Is it worth buying a UPS??

I work with UPS as my job. There definitely better than surge protection and broadly speaking there are two types.

1) Off-line - With these you are mostly connected to the mains and these only provide power when mains fails. ie Still see brown outs and spikes etc (although reduced if UPS has filtering sometimes called Line interactive)

2) On-line - These are expensive but the best. The load always runs off an inverter which is supplied by rectified mains or the battery so there is always a perfect supply no matter what happens. Most data centres for example would run multiple on line UPS in parallel for redundancy.

Most of the APC and other low cost units will be the off line type.
 
For me it's more about convenience than data loss
Difference in convenience is quite major if electric network in area is sensitive to for example bad weather: Wind can keep some tree branch swinging to wires causing repeating short black outs. (that's what "ripened" me to UPS-less PC use)
And that's where surge protectors fail to do anything helpful.


1) Off-line - With these you are mostly connected to the mains and these only provide power when mains fails. ie Still see brown outs and spikes etc (although reduced if UPS has filtering sometimes called Line interactive)
Surge filtering/buck&boost voltage regulation can be equally built to both but especially voltage regulation (as price rising feature) is simply left out of off-line UPSes aiming to lowest price.

Here's quite decent information about differences between architectures and another price difference causing "feature":
http://www.pcguide.com/ref/power/ext/ups/types.htm
http://www.pcguide.com/ref/power/ext/ups/funcOutput-c.html

So there are good reasons behind price differences between various UPS-series.
 
The only reason I am thinking about getting one is simply because I don't want my hardware getting damaged because of power outages.

If a surge protector would give me the protection I need then I wouldn't bother buying a UPS.
Let's start with the many myths. Power outages do not cause damage for a long list of reasons. For example, if power outage causes damage, then so does power off. As the telco installer noted, those telco switching machines do not have a UPS because power off is not destructive.

Because it is called a surge protector, that means it provides surge protection? When did you (and others) jump to that conclusion based on what facts and numbers? 'Protectors' and what provides 'protection' are two completely different items. Most educated only in word association or from a retail salesman never learned the difference. What does a protector do?

Either surge energy is dissipated harmlessly in earth. Or that surge is inside the building hunting for earth ground destructively via appliances. No protector stops or absorbs surges as myths so often promote. Will that 2 cm part inside stop what 3 kilometers of sky could not? That is what some are saying. Will its few hundred joules absorb a surge of hundreds of thousands of joules? Of course not. But again, that is what many claim - by ignoring numbers.

So again, what does that plug-in protectors (or UPS) do? View its manufacturer specs. Where does it list each type of surge and protection from that surge - in numbers. It does not. It does not claim to protect what an overwhelming majority recommend it for.

So what does BT do? With every thunderstorm, that CO (switching computer) is confronted by about 100 surges. How often is your town without phone service for four days while they replace that computer? Never. Because they do not use any plug-in protectors. And they use the only protectors that have performed surge protection for the past 100 years. Yes, we are now about to discuss how all successful surge protection has been done long before computers even existed.

From the NIST (a US government research agency) is:
> You cannot really suppress a surge altogether, nor "arrest" it.
> What these protective devices do is neither suppress nor arrest
> a surge, but simply > divert it to ground, where it can do no
> harm.

Why does the NIST say something completely different from what the overwhelming majority assumed? Again, where does the energy go? Did they even ask that question? What does BT do to not have damage? BT does not spend massive sums on those obscenely overpriced protectors. BT installs protectors that are as close to earth ground as possible and up to 50 meters separated from electronics. Both distances mean better protection. An effective protector makes a 'less than 3 meter' connection to earth. Obviously nothing plug-in can make that low impedance connection. A connection that must be that short, no sharp wire bends, no splices, ground wire separated from all other wires, not inside metallic conduit, etc. All necessary so that a protector connects that energy to earth.

So what does a protector do? It is a connecting device - not a protection device. A protector must connect every incoming wire in every cable 'less than 3 meters' to single point earth ground. How does a UPS do that? It does not - and does not claim to provide protection in its numeric specs.

Only the more responsible companies provide 'whole house' protectors such as Keison, Siemens, General Electric, etc. Earthed protectors for phones are:
http://www.citelprotection.com/english/citel_data_sheets/data_line_protection/B280_B380_B480.PDF
http://www.keison.co.uk/furse/furse11.htm

How to make a protector better? Upgrade (ie shorten) the quality of and connection to earth ground. A protector is only as effective as its earth ground. The NIST makes that point bluntly:
> A very important point to keep in mind is that your surge protector will
> work by diverting the surges to ground. The best surge protection in
> the world can be useless if grounding is not done properly.

UPS has one function. To provide power during blackouts - to protect from unsaved data loss. View that UPS spec. How many joules? It contain near zero joules. But near zero is enough for a sales brochure to convince the naive that it contains "Surge Protection" (in big letters). And so an overwhelming majority say a UPS provides surge protection. Most ignore the numbers that say 'near zero' (ineffective) protection.

Where does that energy get dissipated? Either surges are harmlessly absorbed in earth (as BT and all other phone companies do everywhere in the world to have no damage). Or that energy is hunting for earth destructively via your appliances. A plug-in protector can even give a surge more paths to find earth destructive via adjacent appliances. Yes, can even contribute to damage of adjacent appliances.

The informed consumer installs only one 'whole house' protector so that everything (even the furnace and dishwasher) is protected at about £1 per protected appliance. When surge damage cannot happen, the protector with a dedicated wire to earth ground is the 'always required' solution.

Appreciate how much longer it took to explain technical reality. Again, why does a overwhelming majority make recommendations without knowing this?
 
Last edited:
I have had a belkin 1200va ups for a good 2yrs, the main reason was we had dodgey eletric and our power would trip off at least once a day. A year later we got it fixed(a new tripswitch). But during that year my ups never let me down once, and if it wasnt for the ups Im guessing my pc would be dead with the power going off dayly. now Im still using the ups, cos its monitoring to power before it gets anywhere near my pc.

A UPS won't protect against a lightning strike, there's too much power. I would definately reccomend one though, power outages can cause data corruption because of write caching and data loss if what your working on hasn't been saved for some reason.

http://www.overclockers.co.uk/showproduct.php?prodid=UP-017-AP&tool=3

Something like that should be enough for an average pc.

Maybe not a direct hit,, but I spoke to belkin about my 1200va ups, and they said my ups could withstand a surge upto 39,000,000 watts and should be fine in a bad storm. But just to be on the safe side I have my ups pluged in to a surge protector aswell:D Plus my ups has a £100,000 warranty if it ever fails to protect me
 
Last edited:
Im guessing my pc would be dead with the power going off dayly. now Im still using the ups, cos its monitoring to power before it gets anywhere near my pc.
Even 30 years ago, one international standard had this expression in the low voltage (dodgey eletric) region. "No Damage Region". Low and lost voltage will only damage electronics when that electronics is also damaged by power off.

Belkin did not say it can withstand 39,000,000 watts. In fact, it makes no claims for surge protection as demonstrated by the numbers you did not post. Even the £100,000 warranty is bogus. You are posting urban myths. Belkin does not claim effective protection.

The long post only challenged some of the popular myths. 'Surge protector' sounds like 'surge protection'. Therefore it must be true? A perfect example of why everything you posted is mythical is the warranty. Learn what others discovered - what is routine - when they tried to have that warranty honored. tamaradensh in aus.computers on 12 Jul 2008 at:
> I had it plugged into a Belkin Surge Protector. They have a connected
> equipment guarentee. I put in the claim and was told that it only
> covered equipment for power surges as the products wasn't designed
> to protect from other power malfunctions.

Newsman on 10 Sept 2002 in the newsgroup alt.video.ptv.tivo entitled "SONY TiVo SVR-2000":
> Yet my Tivo's modem still failed. And the '$20,000 connected devices warranty'
> did not help me. I jumped through many hoops, including finding the original
> recept for the surge protector (just under a year old) and I sent my surge protector
> to Belkin (paid for shipping), and was denied my warranty. They gave me a ton of
> crap, ...
> Eventually it boiled down to a line in the warranty that said "Belkin at it's sole
> discretion can reject any claim for any reason".

That Belkin is protection only when Mighty Mouse saves the day.
 
> Eventually it boiled down to a line in the warranty that said "Belkin at it's sole
> discretion can reject any claim for any reason".

To be fair that would probably be illegal under the Unfair Contract Terms Act 1977 in the UK.

Even 30 years ago, one international standard had this expression in the low voltage (dodgey eletric) region. "No Damage Region". Low and lost voltage will only damage electronics when that electronics is also damaged by power off.

Aren't motors often damaged when undervolted?
 
Last edited:
If surge protectors and ups are a load of rubbish, why do the exist?? cos were stupid enough to buy them??

I had my pc took out during a thunderstom once about 5-6yrs ago now and had no "so called" protection that you guys are saying is rubbish. A almighty crack of thunder, the tripswitch when off, and when we turmed the trip back on, my pc was dead.

I might scan the warremty of my ups sometime, and ill post it.
 
Last edited:
Also questioned belkin a few yrs ago and ive buggered my warrenty by plugging the ups into a surge protector.

http://forums.overclockers.co.uk/showthread.php?t=17755292

Heres 2 posts from a tread I made about a yr ago,,, you maybe intrested........

--------------------------------------------------------------------------------------------
Been talking to this guy, and he says my ups can handle 39,000,000 watts, my spec is "420 Joules, Maximum Surge / Spike - 6,500 Amps / 6000 Volts" so if you x the amp by the volts, you get how meny watts it can handle.


Also he says "now a lightning strike (being an instantaneous surge) can be way more than the usual 5,000W maximum that will normally travel through your socket, but no way it will get up to 39 millionnot in a residential home. Too much resistance"

Intresting hey guys

--------------------------------------------------------------------------------


------------------------------------------------------------------------------
Belkin stuff is great and the reason why we rolled them out around our company.
All of the server room are covered by expensive rack-mount APC stuff that is all centrally controlled etc.
However each of our workstations have a Belkin UPS attached.
I myself use a 1200va Belkin at home and over the past few weeks with all the thunderstorms arond at no stage have I worried.
On more than one occasion I heard the unit kick in on either a power surge or power drop.
The one time we actually lost power my machine and ADSL modem continued to function without problem for the 15 minutes that power was out for - and with battery life to spare.

I refuse to buy APC equipment for home as you can get equally and better specification stuff from the likes of Belkin for a lot less money.
And the fact we've got so many (75) at work plus my slightly better one at home gives me the confidence to recommend them.

----------------------------------------------------------------------------------------
 
Last edited:
Aren't motors often damaged when undervolted?
The chart defines electronics. Motorize appliances cannot tolerate low voltages. AC electric voltage must not dip below 5% so that motors are not damaged. Meanwhile standards require electronics to work normally even when AC electric voltage dips 25%.

Meanwhile, a Belkin does nothing for low voltage - that also is not destructive to electronics hardware.

If low voltage is causing electronics problems, then motorized appliances are also at much greater risk. Fix the problem - the reason for low voltage.
 
Last edited:
Quick question. I'm looking to get a UPS. My power supply on the pc is 550W. Do I need to look for higher spec UPS devices our will most do?

Ta
 
So if I've got a low voltage problem in my house, I should buy a UPS for my vacuum cleaner? Hmm... this is a strange turn of events.

Is there any way to prevent surges through telephone wires? We had a lightening strike on the telephone wire outside our house. It buggered all but one of our router's Ethernet ports, and the modem port, as well as the one on my Xbox 360. So unless I get a wireless adapter or send it in for replacement (and naturally I've already voided the warranty, several times) it's no more Xbox live for me.
I'm grateful for wireless now - electricity from the telephone wires can't damage my PC without a wired connection. Maybe we should implement wireless electricity too? This would prevent surges causing damage.
 
Is there any way to prevent surges through telephone wires? We had a lightening strike on the telephone wire outside our house. It buggered all but one of our router's Ethernet ports, and the modem port, as well as the one on my Xbox 360.
Yeah, you can get surge protectors that have phone ports - well worth going for. Personally I think a UPS is overkill, though I consider surge protection essential.
 
Is there any way to prevent surges through telephone wires? We had a lightening strike on the telephone wire outside our house. It buggered all but one of our router's Ethernet ports, and the modem port, as well as the one on my Xbox 360.
First, a surge does not enter on one wire, crash on electronics like a wave on the beach, cause damage, and stop It is electricity. First a complete path from cloud to earthborne charges exists. Same current is flowing through everything in that path - simultaneously. To have damage, something must have both an incoming and outgoing path. An incoming path (ie telephone wire) with no outgoing path (as you defined) means no electricity - no damage. To have a solution, you must define both incoming and outgoing current paths.

Provided in an earlier post were two solutions for phone lines. However, properly earthed phone line protectors means surges on AC electric can obtain earth ground destructively through phone line protectors. Effective protection means every wire in every cable is connected to the same earth ground. If the surge gets inside the building on any wire, then it hunts for earth destructively via household appliances. Effective protection means surge energy must not be inside the building.

Ethernet ports are required to withstand thousands of volts without damage. How much energy did you permit inside the house? Nothing stops that kind of energy. Effective protection means energy gets connected (diverted) to earth before entering a building. What is required on any phone line protector? That essential 'less than 3 meter' connection to earth. Any solution that does not have a dedicated earthing connection will not provide what is requested.

Significant protection already standard on every ethernet port was overwhelmed because you did not earth a surge where it entered the building.
 
I think they are worth while, the only problem I find is that as PC's gradually use more power it's harder to get one that has enough output and is still quiet when mains is on (fan noise during a power fail is fine). Not a problem if you've got somewhere to relocate the UPS, but I've not got the space.

There is also the matter of a pure versus simulated since wave unit and compatibility with some PSUs, but that seems less of an issue with modern PSUs.
 
I'd reccomend the APC smart UPS range,
... which does nothing for phone line protection and does not even claim to "protect your hardware from things like lightening storms and the likes". APC does nothing to protect computer hardware. Its purpose is to protect unsaved data.

How many dishwashers, timer switches, and clock radios have you replaced this month? Oh. Those are on invisible APC UPSes? Using your reasoning, invisible APC protectors must be on all appliances. After all, those invisible UPSes "have never let me down." Your logic - not mine.

It did nothing to protect hardware from lightning and the likes. And it did nothing for phone Therefore it never let you down.

Pure verses simulated sine wave: a typical 230 volt UPS may output two 400 volt square waves with spikes exceeding 500 volts. That is a simulated sine wave that can be harmful to small electric motors. Since protection inside computers is so robust, then 'dirty' electricity output by a UPS was irrelevant to PCs even 30 years ago. All appliances contain significant surge protection – that some instead attribute to invisible or APC UPSes. And does nothing for surges through the phone lines.
 
One thing I will point out is you should be looking for a UPS that produces a pure sine wave on the output (which is what your devices get if connected directly to the mains, or through a surge protector). Most of the cheap ones produce a square/step sine wave, which can cause problems and stability issues with power supplies that use active power factor correction (i.e. pretty much any PSU you can buy these days).
 
Westom, I don't know how it happened but it did. I did not imply that the electricity magically traveled up a dead-end wire with no p.d acoss - that would be a dumb thing to say. Obviously, seeing as every Ethernet port on every device is wired (via the appliance) straight to the mains (or did you forget that?)*, the electricity took this path and the only surviving port was the one which had nothing attached. To break an Ethernet port, only one little wire from somewhere on the board has to get burned-out for it to stop working. And why would an Ethernet port be required to withstand thousands of volts without damage? The standard does not send thousands of volts down a cat 5/5e/6 cable when all a computer needs to measure transfered data is the difference between +5V and 0V.*

(* Or are you under the impression that the Ethernet port provides power? That is the only explanation that fits these two misconceptions)

Anyway [/rant] - so if the surge is strong enough to damage my hardware, I'm completely buggered no matter what I do, unless I have a dedicated earthing wire less than 3m long attached to all (actually, that's only 4) of the wires coming to my house more than 30m away. (OR if we ran off a generator and the only path in/out is one phone line. Then there could be no p.d across the house and hence no destructive discharge).

Fair enough, but I resent the implication that I'm "permitting energy inside my house" :p

Seems a silly question, but if I have a wire from +ve to ground (ie with a p.d across) surely that channels current into the ground? I know you can get trip switches which deal with current, is there a similar jobby for voltage?
EDIT: Another question. If I had a set of very expensive capacitors, then that too would presumably work?
(I'll answer my own question: ) If the average, negative, ground-earth lightening strike transfers an energy of 5 coulombs, then you'd need a capacitor capable of holding 5 coulombs. As 1 coulomb = 1 farad of capacity, and a 1 farad cap is the size of a 2 litre bottle, and costs £30 (what? I'd expect it to cost more) Then with a little redundancy and £180, you could fill six caps each strike, and sell it back to the grid for however much it's worth. My maths (or theory) must be wrong - if it was that easy, people would have been doing it for ages.
 
Last edited:
Most of the cheap ones produce a square/step sine wave, which can cause problems and stability issues with power supplies that use active power factor correction (i.e. pretty much any PSU you can buy these days).
And as nearly all consumer UPSes are "modified square wave" ones it's rather obvious that problems aren't probable.
Active PFC itself is just step up switcher charging primary capacitor also when sone wave's voltage drops from its peak value so there really shouldn't be much reasons for it to not work with squarish input wave form for as long as its controller is doing its job.

Same current is flowing through everything in that path - simultaneously. To have damage, something must have both an incoming and outgoing path. An incoming path (ie telephone wire) with no outgoing path (as you defined) means no electricity - no damage.
Actually in smaller scale lightning strike can probably be very filamental and go along multiple paths. Like single stroke bolt leaving multiple incisions to tree.

And like you said earlier if it's already coming into house then it will find a way forward regardless does that wire end... like in classic situation of telephone blowing up.

I think they are worth while, the only problem I find is that as PC's gradually use more power
That happens only because you people keep insisting on multiple graphics cards and maximum overvolting.
In fact it's still those old P4s which hold the honour of most power hungry CPUs at stock. (although Phenom II 965 is at same level)
 
An incoming path (ie telephone wire) with no outgoing path (as you defined) means no electricity - no damage. To have a solution, you must define both incoming and outgoing current paths.

Interesting, but in that case you will be telling me that phone lines do not consist of two wires and that BT never used gas discharge tubes to protect against lightning strikes to phone lines? :confused: Also, the amount of systems (or components) we replaced due to lighting damage obviously was a figment of my imagination too.
A uni understanding is one thing, but I just see what happens in the real world. And many brighter people than me and you define what happens in the real world.
 
Last edited:
Back
Top Bottom