UPS lighting?


<snip> >>> What's strange about being fed up to the teeth (is that the right phrase)

I would say from the bottom of your stomach as I believe it's to do from the overconsumption of food.

<snip>

You 'have to'?

<snip> >> Sure, if it's an output overload trip. I'll check that out to confirm

Quite.

Yes, but what damage might it do in the process? If it's a thermal overload trip it may be all the electronics will be damaged before it heats up sufficiently to trip?

That's the idea (to not rely on just the fuses in the plugs relying the lighting circuits).

Not sure.

I'll look into it. Worst case was just a housing to take two std DIN MCBs.

Lets hope, although surge loads like inrush don't quite have the same instantaneous current peaks (that kill electronics) as real spikes caused by stuff failing.
Cheers, T i m
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

hrase)

Huh?

s irritating.

Hate. Typographical error.

m

trip.

I doubt it. The thing is quite happy starting big motors and the inrush current in loads of computer PSUs (which can easily trip bloody circuit breakers - bring back the fuse!).

gh to trip a health and softy breaker but not a proper manly fuse.

But bulbs blow so quickly that the short isn't big enough to knacker anything.

circuit, so anything really bad would likely blow that.

t

nd inrush things on big server power supplies.

The only thing I've broken with a blowing bulb was a 5 piece of shit thyristor output PIR detector from China. I always use relay ones now (mainly because they cause LEDs to flicker as the current isn't high enough for them to work properly).
-- Abdicate (v.), to give up all hope of ever having a flat stomach.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

I've already done this. I didn't use the lighting fuse in the fusebox at all. I have a 1500VA UPS that powers my computer/printers/monitors/stereo, which is plugged into a normal 13A socket in the ring main, and coming out of that are a few 6 way 13A strips for the computer etc. Into one of those sockets is a plug with a 5A fuse, and the cable runs into the fusebox and connects to the circuit that already fed the lights. Whatever it produces when on battery doesn't piss off any of the incandescent, halogen, CFL, or LED bulbs at all. And in fact since installing it, bulbs last longer because it steps down the voltage when the mains is too high (I've seen it as high as 256 volts). Probably filters other shit out too. It's an old American Power Conversion Smart UPS 1500 I got 2nd hand on Ebay, and fitted with new batteries. One odd thing it does is when on batteries, it outputs -110 and +110 volts on the live and neutral, but unless you're referencing it to earth, the bulbs aren't going to know that. I guess it's because it's American and that's how their electricity is.
--
Bills travel through the mail at twice the speed of cheques.

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On 08/02/2016 01:41, Mr Macaw wrote:

That's fine for lighting but may cause problems if you try to power the electrics of a central heating boiler with it. Some flame detection devices definitely require neutral to be referenced to earth. My Honda genny has a floating output, and I have to strap N to E to power the boiler.
--
Cheers,
Roger
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
wrote:

I wondered about that (not that we have any other appliances that would be powered. Gas fires, gas cooker (battery ignition) and a 'Multipoint' water heater with pilot. ;-)

Understood. Would you know if all genny's (I have a 650W 2/ and a 3KW Honda 4/) would be like this and the same question with the likes of the APC UPS's?
Cheers, T i m
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On 08/02/2016 11:09, T i m wrote:

I suspect that inverter-based generators usually have floating outputs - but I don't know about other sorts, or about UPS's.
--
Cheers,
Roger
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
wrote:

Ok cheers.
T i m
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

I guess if I connected my boiler, and if it had something that required what you say (possibly not as it's an old Baxi system boiler, non condensing, non combi, made before 2000 as it was here when I moved in), I'd need to strap the two together. But that would short out half the UPS output, so I'd have to give the boiler a fake earth, meaning touching it when on battery might give me a 110V shock if I touched a grounded washing machine. No, that wouldn't work either, as the pipes are most likely connected to its chassis and probably grounded elsewhere in the house. Perhaps a small isolating 1:1 transformer just for the boiler? Then I could strap one leg of its output to earth.
--
What's the difference between a church and a cinema?
In a church they say "Pray in the name of Jesus!"
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

I think the idea of bonding the output of a generators neutral to earth is because it's otherwise 'floating', rather than the neutral being referenced to ground as I believe it generally is (here in the UK).
Cheers, T i m
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

I assume the UPS we both have references the live and neutral outputs to ground. I tested them with my multimeter and got -110 , 0 , +110. But perhaps they were floating and the meter was just guessing? One of us would have to try connecting a small load between live and earth of it's output to see if it clamped it.
--
When shagging a goat you are best taking it to the edge of a cliff because they push back harder. -- Billy Connelly

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Well, I think it only needs to do one and the other follows as such.

Ok.

If you did that to ground then it could be floating or centre linked I guess?

Yes, I'll try and do that later.
Cheers, T i m
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Dunno how it works. Maybe there are two invertors? Maybe one invertor that makes a +ve and -ve output and has the earth connected in the middle?

It was exactly half on each side, so I guess it wasn't just floating.

Assuming it is 110-0-110, I guess you have a handy 110V supply. If it doesn't mind you drawing power off half the cycle only. If it's designed to be American then perhaps that would be ok, as thy have two voltages present in their houses.
--
A weekend wasted is not a wasted weekend.

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Interesting.

Ok.

Ok.

Yup, that was what I was hoping to do but with the UPS under the stairs etc.

Cool.

Yup, they can 'bang and buck I understand (online UPS's that is).

Same here, and I have the 1Kva and several of the 650 CS models.

I believe the 1000 and 1500 units can both take the same battery (just by removing some retaining plates from the battery bay on the 1000). I was waiting till I actually installed the system before buying the batteries, or, use some external ones.

Quite, and as I only intend powering the lights from my solution (I might also have the other unit on the computer gear but with several laptops and a mobile hotspot on my phone, I shouldn't need to).

Possibly. Thanks for the positive feedback.
So, assuming you use low energy lighting and the other equipment doesn't draw that much, how long do you get reserve lighting would you say?
Cheers, T i m
p.s. I can see why people wouldn't like this sort of thing installed for their elderly parents? Much safer than having them plunged into darkness and then faffing about with torches and candles etc?
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

-110 and +110 is actually safer and only a problem if some earths and neutrals have been connected or swapped somewhere. I did used to have a PIR powered off live and earth as there was no neutral where I put it (in place of a lightswitch), so that might not have worked on battery as it would only get half the voltage.

Not all low energy lighting yet. I'm perhaps stupidly wanting to wear out the old incandescents and CFLs. Mind you if I'd have changed everything to LED at once, I'd have the shitty LEDs, as they've improved gradually, and they would have cost a lot more to buy.

I take it you meant "can't" or "would".

Indeed. Although depending where you live there aren't powercuts like there used to be. 30 years ago where I used to live as a kid, we could have powercuts for several hours. I'd often produce an old car battery and a car headlamp as I've always liked fiddling with stuff and got stuff from the local scrapyard and skip.
--
I had amnesia once -- or twice.

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
wrote:

Presumably you meant can't and not can at the start of that para.
I think with elderly parents it would be better to have proper emergency lighting that comes on auto on a power failure if its dark where it is rather than to use a UPS, just because the UPS is a single point of failure and emergency lights aren't.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Tue, 9 Feb 2016 05:54:41 +1100, "Rod Speed"

I did indeed, thanks.

True, however, it might be quite easy to fit a UPS in the lighting circuits and therefore 'more likely' to happen than dealing with many different (and possibly considered 'unacceptable in their design' emergency light fittings?
Cheers, T i m
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Mon, 08 Feb 2016 19:06:28 +0000, T i m wrote:

Well, after reading through your back and forth with "Mr Mccaw" (interrupted towards the end by your exchange with the phantom killfiled "Rod.Speed") on this business of using a UPS to provide emergency lighting power over the existing lighting circuits, I thought I'd add some thoughts of my own to this discussion.
A few years back, I thought I'd upgrade my existing "Protected Supply" by adding a 2.8KVA 4 stroke single cylinder petrol generator (a cheap Aldi purchase for a mere 150 quid) to my existing SmartUPS2000 (an ancient APC 2KVA/1500W UPS) that I'd bought at the Blackpool Amateur Radio Rally over ten years ago, sans its battery box (hence its modest 35 quid price). One way and another (the details are hazy now) I wangled 48 volts worth of SLAs to run it from.
I parked it on the top shelf of battery shelving I'd originally fabricated out of dexion shelving to house two 12v 100AH banks of ex- PABX batteries in my basement cum radio shack (they originally had a 125AH rating but when I rescued the best 12 out of the 25 strong collection of completely disconnected and totally flat Tungstone manufactured cells I wasn't expecting them to still have more than a fraction of that capacity even if I could get them to recharge.
They were so completely flat, I'd had to revive them one at a time using a 21W light bulb in series with a 12v charger just to convert the distilled water back into enough dilute Sulphric acid before they'd start to accept charging current when assembled into 12 volt battery strings.
After a week or two involving the use 3 large 0.3 ohm 100W ceramic tubular resistors in series to boil a few bucket's worth of water by way of a test discharge load and the use of reverse polarity charges in- between from a 10A charger, I finally got both banks into a serviceable state which, on a very low 1A 200 hour discharge from the paralleled banks, indicated I'd managed to restore almost 80% of their original capacity in a neat demonstration of the difference between stationary Lead Acid batteries and SLI car batteries.
However, I only got another decade's worth of service out of them before the plates had shed enough material into the bottom of the cells, thereby increasing the leakage current causing state of charge imbalance and the ensuing woes of accelerated water loss in the remaining 'good cells'.
I was then reduced to picking out the 6 best cells to rearrange as one good bank, retiring the other 6 to stand in a corner of the basement until I could get a Round Tuit to weigh them in at one our local scrap metal merchants. It must have been around that point in time when I acquired the SmartUPS2000 because I then had the middle shelf clear to take a set of four 36AH car batteries supplemented by sets of 7AH 12v SLAs (including a bank of four 25AH 12v SLAs later on) none of which survive today.
In hindsight, the relatively short life of a typical SLA oft complained about by APC's customers is due to the choice of 13.8v float charge per 6 cell SLA. Experience suggests a better trade off between battery life and useful capacity would be the use of 13.5 volts for continuous float charging duty. This sacrifices some capacity but greatly increases battery life. The upper voltage limit of 13.8v per 12v SLA is chosen by APC (and others) to maximise autonomy performance at minimal initial cost.
Perhaps most surprisingly (and please take note) the battery type least suitable for such continuous float charging duty is the SLI car battery where you could well be forgiven in thinking the much 'easier life' (no vibration, no two or three hundred amp starter loads, no extremes of temperature and humidity or very high initial charging current from an alternator designed to charge at 14 or 14.2 volts to contend with) but you'd be dead wrong.
It actually took 3 "Experiments" in total to discover that a continuous float charge off a 13.8v charger is the worst thing you can subject a car battery to. The first 'experiment' took place two or three decades earlier when I used the classic CBer's trick of using a 13.8v 4A supply to float a 12v car battery to power a higher output power CB rig than the CB power supply could handle alone. At the time when the battery became 'goosed' just 6 months or so later, I just put it down to 'bad luck' with my battery and thought nothing further of it until decades later, when I was pondering the same fate that had befallen my *second* set of car batteries on the SmartUPS2000.
The plain fact is, a car battery is rarely subjected to more than 6 hours or so of continuous float charge at any one time. Most of the time it rarely sees more than an hour's worth per day at a time.
The optimum continuous float charge voltage depends on battery type and service. The gel type of SLA won't tolerate a gassing charge since the gel will trap the resulting bubbles, increasing the cell's effective series resistance. The fibreglass matt SLAs are more tolerant but even here it's best to avoid 'gassing' events since this involves an irretrievable loss of electrolyte, notably the water content. SLI batteries are more tolerant of gassing and short term excess float voltage charging but the problem with excess float charging is the accelerated corrosion this introduces.
In fact, when it comes to choosing the optimum float charging voltage with lead acid cells, you're rather caught between the Devil and the Deep Blue Sea on this matter. Too low and you risk sulphation, too high and you risk accelerated corrosion and all compounded by variables such as battery type and temperature.
The main reason why I attached my spare 7AH SLAs to the UPS was to keep them 'conditioned'. Now I know that I should have left them disconnected after just a day or two of charging on an annual cycle. They'd have lasted a lot longer as a 2nd hand 12AH SLA flea market 5 quid bargain I bought 5 or 6 years ago has proven by maintaining its open circuit voltage at 12.71 some 8 months after its last annual charge from a couple of small 1.2Wp solar panels hung out of the office window during summer for a week or so until the voltage hits or exceeds the 14 volt mark. Indeed I've just checked the battery voltage and lit up a 55W halogen headlamp capsule lamp to verify that the voltage isn't just all show and no go.
If you're planning on substituting the specified SLA battery pack with a bunch of car batteries please be aware that unless you can knock the float charging voltage down to 13.5 per 12v battery (27.0v for a pair, 54.0 volts for a set of four), this won't turn out quite as 'economic' as you may think.
A better option would be proper deep discharge marine or leisure batteries. Even better, if you can spare the room and can locate a cheap source, are traction batteries. If you're going to use a much bigger capacity battery than originally specced, you can afford to forgo the small extra effective capacity that excess float charging voltage brings to the UPS manufacturers' spec sheets.
Oh, and one final remark on the subject of using a large external battery pack in place of the original internal battery. Don't double or treble the capacity in the smaller 500VA/300W UPSes since they rely on the limited autonomy offered by their small internal battery to limit the overheating of their over-run transformers to a safe level. Replacing a pair of 4.5AH SLAs with an externally connected 25AH SLA battery pack is not a good idea, DAMHIK, I just know. :-(
Getting back to the mention of an emergency genset, I too then wondered how I could make use of the 'spare' KW of capacity and looked into using it to provide power to the lighting circuits. My idea was to wire up a changeover switch in each lighting circuit so I could divert the lighting wiring from the CU fuse links to the generator output which plugged into a main lever operated changeover switch which also diverted the feed to the UPS from mains to generator and back again. I only got as far as mounting the big switch and a couple of surface boxes on the CU backboard before abandoning the whole project due to a nasty interaction between the UPS's capacitive loading and the voltage regulator of the genset.
The problem with all such mains voltage alternators is their extreme sensitivity to modest capacitive loading which causes them to completely lose voltage regulation control and start overvolting (a 4.7microfarad Fluorescent lamp PF correction capacitor was all it needed to send the 230v AC climbing north of the 275 volt mark.
Since the SmartUPS2000 had some 9 microfarads effectively in parallel with the mains supply input which became isolated when running off the battery, I suffered the very puzzling effect that as soon as I switched over from mains to genset power, the UPS would kick out and run off the battery, then a second or three later switch back to its mains supply (in this case the genset) before 'rinsing and repeating' the whole process until I switched back to the mains.
It took quite a long time before I finally figured what the *true* cause of the problem was. It most certainly wasn't down to the "dirty" genset supply (harmonic distortion, poor voltage regulation on resistive and inductive loads and unstable frequency) but rather down to the one thing that all such alternators are prey to which is excess capacitive loading (FVSVO 'excessive' as it turns out) causing the alternator to go into an overvolt condition independent of any built in voltage regulator's best efforts to stabilise the output voltage.
The very latest switching inverter design of UPS (the type that does away with the heavy 50/60Hz mains transformer) may be usable with such cheap low tech gensets but this still leaves the issue of capacitive loading from the PSUs used by desktop PCs and other IT kit.
The only type of genset that's free of this unexpected voltage instability is the inverter type (eg the Honda e-3000 and similarly based competing designs). These typically use a permanent magnet alternator to generate a DC voltage to power the switching technology based 50/60 Hz inverter circuit and usually feature an 'eco-throttle' since the machine rpm no longer has any relationship to the inverter's output frequency, only output voltage and current which feeds the inverter allowing even lower fuel consumption to be achieved at no and low loads compared to the conventional alternator design.
Since I couldn't justify the expense of a better genset, I abandoned my emergency lighting power project and sold the genset on to my stepson who had a much better use for it than me (something to do with a portable mains supply for use at Go-Kart racing events iirc).
Apologies for the long post but I'm sure it'll save a few people contemplating emergency gensets and alternative battery packs for UPSes a lot of grief. I've done the work so you don't have to. :-)
--
Johnny B Good

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Mon, 15 Feb 2016 02:45:44 GMT, Johnny B Good

Thanks.

Ok.

Hey, I did similar. ;-)

Yup. Many of mine were 'DOA' but a few recovered sufficiently to be played with.

Quite.

Some people wash that out. Drain the electrolyte, wash out with distilled water and re-fill with electrolyte.

Scrap LA batteries went from a few quid each down to next to nothing and now seem to be up to something worth bothering with again. ;-)

Ok.

And I think I have noticed a higher float voltage on 'gel' LA than 'wet'.

I don't need that explaining mate. ;-)

Ok.

And is rarely float charged in any case. As soon as you have the (head)lights, rear screen heater, blower, electric fan, brake and indicators on whilst at the lights, there is a fair chance you may be taking some power from the battery.

Yup.

Many sealed batteries are able to recombine the water via special cell caps.

Some chargers can pulse / cycle when on float.

Ok.

;-) I use a 120W car fan heater to give my batteries a reasonable test load.

Understood.

Agreed.

Understood and I wouldn't do that. ;-)

I think that's quite a common solution. The switch has to be 'break before make' though.

Ah.

Ouch.

;-)

Interesting.

Ok.

Yeah, I've read about those (and would love one).

Possibly running tyre warmers?

No, thanks for that mate.
Reading the ever lengthening list of power stations that are being closed down and no matching list of those being built, I'm even more keen to do something automatic / central to at least keep the lights on for a while. This is now easier than it might have been years ago because of the use of low energy lights.
Stage one is to remote the output of the lighting circuits from the CU slightly with a couple of 13A mains sockets. Then convert the 1.5mm T&E lighting rings to a short flex with 13A plug via a suitable flex inlet / gland and junction box. That would allow me to have the lights running again 'as normal' but give me a (safe) 'flexibility point'. I can then easily plug a UPS into one or both circuits and do some tests.
Cheers, T i m
p.s. We have a low energy side light on a time switch that is very close to the APC BackUPS 650 that supports the wife's PC. It would be very easy to move that across to the UPS side, pull the power and see just how long it will run. I might need to open it up and put a switch in the warning buzzer first though. ;-)
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Mon, 15 Feb 2016 10:20:31 +0000, T i m wrote:

I could have tried the same remedy but with the cells being designed to cope with such shedding of plate material before it becomes an excess leakage or bridging problem, it meant that there wasn't likely to be a lot of active material left in the plates after such an exercise. I felt I'd well and truly gotten my 'money's worth' out of them by then. :-) Besides, there wasn't really the same need for a battery backed 12 volt shack supply, by that point in time.

That's handy to know because I have quite a collection of dead SLAs waiting to go to the scrappy. :-)

They seem to specify the same 13.8v per 6 cell SLA regardless of type as far as I've observed. Mind you, I have seen slight differences in the recommended 'optimum' float voltage per cell between the two different SLA types but these recommendations don't seem to be slavishly adhered to by the UPS manufacturers.

That suggests you've made the same (unfortunate) discovery. There's nothing like real world experience to prompt a closer look at "The Theory of How Things Work In Practice". :-)

What I was saying was I felt I was a little bit slow on the uptake in regard of this problem (subjecting SLI batteries to continuous float charging 24/7 month after month - not 'year after year' - they don't last long enough to make that phrasing appropriate to the reality).
In my case, with both sets of car batteries, they were 'shagged' after little more than 6 months of duty as UPS batteries. :-(

Well, add to that that unless we're talking about a very long journey along motorway and A class roads where the engine rpms rarely drop below the 2000 mark, most every day journeys involve rather more frequent and longer periods of idling rpm where the alternator can only supply a limited charging current which will often be cancelled out by extra lighting and wiper and fan heater motor loads (aggravated in very cold weather by electric screen heater loads).
Even when the alternator output brings the battery terminal voltage back up to its set 14 or 14.2 'float charge' voltage limit, the battery will still be absorbing charge current which mitigates the 'corrosion effect'.
Although the 13.8 volt float charging voltage of a UPS is significantly lower than a car alternator's 14 or 14.2 'float voltage' level, it's the unrelenting 13.8 volt charging voltage, day after day, month after month (forget 'year after year') that finally does an SLI battery in. Unfortunately, not an obvious limitation to most people looking for a cheap alternative to all those over-priced UPS 'battery packs'. :-(

I think most incorporate some form of catalytic mechanism to safely recombine the hydrogen and oxygen back into water but only to cope with moderate gassing rates before the cell's safety pressure relief caps start venting excess gas to the atmosphere.

That's a much better way to 'manage and maintain' battery condition, hence the description, "Intelligent Battery Charger". Unfortunately, the battery charging circuits used in the older types of APC SmartUPS model ranges mentioned here are rather 'dumb' being set to the upper end of the recommended float charging voltage range because it allows the manufacturers to specify the smallest capacity battery rating that will allow them to meet their claimed max loaded autonomy. They don't care that their customers don't discover the short two or three year life span of this particular "Consumable" until well past their generous "Two Year Warranty" period.

I wanted to avoid running any more 'deep cycling' tests which would 'burn up' any remaining charge/discharge cycling hours. The battery was 2nd hand to begin with and, despite the 12.8 volts measured on a borrowed DMM at the time of its purchase, it was actually only just over the 12 volt mark when I retested with my own DMMs (I'm guessing the borrowed DMM's battery was low, causing the over-read behaviour - I'm also guessing I must have overlooked the 'low batt' indication on the borrowed meter). Despite this unpromising start to its 'career' as a spare 12v SLA come occasional 12 volt test PSU, it did prove an effective jump starting battery a few years back.

The "Big Switch" consisted of a 2 pole 63A rated contactor with a 10A rated auxiliary switch which I'd modified to provide the necessary break before make changeover function. I never got round to wiring it (or the lighting circuits switches) up before the problem with the genset and the SmartUPS2000 put paid to the whole idea. Possibly a good thing because the 10A rated light switches may not have taken too kindly to a possible 480v rms between the N/O and the N/C contacts. They were, afaicr, just standard lighting switches for use on single phase 240v lighting circuits and may well not have had a large enough margin of safety to handle 480 volts rms. I think the contactor was rated for 600v rms so would have been ok.

Possibly but I think it may had more to do with allowing the use of power tools in the absence of a handy mains supply. Tyre warmers seems a bit too sophisticated for mere Go-Karting but I'm not that familiar with the sport.

Actually, low energy lights have been around for a surprisingly long time now (just over quarter of a century ago afaicr). I can remember the one and only time we had an actual outage on our mains supply. It must have been around 1990 when I was still interested enough to regularly attend my local Amateur Radio club meetings and shortly after acquiring my first UPS (an Emerson 30 450VA/300W 'quasi sinewave' output UPS which used a pair of 7A SLAs).
On the evening in question, just before I was about to set off for the club meeting, the mains dropped out. Since it soon enough became apparent that it wasn't only our house in blackout, I shutdown the PC the UPS was connected to and moved it and a lamp stand fitted with a CFL (but with its shade removed) onto the half landing outside the toilet where it could provide light to the hallway and the stairs and allow the children to sit on the toilet seat lid to do their homework by before I disappeared off to the club meeting.
When I returned about 3 hours later, the mains supply had been restored and the UPS was still running (but I think the lamp stand had been switched off when the supply returned about 2 hours or so after I'd left the house). Back in those early days of CFL technology the ballast was just a miniaturised (and weighty) inductor sans PFC capacitor which was allowed for sub 20W fluorescent lamps. Afaicr, it was a Philips SL18 lamp.
Later testing with a 14W load allowing this UPS to run off its battery for a whole 7 hours rather suggests it could have kept that 18 watt lamp going for another 3 hours. When I moved the UPS, I had to be careful not to shut it down because it couldn't be restarted to run off battery power in the absence of mains voltage, unlike later UPSes which could be triggered to run the backup inverter without first seeing mains voltage.
We've never experienced any further blackout events in the intervening 26 years since (hence my difficulty in justifying the expense of a 3KVA inverter type genset just a few years ago).

That sounds like a safer way to do the job (assuming the usual caveats about wiring safety). No worries about making sure any lighting feed changeover switches are rated for better than 480v rms between the n/o and the n/c contacts. :-)
Also, lighting circuits aren't normally arranged as a ring circuit so there's normally only the one cable per lighting circuit going into the CU. If you have enough slack and space on the CU backboard, you could wire them into an FCU and join your short length of plug ended flex onto the lighting cable end there.

I think my experience with a 14W load and the 7 hour run time I got out of the Emerson 30 should give you a good idea of what to expect. I've got a BackUPS500 protecting the 50W NAS box load for a mere 3 watts maintenance consumption and I'm guessing the Backups650 is merely an uprated version of this single 7AH SLA powered 'quasi Sinewave' inverter based UPS, most likely using a pair of 7AH SLAs instead of just the one.
Assuming I'm right about the use of two 7AH SLAs, I'd expect a run time with a 13W CFL load to be around the 6 1/2 to 7 1/2 hour mark with new batteries. The older type of synthesised Pure Sinewave UPSes aren't quite as efficient, only the later types that forego the need of a heavy mains transformer are likely to better the simpler quasi sinewave types and then, only by a small margin.
The older APC SmartUPS range are horribly expensive of electricity cost per Maximum VA's worth of protection. I have a SmartUPS700 now retired on account of its 20W maintenance consumption, none of which can be accounted for by the float charging current into the battery pack. The SmartUPS2000 is even higher at some 33W on average but its cost per VA's worth of protection is about half of the SmartUPS700's cost (I'd need 3 SmartUPS700s to provide as much protected output as a single SmartUPS2000 which would cost 60W in maintenance consumption versus the 33 or so watts of the larger 2KVA UPS.
Unless things have changed radically over the past few years, discovering these hidden costs was like trying to track down rocking horse shit. They were never revealed in the product spec sheets of the UPS manufacturers no matter how hard you looked.
The later, modern switching technology based designs that eliminate the expensive, bulky and heavy (and often over-stressed) mains transformer, may actually include such critical information, possibly along with battery pack life enhancing intelligent battery charge management.
Now as UPSes become ever more commoditised and 'consumer' orientated than they were 5 or 6 years ago, as more and more home users start realising the situation where the government's ill advised energy policies are leading us into, and take to buying UPSes in their droves, the manufacturers may actually start paying some attention to this new mass market demographic and take some measures in reducing these hidden costs and advertise the fact in the same way that BLMC did when using MPG figures to help sell the "Mini".
In the meantime, if you're checking out 2nd user UPSes to keep initial costs down, you'd be well advised to track down technical reviews as the only likely sources for such information to make sure you're not going to be adding tens of pounds worth of extra expenditure a year on your electricity bill. You might also, whilst you're at it, try and verify whether your UPS candidate has some sort of accessible preset to adjust the battery float voltage to a less damaging level, especially if you're planning on fitting a large and expensive battery pack (assuming the use of a UPS that permits such battery capacity expansion without detriment to its reliability - typically 2KVA or greater rated units with built in cooling fan or fans).
If I ever manage to get hold of another set of batteries at a knock down price to recommission the SmartUPS2000 as my 'Protected Supply', the first thing I'll be doing is to cut out the 'so called fan vent slot area' to fit a wire finger guard and then drill additional holes elsewhere in the plastic casing to allow said fan to properly earn its keep.
As it stands, the vent slots have so little effective area they must have been designed by a stylist rather than an engineer mindful of the needs of the typical axial fan which has resulted in a forced ventilation system that's little better than the indifferent efforts of an asthmatic mouse(tm JC).
--
Johnny B Good

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Mon, 15 Feb 2016 22:46:25 GMT, Johnny B Good
<snip good stuff for brevity>

Yes, I replaced a large Phillips CFL (the one in a glass jar) a while back that we had had for over 15 years (I try to write the date I install them on them somewhere). It was on a time switch, on from dusk to about midnight.
<snip> <snip>

Of course.

Hmm, you may be right but I thought mine were (but it was a long time since I wired this house). ;-)
<snip>

In the BackUPS 650's I have here there is only one 12V 7Ah SLA.

Even half that would buy you quite a bit of time and comfort.

And this (standby consumption) is something I will measure on the several UPS's I have here before doing anything.

<snip>

I hate that sort of thing and have been know to use my 'nibbler' and cut out the often restrictive and noisy 'perforated vent holes and replace them with finger guards.
Cheers, T i m
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

HomeOwnersHub.com is a website for homeowners and building and maintenance pros. It is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.