operating correctly. Network Rail requires that the batteries in
the UPS systems of its data loggers are renewed at no more than 10 yearly
intervals.


For level crossing UPS systems, Network Rail required them to operate for 20% of
their expected performance.

The UPS system at Butterswood level crossing was believed to be capable to
operate for up to 12 hours, therefore the permitted performance time was 2.4
hours.


83. Records indicate that the batteries were replaced in 2012. This means the
batteries only lasted around a year before they failed. Information from the
data logger manufacturer suggested that the batteries may only last 5 years,
depending on how many channels the data logger is monitoring. It was the belief
of the local signalling and electrification and plant teams, that it was often
the case that new batteries can spend significant periods of time ‘on the
shelf’ in store rooms, and could thus be several years old before they are used,
sometimes leading to premature failure.

That's a pretty accurate view regarding backup batteries. However, a
12AH 12v SLA I picked up in my local fleamarket for a fiver about two
years ago has proved to be a rather remarkable example of SLA
technology. It showed 13 point something volts on a borrowed digital
meter at the time of purchase but actually showed only about 12v when
I tested it at home (clearly the borrowed meter's battery was low).

I left it connected to one of those car battery maintainance solar
panels for nearly 6 months to fully charge it up last year, which it
did. It's basically been sat on my office windowsill for nearly a year
without being recharged because every time I checked its voltage, it
was showing 12.78 or so volts, even after using it for brief periods
testing 55W halogen capsule lamps for a minute or two at a time and to
use it as a test supply voltage for other kit every so often.

The total lack of care doesn't appear to have done it any harm and
when I checked it again today, it still showed a healthy 12.76 volts
and still fully lit up a 55W test lamp. I put it on charge from a pair
of those panels mid afternoon today (100mA) and the voltage peaked at
13.25v before settling back to 13.02v after sundown. A test just now
shows a voltage of 12.99v (1:30 am).

Mindful of the comments I'd seen recently about these solar panels
having high reverse leakage sufficient to undo the benefit, I checked
and measured a mere 2 micro-amps (using both a digital meter and an
analogue one on its 30 micro-amp range as a sanity check).

This was for two units in parallel in total darkness (the 33
microvolts open circuit reading probably being due to sodium street
lighting spillover) so whatever others experience might be in this
regard, this is certainly not true for the two units in my possession.

As far as charge retention performance goes, the difference between a
wet cell 12v motorcycle battery bought brand new last year and this
SLA is like night and day. The motorcycle battery (removed from the
T120v) only took a month or three to self discharge from full down to
a mere 6 or 7 volts!

I'm not sure if this is typical of a decent quality of SLA but if it
is, it would seem best _NOT_ to keep them on a float charge long
term. Just a refreshing charge once a year might prove to be the best
bet. Perhaps a float voltage of 13,2v with a brief monthly top up to
13.8v might turn out to be a better solution to achieving longevity in
standby service.

For anyone interested in a 12AH 12v SLA that can be left a whole year
between refreshing charges, the battery is a Chines made 'Maplin'
branded battery also marked as "AINO MICRO" underneath which it has
the following: "AM12-12 (12Volt 12AH)".

I can't spot any obvious date code except possibly one encoded on the
made in China sticker bar code - "5 026686 823035" for the benefit of
anyone who may know how to interpret it. The only other sequence of
digits visible is "0908051" embossed into the top cover plate.
--
J B Good