Hi
What are peoples recommendations on a UPS, living in near Ely now I have found that the power has dropped twice in the last couple of months due to 'substation faults' and would like to get something to protect my equipment and allow for graceful shut-down.
Ideally I would need power for my workstation, router, network switch, phone, 2 monitors, external drive bay and laptop.
Ideally 5-10 mins, minimum power to allow for a quick pause to see if the power is coming back followed by a manual power down.
All help, suggestions and pointers gratefully received.
Ben
Hi Ben,
I've been using APC UPSes for a few years - I get at least an hour out of mine which is plenty long enough to cover the DVR, ADSL and WIFI during the occasional dark interludes I've experienced out here in the fens.
https://www.google.co.uk/search?q=apc+ups+es400
I bought the cheaper ones with standard 3 pin UK power sockets because I couldn't face converting kettle leads to work the ADSL modem's 5V power adapter (wouldn't it be neat it they had 5V DC outlets). Also I didn't feel the need for smooth sine wave AC when on battery (only equipment I've seen which didn't like stepped approximation to sine waves was a Xilinx FPGA power supply).
Hope that helps, Tim.
On 11 November 2013 17:45, Ben Whyall ben@whyall-systems.co.uk wrote:
Hi
What are peoples recommendations on a UPS, living in near Ely now I have found that the power has dropped twice in the last couple of months due to 'substation faults' and would like to get something to protect my equipment and allow for graceful shut-down.
Ideally I would need power for my workstation, router, network switch, phone, 2 monitors, external drive bay and laptop.
Ideally 5-10 mins, minimum power to allow for a quick pause to see if the power is coming back followed by a manual power down.
All help, suggestions and pointers gratefully received.
Ben
main@lists.alug.org.uk http://www.alug.org.uk/ http://lists.alug.org.uk/mailman/listinfo/main Unsubscribe? See message headers or the web site above!
Thanks
Supplementary question, how can I work out how much power I need to keep stuff going as I ideally want to keep my desktop from dying spontaneously along with the drive bay.
Ben
On 11/11/2013 18:28, Tim Green wrote:
Hi Ben,
I've been using APC UPSes for a few years - I get at least an hour out of mine which is plenty long enough to cover the DVR, ADSL and WIFI during the occasional dark interludes I've experienced out here in the fens.
https://www.google.co.uk/search?q=apc+ups+es400
I bought the cheaper ones with standard 3 pin UK power sockets because I couldn't face converting kettle leads to work the ADSL modem's 5V power adapter (wouldn't it be neat it they had 5V DC outlets). Also I didn't feel the need for smooth sine wave AC when on battery (only equipment I've seen which didn't like stepped approximation to sine waves was a Xilinx FPGA power supply).
Hope that helps, Tim.
On 11 November 2013 17:45, Ben Whyall ben@whyall-systems.co.uk wrote:
Hi
What are peoples recommendations on a UPS, living in near Ely now I have found that the power has dropped twice in the last couple of months due to 'substation faults' and would like to get something to protect my equipment and allow for graceful shut-down.
Ideally I would need power for my workstation, router, network switch, phone, 2 monitors, external drive bay and laptop.
Ideally 5-10 mins, minimum power to allow for a quick pause to see if the power is coming back followed by a manual power down.
All help, suggestions and pointers gratefully received.
Ben
main@lists.alug.org.uk http://www.alug.org.uk/ http://lists.alug.org.uk/mailman/listinfo/main Unsubscribe? See message headers or the web site above!
On 11/11/13 18:40, Ben Whyall wrote:
Thanks
Supplementary question, how can I work out how much power I need to keep stuff going as I ideally want to keep my desktop from dying spontaneously along with the drive bay.
Ben
I think they explain that somewhere on the APC website.
I have an APC Smartups 700 and it keeps an old PC (with monitor off), external drive, wireless router and cordless phone going for about half an hour.
The lower the load, the longer the battery life. The higher the VA rating, the longer the battery life.
I brought a reconditioned APC UPS from ebay. It works well, but replace the batteries regularly - e.g. every 2 years - depending on the quality of the batteries.
Use APCUPSD (APC UPS Daemon) to monitor the UPS. This means that when the UPS detects power loss, you can configure what to do. The sensible thing to do is run either for a set time, or until say, 10% power remaining, then trigger a system shutdown. Configure your BIOS to power on on mains power returning, and Bob's your uncle. (You also need to configure the UPS and / or APCUPSD.)
This way, you can't ride out a long power cut, but you get controlled shutdown while you still have power, and smooth reboot when power returns. Data loss should be minimal - only any apps that are running with unsaved data on an unattended PC. Disk integrity should be preserved.
I suggest using an APC UPS because of the compatibility with APCUPSD.
Look at the APCUPSD website for UPSes to avoid. http://www.apcupsd.com/
Different UPSes use different cables to talk to the host PC. Look at the website for info. I'd suggest that a UPS that uses an RS232 to communicate to the host PC is too old - USB ones should work well.
Beware - you may need a custom USB or RS232 cable to put between the UPS and the PC. Check that the UPS comes with one if buying second hand, or be prepared to make one (see apcupsd website manual for info)
I got a 4 way extension cable, and a Kettle lead plug from Maplins, and replaced the standard plug on the extension lead, with the Kettle plug. That meant that I can run things with standard plugs via the USB. However, I am VERY VERY careful with it. I only plug in computer peripherals that are essential to the running of the PC and must be on while it is, and I make VERY VERY sure I don't overload it.
Hope that helps.
Steve
On 11 November 2013 20:30, steve-ALUG@hst.me.uk wrote:
The lower the load, the longer the battery life. The higher the VA rating, the longer the battery life.
At least as I understand it, VA doesn't give any indication of battery life, and it's quite common to find cheap UPS's differentiating themselves on VA rating (1500VA is better than 1200VA, right?).
750 VA means it can deliver 250V at 3A. If your hardware needs 4A the UPS will fail to provide it and presumably go overcurrent and shut down. Whether it can provide 250V at 3A for 5 seconds or 5 hours is immaterial. So, for the same load, a 750VA UPS and a 1500VA UPS will last just as long (assuming all else is equal), but if you have a 4A load the former will fail to work at all where the latter will.
"Decent" UPS manufacturers like APC tend to put bigger batteries in to a UPS with a higher VA (logic: bigger battery means more stored energy, and whether you want it for a low load over a long period of time or a high load over a short period of time is up to you, so make it capable of supporting the higher load).
Frustratingly even the likes of APC seem pretty vague about battery capacity (or they did the last time I looked at it) making it quite hard to compare. Generally they'll tell you how long the UPS will last a full or half load (on which basis a 1500VA UPS would appear half as good as a 750VA UPS on the same battery capacity).
I'm sure someone will correct me if I have all this wrong.
Mark
On 12/11/13 09:31, Mark Rogers wrote:
On 11 November 2013 20:30, steve-ALUG@hst.me.uk wrote:
The lower the load, the longer the battery life. The higher the VA rating, the longer the battery life.
At least as I understand it, VA doesn't give any indication of battery life, and it's quite common to find cheap UPS's differentiating themselves on VA rating (1500VA is better than 1200VA, right?).
750 VA means it can deliver 250V at 3A. If your hardware needs 4A the UPS will fail to provide it and presumably go overcurrent and shut down. Whether it can provide 250V at 3A for 5 seconds or 5 hours is immaterial. So, for the same load, a 750VA UPS and a 1500VA UPS will last just as long (assuming all else is equal), but if you have a 4A load the former will fail to work at all where the latter will.
Really????
IMO, provided that you're not trying to draw more current than the UPS can provide, a bigger VA rating means a longer battery life.
This is broadly supported by this page on the APC website http://www.apc.com/products/family/index.cfm?id=165&tab=models
where the bigger VA numbers correspond with a longer run time. Also, a lower load power rating results in a longer run time.
I'd suggest going to the website http://www.apc.com/site/apc/index.cfm
Clicking on the products tab, then click on the relevant Sector e.g. (probably) Home/Home office then click on "Uninterruptible Power Supply (UPS)" and see what's available and get an estimate of capabilities. Also remember the APCUPSD website for models to avoid.
HTH Steve
On 12/11/13 16:44, steve-ALUG@hst.me.uk wrote:
On 12/11/13 09:31, Mark Rogers wrote:
On 11 November 2013 20:30, steve-ALUG@hst.me.uk wrote:
The lower the load, the longer the battery life. The higher the VA rating, the longer the battery life.
At least as I understand it, VA doesn't give any indication of battery life, and it's quite common to find cheap UPS's differentiating themselves on VA rating (1500VA is bett
750 VA means it can deliver 250V at 3A. If your hardware needs 4A the UPS will fail to provide it and presumably go overcurrent and shut down. Whether it can provide 250V at 3A for 5 seconds or 5 hours is immaterial. So, for the same load, a 750VA UPS and a 1500VA UPS will last just as long (assuming all else is equal), but if you have a 4A load the former will fail to work at all where the latter will.
Really????
IMO, provided that you're not trying to draw more current than the UPS can provide, a bigger VA rating means a longer battery life.
AIRI that was the point: if the UPS is rated at 3A, it will not deliver 4A (for long). It's not the UPS per se, but the inverter in it that matters in this case.
Otherwise - I can't argue.
On 12/11/13 09:31, Mark Rogers wrote:
750 VA means it can deliver 250V at 3A
*if* everything connected has a power factor of 1 which is unlikely
As a guide.
A SMPSU with no power factor correction (anything cheap and under 75W this is quite likely) bet on a power factor of .5 to .6
A SMPSU with basic PFC is maybe .75
The very best Active PFC SMPSU's are about .9 (+85 spec PC power supplies and anything over 75W with the most recent Energy star rating is a good bet)
Just to confuse matters more even with PFC it is only designed to be optimal at 100% load as that is what is measured.
Short version, plugged into a selection of PC kit you should expect to get an average Power factor of around .75 so only 75% of that VA is turned into usable energy when converting to Watts or Amps.
As others have said VA rating has nothing to do with battery life..there is no time factor in a VA measurement so how can it. It's just coincidence that the larger models happen to have larger batteries so for the same load you'll get more runtime, but it's possible to buy 1500VA UPS's with different runtimes.
On 15/11/13 20:04, Wayne Stallwood wrote:
As others have said VA rating has nothing to do with battery life..there is no time factor in a VA measurement so how can it.
Well, unless I'm very much mistaken (which I may be), the VA number is an indication of the amount of power the UPS stores. An indication, not an exact figure. Also by extrapolation, it gives an indication of how much power the UPS will be able to provide. If it didn't, why would they use the VA figure?
I would expect that given two UPSes from the same manufacturer, with similar electronics, and similar battery technology, providing a load within their capabilities, one with twice the VA of the other, the one with twice the VA would last roughly twice as long as the other.
There is no indication of run-time, because the run-time depends on the load. The lower the load, the longer the run-time.
Checking the APC website's quoted figures. http://www.apc.com/products/family/index.cfm?id=165&tab=models
I looked at the APC Smart-ups 750VA LCD RM 2U 230V and the APC Smart-UPS 1500VA LCD RM 2U 230V. I discounted the cheapest UPS, as it would probably be a cost-cutting version. I picked those two because one is twice the capacity of the other, and they are both LCD RM 2U 230V (whatever that means)
Load in Watts, Runtime in Minutes
Load 750 1500 Runtime Ratio 50 139 335 2.4 100 65 190 2.9 200 23 92 4
I must admit I'm surprised that the ratio of the larger run-time to the smaller is not constant, but in all cases it's at least double. But it confirms that same manufacturer, similar technology, bigger VA => longer run-time.
It's just coincidence that the larger models happen to have larger batteries so for the same load you'll get more runtime
How is it a coincidence? Given the same battery technology, if you double the number of batteries, you're more or less double the storage capacity, and probably the VA. Or, if you want to increase the VA, you need to increase the battery capacity, either by increasing the number or size of the batteries, assuming the same battery technology. I can't see that as a coincidence at all.
, but it's possible to buy 1500VA UPS's with different runtimes.
Absolutely: different electronics, different types of battery, different manufacturers, a "cheapie" vs a "deluxe" version etc.
Not trying to sound argumentative, but what you're saying just doesn't make sense to me. Am I missing something?
Steve
On 16/11/13 00:23, steve-ALUG@hst.me.uk wrote:
I must admit I'm surprised that the ratio of the larger run-time to the smaller is not constant, but in all cases it's at least double. But it confirms that same manufacturer, similar technology, bigger VA => longer run-time.
It's just coincidence that the larger models happen to have larger batteries so for the same load you'll get more runtime
How is it a coincidence? Given the same battery technology, if you double the number of batteries, you're more or less double the storage capacity, and probably the VA.
Because The VA rating has nothing directly to do with the number of batteries. It is a rating of the AC inverter and associated electronics. A double rated UPS in terms of VA may or may not have twice the battery capacity. Even within a particular VA rating you can get wildly different battery capacities,
Take two models from APC's site, Compare SUA750ICH and say SMX750I (both 750VA)
If you were comparing over different makes you CANNOT rely on there being anything like a 1:1 relationship on runtime to VA. It just happens that APC like to spec *most* of their UPS's to 5 minutes at full rated load so tend to double the batteries for a double in VA, and then you add batteries via the expansion kits if you want more. But a 750VA UPS with 2 battery expansions is still a 750VA UPS.
Can increasing (lead-acid) accumulator capacity overload the charger circuitry?
On 17/11/13 17:16, Anthony Anson wrote:
Can increasing (lead-acid) accumulator capacity overload the charger circuitry?
In theory it might but you'd probably get away with it to a reasonable degree. Pb batteries are supposed to be charged constant voltage so in theory doing things like running packs in parallel would lower the series resistance and thus make the charging circuit work harder. Though I think on heavily discharged packs the charger starts off CC and then moves to a CV model once some base charge is in.
In practice a lot of the APC UPS' are designed to deal with additional battery modules so you might get away with it.
Couple of things though
Charging parallel packs isn't good..running/charging parallel packs of different age/type/capacity is very bad. Running a bigger battery or packs in series to make up the correct voltage is better than two packs of the right voltage in parallel, generally UPS's tend to run batteries in series to make up a higher voltage if they have multiples.
You might confuse the life out of any capacity monitoring circuit, though I'd expect anything good to be using a coulomb counter which should cope.
Charging leisure type batteries and Pb "Gel Cells" as used in UPS kit is different. I think putting a leisure battery on a gel charger is better than the other way round but I'd check on that before you explode yourself due to excessive venting from the pack :-)
On 19/11/13 21:29, Wayne Stallwood wrote:
/snip/
Charging leisure type batteries and Pb "Gel Cells" as used in UPS kit is different. I think putting a leisure battery on a gel charger is better than the other way round but I'd check on that before you explode yourself due to excessive venting from the pack :-)
Tanks for that - I was considering the possibility of using two much heavier batteries instead of those in the pack.
I have one which needs new batteries, and that would/would have removed the need to buy two small ones as well.
While the bigger batteries would be OK to use while charged, discharging them to some extent in a power out(r)age might have nasty results on restoration of the service, then?
On 16 November 2013 00:23, steve-ALUG@hst.me.uk wrote:
Not trying to sound argumentative, but what you're saying just doesn't make sense to me. Am I missing something?
Yes, you're missing something, but that seems to be "by design". Battery/UPS technology seems devoid of meaningful marketing information, and I suspect that "most people" believe that a 1500VA UPS will last twice as long as a 750VA UPS. Quite why the likes of APC don't do more to dispel this myth defeats me - their low-end 750VA UPS is otherwise easily beaten by a cheap 1500VA unbranded UPS.
At least with other rechargeable batteries, such as phone or NiCd/NiMH AA/AAA/etc, you do get some idea of capacity in the mAh values - mAh again isn't sufficient alone to tell you how much capacity a battery has, but combined with knowing the voltage of the battery it is (and at least that is also specified on the battery). The cordless phone next to me has 800mAh 1.2V NiMH batteries in it - ie 0.8A (800mA) at 1.2V for 1hr, or 0.4A at 1.2V for 2hrs, etc.
I did once ask an APC salesperson about this and the response was something detailed and helpful like "I dunno" so I gave up.
On 15 November 2013 20:04, Wayne Stallwood ALUGlist@digimatic.co.uk wrote:
On 12/11/13 09:31, Mark Rogers wrote:
750 VA means it can deliver 250V at 3A
*if* everything connected has a power factor of 1 which is unlikely
Indeed, that was lazy (and indeed it's unlikely it would be supplying 250V either...). It was just to make the point that VA is basically "volts times amps" and neither of those include a time component.
As others have said VA rating has nothing to do with battery life..
The "others" in this case were just me and the lack of anyone confirming it was starting to make me doubt myself, so thanks for jumping in!
On 11/11/13 17:45, Ben Whyall wrote:
Ideally I would need power for my workstation, router, network switch, phone, 2 monitors, external drive bay and laptop
Why do you need to power your laptop from the UPS ?
As others have said go with APC and use Apcupsd to manage it and tell your workstation to shutdown gracefully when there is only enough power left to allow it to do so.
As for working out what you need, APC have calculators though they tend to be a bit on the pessimistic side ( I wonder if the marketing team had a hand it that to push you up the range )
I'd say finger in the air that you could get away with a 750VA but I'd go for 1500.
Rule of thumb calculations below
Screens 50W each (could be an overestimate, I don't have any here to look at)
say 30W for external HDD Router and switch 10W each
To get the W rating of a UPS from the VA we are either going to have to go to great pain to calculate the power factor of all your kit combined or I guess and say it is likely to be about 0.7 overall.
W = VA × PF would then give you about 500ish Watts of capacity minus the 150 above that doesn't sound like lots left over for your workstation and laptop? if you get the 750.
-
Anthony Anson -
Ben Whyall -
Mark Rogers -
steve-ALUG＠hst.me.uk -
Tim Green -
Wayne Stallwood