A couple of weeks ago they were fiddling with our local electrical supply, so they decided to run the area from a local diesel genny which they parked in a field and lashed up to a power pole. Aside from being a tad noissy 24/7 it was mostly ok (only conked out completely twice in about 10 days). However there were times that the power was somewhat "variable" - you could tell that the UPSs for my computer stuff were working hard to keep things normal. By the end of it one was starting to have difficulties holding up the supply at all, so I figured it was due a new set of batteries.
Anyway, got round to installing them today. The "weak" one had batteries last changed in March 2013, and you could tell it was not happy from how hot the case was. I had some difficulty prising them from the unit:
formatting link
formatting link
Half an hour later, they still felt a bit warm:
formatting link
Still on the bright side, they did not actually leak!
(A pair of probably a 20 year old 1kVA APC UPSs - not very kind to batteries, but they do seem to soldier on without complaint)
I have a feeling the same artist has been at work in one of my electric shavers, but not figured out how to get it apart without all the little bits going everywhere yet. Brian
Had they not already been 4.5 years old, then it might have been worth asking.
(the 2 year old set in the other UPS (which has the higher load on it) worked flawlessly and managed a run time of over 1h 10m on one occasion with two PCs supported).
I noticed that the genuine APC model I bought for another smaller UPS, was a bog standard CSB battery under the APC sticker. If you buy the CPC "unbranded" compatible packs, then they are also usually CSB batteries (and come stuck together and with the joining "fuse" to bridge them), at half the price.
I suspect that that is effectively the case here (and all done in the name of maximising the autonomy of the UPS on a brand new battery pack to gain "PR Bragging rights").
There is a range of float charging voltages for a 6 cell LA battery that goes from a low of 13.5 to a maximum of 13.8 volts. The lower limit is set by the need to avoid sulphation whilst the upper limit is to avoid corrosion.
Since I've been able to maintain a 12AH SLA in good condition for the past 5 or 6 years simply by using just a single annual charging session off a couple of 1.2Wp solar panels hanging out of my west facing office window for about a week's worth of summer time with maybe 3 to 6 hours of sunshine a day until I see the voltage hit or exceed the 13.8v mark, I'm rather inclined to think that the sulphation risk at the 13.5v mark is way overstated since the battery then spends the remaining 11 1/2 months of the year resting at a voltage of 12.85 to 12.75 volts. IOW, I suspect that dropping the float charging voltage down from 13.8 to 13.5 volts per
12v SLA will likely double the *useful* service life at perhaps just a 5 to 10% reduction in autonomy which seems a worthwhile trade off to gain another 3 years or so of extra service life.
Funnily enough, using a set of car batteries (SLI) as a substitute for the SLA types normally used by UPSes gains you an even shorter battery service life despite the seemingly less harsh treatment they enjoy compared to their use in a car where they're float charged at 14 to 14.2 volts and have to supply up to a couple of hundred amp discharge loadings
2 or more times a day as well as endure vibrations and various acceleration/decelerations throughout each journey.
Amazingly, SLI batteries seem to thrive on all this "rough treatment", lasting anywhere from 5 to 10 years whereas when "mollycoddled" by a UPS charging regime, they'll be lucky to survive a mere 6 months before becoming very thirsty and short of capacity, DAMHIK - IJK :-(
Using deep discharge leisure/marine or a pack of 2nd hand traction batteries would more likely prove a suitable alternative to the typical SLA type without suffering the shortened service life of an SLI battery but I'd recommend lowering the float charging voltage to 13.5 volts, whenever it's possible to do so.
The temporarily decommissioned (for want of yet another battery pack) SmartUPS2000 that I've bypassed to restore the power feed to my "Protected Sockets" will be having its charging voltage knocked back from
55.2[1] to 54.0 volts when I eventually get my hands on a cheap inverter genset and a justification to re-equip the UPS with a battery pack once more.
[1] Original voltage setting with a new battery pack but later testing revealed it was ultimately reaching a high of 55.5 volts once the battery pack had finally reached full charge. It turns out the "Constant Voltage/ current limited" charging circuit in a SmartUPS2000 isn't quite as "Constant Voltage" as I was expecting. Never mind, I'll make damn sure it can't go above the 54.0v mark when I set it up with the next battery pack.
They charge them fast so the UPS is ready for another outage ASAP, but why do they then continue to over charge them after that? I suspect just to generate a consumables revenue stream. You tend to find the capacity drops pretty linearly to nothing at 4 years old, whereas this is not the case with SLA batteries which are not overcharged.
It's on an APC UPS it will always without exception be overcharged as that design feature comes as standard regardless of price point. It's as if no one at APC knows anything about battery chemistry and that ramming almost unlimited amps up its chuff regardless of cell temperature is the best way to get the UPS ready for the next power supply issue. But it's not just the recovery to near full charge, it's years of trickle charging at too high a rate for the battery construction. It's only a few handfuls of millivolts but that is all it takes to reduce cell life from half a century to significantly less than half a decade.
That was the case nearly 30 years ago when they were first informed of this design issue, but the battery manufacturer probably told them that it was ok so the cycle repeats itself.
Well, that strikes me as the result of 'vested' interests by both parties (UPS manufacturers who can exploit a 100% or more markup and the battery manufacturers who will benefit regardless).
The choice of 13.8v over 13.5 has very little impact on recharging times. However, what I suspect (but have yet to find actual figures for) is that the lower limit reduces the autonomy of a brand new battery pack by something in the region of 5% to 10% for a chosen battery pack watt hours rating as the trade off for a doubling, perhaps even a tripling, of effective service life. Neither feature of which (slightly reduced autonomy and longer service life) suits the vested interests of either party (UPS and Battery manufacturers).
In theory, 12v SLAs can be charged safely from a constant voltage source of 13.8 volts no matter how low the dynamic impedance of the charging supply. The "Constant Voltage" charging circuits of all the SoHo grade UPSes I've had the displeasure to use fall far short of that ideal, behaving more like current limited constant voltage sources with relatively 'soft' end voltage regulation characteristics.
Even when the top of float charging current is a mere mA, the cells begin to suffer from corrosion effects given enough time (years of 24/7 charging versus a car battery's half an hour a day of such abuse at the higher 14.0 to 14.2 charging voltage over a period of a decade or so). It's the fact that UPS battery packs are subjected to such an unrelenting regime of constant float charge that makes their service life so critical of the chosen float charging voltage.
It's almost certain that the *only* downside to choosing the lower voltage option of 13.5v per half dozen cells, will be a modest reduction in autonomy with a brand new, as yet unabused, battery pack. Unfortunately, I haven't seen any figures on exactly how much of a reduction this might be. I think a reasonable guesstimate would be somewhere in the region of 5 to 10 percent which, even at the higher figure, would be a more than worthwhile trade off to gain a doubling or more of service life out of a not so cheap "consumable"
I personally believe that the whole concept of relegating such an expensive and critical part of a UPS as its battery pack to that of 'mere consumable' is a total and needless travesty. Sure, they do have a limited life, counted in discharge/recharge cycles but most UPS battery packs are rarely subjected to anywhere near that limiting level of stress.
I saw one contributor mention an option to adjust the charging voltage in software and my ancient SmartUPS2000 has a hardware adjustment pot on the circuit board so I suspect most, if not all, modern UPSes will have some provision to choose a lower than the "default" charging voltage as a "get out clause" to put the responsibility for ensuring a longer battery life firmly in the hands of the end user.
No doubt any mention of this feature, buried somewhere in the depths of a user guide (printed or an on line pdf), will mention caveats of reduced autonomy and underestimate the increased lifetime to be gained to help dissuade said end users from doing *anything* to bolster the principle of "user inertia" which AoL put to very effective use in its "Free Trial" customer recruitment campaign so very many years ago (some two decades back, iirc). It's worth bearing in mind that *all* manufacturer's "defaults" are chosen purely for *their* benefit alone, often against the best interests of their end users.
Witness all the kakamaimee defaults in a freshly installed copy of windows XP, patently designed to swiftly remove the shine off a recently purchased PC in order to persuade the end user to "upgrade" to a newer model a year or three sooner than is really necessary. Microsoft aren't the first corporation to exploit their customers' abject ignorance of their product's inner workings, they're simply the most shining example of such exploitative practices that other mass market orientated corporations can only dream of inflicting upon their hapless clientèle.
This business of choosing 13.8v as the optimum float charging voltage for a 12v SLA seems to be the result of both UPS and battery manufacturers closing ranks simply to maximise profits at their customers' expense. It's the standard trick of 'externalising' the costs of any "benefits" such as maximising autonomy of a brand new battery pack at the expense of shortened life, the cost of which ultimately impacts the environment.
Even when properly recycled, LA batteries still create an environmental impact from transport and handling of worn out batteries as well as that of the recycling process itself. APC offer a "Free Recycling" service but, quite frankly since this is only when you pay their grossly inflated battery pack prices, this is really more of a Piss Take than a service.
Absolutely. We spent well over a million quid in the early 1990's on APC products, we found out how bad they were after a few years in service and the swollen and often unremoveable battery without very brute force was the norm.
It's one of the pitfalls of buying commercial off the shelf kit based on a functional specification. Their response or more importantly the lack of it ensured they are forever blacklisted.
The same happened to some UK manufactured kit (mains fed 48v and 110v DC chargers with cells in racks) the battery manufacturer eventually acknowledged a problem, the charger was modified to have a lower voltage at most stages of the cycle together with temperature compensation and the batteries went from less than three years usable life to around twelve years at 80% capacity.
Ditto.
Piller (German made but a British owned company) don't cook the cells but despite that they are currently on a five year cell replacement cycle.
Their rotary UPS's are built to last 'forever' somewhere around 40 years is one of our oldest. Not practical for a domestic situation though :)
Also made harder because their kit is very well supported in terms of software compatibility while some of their competitors less so. I have used some other "off brand" UPS in the past, but you can run into problems getting things like NAS boxes to interface with them.
Early indications indicate it works ok, but offers slightly less run time (although its a physically smaller unit - which is one of the reasons I change it anyway)
HomeOwnersHub website 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.