UPS that is UN-interrupt-able;!..

[D.M.Chapman]

Not always. We have plenty of UPS around the place that will supply power to keep core network kit running for many hours. Something in excess of 10 hours on most.

These are APC UPS - and are pretty good. Mind you, not the small home/small office ones, these have shelves of batteries.

Also, don't expect too much from the batteries - you are talking of a small number of years tbh.

We did have a minor issue with one of our main UPS (many KVa - not APC ones). It was installed with the wrong charging controller. First night this happened...

hadn't powered anything at that point. Data centre smelt of sulphur for a while after that one - and some poor sod had the job of chiselling them out of the racks (the photos don't show the worst - the installer didn't like us taking photos of their expensive self destructing UPS ;-))

Darren

[Espen Koht]

I wonder why. We cycle the batteries on our solar panel system daily all through the summer, and for the rest of the year they stay on maintenance, yet after 15 years of operation we've only on our second set. Do infrequent high loads really make that much of a difference?

[Espen Koht]

It's a bit like getting your first mobile phone and realising that if used properly, rather than subjecting you to the tyranny of making your reachable at all times if liberates you by giving you more control over when and where you can be reached. The same can be applied to email, and once you have a capable email device in your pocket that might as well also be your phone.

[John Williamson]

They can. I'd guess that your solar array takes a few hours to charge from flat, and that the load you put on the batteries would take a few hours to flatten the batteries completely. I'd also guess that your batteries are rated so that normal use only takes the charge down to fifty percent or thereabouts. All things designed to get the best life out of a set of batteries.

Most standby applications run the batteries at a rate that will flatten them in less than an hour, and charge them in the same time frame. Add that to what is normally a not very well controlled float charge regime, and that will kill 'em quite quickly by comparison. Another thing that kills UPS batteries is that they can't be serviced, as wet cell types can. Solar installations I've seen have wet cell batteries that can be topped up with distilled water, while I've only ever seen very old standby installations with that type of cell, and can they last for ages, with care. Ask BT......

[Dave Liquorice]

I have mate who lives in the out skirts of Bristol his power is terrible, voltage flucations and cuts several times a year. Basically the infrastructure hasn't kept up with development and increased demand.

Out here you can probably measure the amount of undergound supply bewteen us and the 125kV grid in feet on the 25 mile journey to us. The single 33kV feed (and 11kV backup) comes over Hartside at 2,000' and is very exposed. Our power is pretty stable and reliable, one cut of hours duration every couple of years and the occasional trip of the auto-reclosure in late spring (tree growth).

[Dave Liquorice]

The electronics side of my little (700VA) APC just works, must be 10 years old or so now. On it's third set of batteries...

I think the problem with the batteries is that they are in the case, thus warm (40C ish) and the charging is not clever enough to switch from rapid to float. What I have found is that switching off the automatic battery test extends thier life noticeably, to the plus side of 4 years rather than 3.

[Tim Watts]

Dave Liquorice ( [e-mail]) wibbled on Saturday 29 January 2011 12:11:

I'd wondered about that battery test.

Given you can buy a box of electronics that performs a capacity test on lead-acids (it does it by some clever means involving AC currents to infer things, rather than doing a drop-test) I wonder why such a device couldn't be incorporated into the unit?

Test would involve: switch out the cells for 20 seconds, switch in the tester module, take reading, switch cells back into circuit. No damage to the cells.

[Roland Perry]

In message , at 09:09:25 on Fri, 28 Jan 2011, Jeff Strickland remarked:

If you look at the data from UPS manufacturers (and I'm talking about professional UPS) then the shortest time they expect you to drain the batteries is about ten minutes. And the fact they give data up to an hour or more means they expect some of their buyers to be using them to bridge gaps like that.

[Roland Perry]

In message , at

15:21:20 on Fri, 28 Jan 2011, Espen H. Koht remarked:

Mainstream PDA - yes I agree. Mainstream desktop computing - no chance.

This appears to be some sort of anti-Windows religious rant. Like it or hate it, Word and Excel are the world standards for a huge amount of desktop computing. Agreed, they don't have to be running on Windows, but most people do. I've described some of the advances that I think need to be made for them to be realistically the choice of tablet users.

And the need for "universal connectivity" to make applications usable is very much an urban myopia. The cost (and reliability) of communications for tablets is still way outside the parameters where I could recommend that people buy into what's a modern hostage to fortune. Unless it works for you, when it's fine.

[I failed to connect to the wifi-enabled train I was on yesterday. Coupled with the impervious-to-GSM train design, one begins to give up on the concept. And that's without trying to roam overseas)]

[Roland Perry]

In message , at

01:08:34 on Sat, 29 Jan 2011, Espen Koht remarked:

The only "capable email device" I've ever been able to fit in my pocket is a 2002-vintage ultra-portable laptop. Phones might be capable of receiving and dealing with trivially small amounts of trivially non-complex email (which can still be useful in limited circumstances) but in effect they are just 2-way pagers.

[Espen Koht]

I expect there to be some convergence here, but I'm talking about 'tablet computing' here, not desktop computing.

That's a facile argument. I will readily admit to not being particularly fond of Windows, but this is a considered opinion based on my observation of the trajectory of 'tablet' computing going back the early

90s (when someone tried to recruit me for a startup basing its business around 'Windows for Pen Computing' tablets). Let's just say that the importance of the current developments is not just that it is not Windows based, but that the front-end experience isn't based on a desktop OS of any flavour (including MacOS X) either. Looking back the desktop/start menu (or apple menu) metaphor crippled most tablet attempts from the outset.

The lack of Word for the iPad doesn't appear to have become an insurmountable obstacle.

Most iOS apps don't require 'universal connectivity'; that's the google model. A thin client model for tablet PCs is probably some way off, as you've identified.

[Espen Koht]

The first set was wet cell, but we're now using dry AGM batteries, and the life expectancy in our application is supposed to similar. We'll see.

[Skipweasel]

Not having Office 07 is a positive benefit, having used it.

[Jules Richardson]

A proper 'open' well? We're out in the sticks here, so everyone has a private well on their land - but they're just drilled pipe of around 4" max diameter, rather than a big ol' hole and a bucket (which, prior to moving over here, was what I always thought of whenever someone mentioned wells)

The disused well-shaft I've got is a box around 4' on each side, and doesn't go down too far - maybe 12' or so. At the bottom is a 4" well cap which has been filled with concrete; presumably there was once a pump in there too (so still a "surface" pump, just below the frost-line). It'd be a good spot for housing big flywheels, though :-)

Our current well out back is 85' deep, 2" pipe, with underground lines running to a jet-pump situated in the house basement. That was put in in

1986 (long before we moved here), although the pump is dated 1977 (and was possibly the one that may have once been in the disused shaft). Info that came with the house says there was a 53' well here before the current one, which may refer to the capped one in the well-shed. There's also a capped well in the house basement itself, which probably dates from when the house was built in 1949 (I've heard they generally have about a 20 year lifespan before silting and corrosion kills them, so our current one's a little past its sell-by date)

cheers

Jules

[Roland Perry]

In message , at

17:20:18 on Sat, 29 Jan 2011, Espen Koht remarked:

Tablet computing (as we see it today) has its uses. But it's very far from being good enough to displace a laptop which is itself in effect a portable desktop machine.

It would be for people whose job involved reading and editing Word documents.

It's not just thin clients - my new phone has Google maps linked to GPS, but there's no mapping data *in* the phone (although it seems to be building a cache). Contrast with a conventional SatNav which has the maps built in. Or maybe Google Maps is a thin client (the definition seems a bit blurry to me). I expect many of the other apps in a tablet really need the connectivity to be able to do anything useful at all.

[Skipweasel]

Yup - dated back a century or more.

[Tim Watts]

Skipweasel ( [e-mail]) wibbled on Saturday 29 January 2011

20:36:

I read some bits on 'tinterweb about how they used to hand dig wells. Very interesting.

One method was to take a 1/2 barrel (big barrel), dig a hole a few feet deep and the diameter of the barrel. Then they would inset the half-barrel, and build a few courses of circular brick wall round inside the hole, supported on the barrel.

I guess when they had enough courses to hold together in a stable fashion, a bloke would go into the hole and dig. As he dug, the barrel would slip downwards into the newly formed hole and the brick "tube" would follow.

Bloke up top continues to add courses and bloke down bottom keeps digging - using a bucket on a rope (and presumably more blokes) to haul out the earth.

Relatively safe, I would guess, provided the brick tube didn't collapse, as the barrel (and bricks) would fill the excavation as soon as it was extended leaving no unsupported earth walls to fall in.

That's the theory - not sure I'd fancy it - but I guess they made plenty that way.

In more modern times, I have seen hand dug wells lined with concrete sewer pipe sections which would probably be a pretty bomb proof method - wasn't deep though, the one I saw in Latvia - probably about 50 feet if that.

[The Natural Philosopher]

Yes. we have a really hard time on usenet without all the features of Word to make our outpourings formatted, don't we?

[mick]

Virgin Media used to use 3 big diesel-driven flywheel ups sets (1 running constantly, 1 on standby and 1 under maintenance - so only 1 at once). They kept a lot of the gear up for long enough to get the emergency generators synchronised & on line. I don't know if they are still in use though. They have the advantage that you can hit them with big instantaneous loads without the frequency taking a huge nosedive. The engine is on tickover once the flywheel is up to speed, so pretty efficient for big systems. The big flywheel is scary though!

[mick]

You are right in a lot of respects!

The main rule for *any* standby power system is TEST IT REGULARLY - ON REALISTIC LOAD.

You wouldn't believe the number of places that stick a ups or standby genny in and then don't do proper tests. They will probably do the simple stuff (e.g. read the battery status off the ups panel), but nothing that involves going in on a pre-arranged sunday with a load bank or three and doing a full run by turning off the mains incoming switch/breaker. Then they moan that something doesn't work when they have a real mains outage. It only needs that sort of test quarterly, six monthly or something like that and it's cost effective to hire the load bank(s) for a day.

FUs trimmed to uk.d-i-y only