jonnyGURU Forums
Home Site Search Reviews Articles Contest Links PSU FAQs  


Go Back   jonnyGURU Forums > General Electronics > Electronic Component Discussion

Electronic Component Discussion Discussion of anything pertaining to electronics and the components that make them work

Reply
 
Thread Tools Display Modes
  #1  
Old 03-19-2013
Kougar's Avatar
Kougar Kougar is offline
500W User
 
Join Date: Nov 2012
Posts: 64
Thanks: 4
Thanked 9 Times in 8 Posts
Default UPS Electronics Lifespan

So, as a layperson I understand the lifespan of electronics is generally limited. But what would generally be considered average, good, and max life for the UPS electronics, completely ignoring the batteries? Tried asking a local electronics/computer shop that sells a lot of 'em, but got some pretty general estimates as they didn't know, most sure answer I found was "10 years+ usually".

I've seen enough oddball things online regarding old surge-protected power strips that go bad after 15+ years, so I'm generally curious. Just went through an episode with an older UPS and I'm eager to not repeat it. In the end it ruined a new set of large batteries before I figured out the unit itself was bad.
Reply With Quote
  #2  
Old 03-20-2013
Digerati's Avatar
Digerati Digerati is offline
Built, Broke, Fixed, Learned.
 
Join Date: Jan 2011
Location: Nebraska, USA
Posts: 149
Thanks: 0
Thanked 5 Times in 5 Posts
Default

I think "10 years+" is a fair statement - at least for a "good" UPS. That said, I have an old APC UPS that is pushing 20 years old. But I also recent tossed an APC UPS that gave out after 6.

Quote:
I've seen enough oddball things online regarding old surge-protected power strips that go bad after 15+ years
Actually, I doubt those strips were doing any good. Most surge and spike protectors use MOV (Metal Oxide Varistor) devices. These devices are excellent at absorbing excessive transient voltages however in doing so, they convert that excess into heat. And as noted in my sig, heat is the bane of all electronics. This means every time the MOV takes a hit from an excessive transient, it ages at a faster rate, which degrades its efficiency.

Many say MOVs, and therefore a surge and spike protector is more like a motorcycle helmet - if it protects your noggin once when smashing into a curb, it did its job, but is now weakened and its time to get a new helmet.

A "good" UPS, on the other hand, uses intelligence to compensate for such variables, including component degradation.

My personal opinion are that surge and spike protectors are little more than fancy and expensive extension cords. For severe, abnormal high voltage events like surges and spikes, the S&S protector chops off the tops of the sinewaves ("clamp") leaving a not so pretty voltage waveform for the UPS and other connected devices to clean up.

And of course, a S&S protector does absolutely nothing for abnormal low voltage events like dips (opposite of spikes), sags (opposite of surges), or long duration sags (brownouts) which, if left untreated, put undue strain on your component power supplies and regulator circuits, which generates more heat, which increases component aging.

A "good" UPS with AVR (automatic voltage regulation) not only treats the surges and spikes, the AVR portion cleans up the waveform pattern (somewhat) making it easier on power supplies and regulator circuits. And a "good" UPS with AVR will use the batteries to boost voltage during those abnormal low voltage events.

Note I keep saying "good" UPS with AVR - this is because like PSUs, there are good UPS and there are cheap, lousy UPS.

Are you sure your new batteries were damaged? I ask because SLA batteries are pretty robust. They do wear out and need to be replaced after about 3 years, but they typically pretty tolerant of abuse.

FTR - I think all computers should be on a good UPS with AVR - at least if like me, you value your data more than your hardware. And I recommend all big screen TVs and expensive home theater audio equipment be on a good UPS with AVR.

Note, until now, I have not mentioned backup power during a power outage - that's because backup power is only the icing on the cake. It is the AVR that makes a "good" UPS worth its salt.
__________________
Bill (AFE7Ret)
Freedom is NOT Free!
MS MVP Windows Expert-IT Pro 2007 - 2015

Heat is the bane of all electronics!

───────────────────────────
Reply With Quote
  #3  
Old 03-20-2013
Kougar's Avatar
Kougar Kougar is offline
500W User
 
Join Date: Nov 2012
Posts: 64
Thanks: 4
Thanked 9 Times in 8 Posts
Default

Quote:
Originally Posted by Digerati View Post
I think "10 years+" is a fair statement - at least for a "good" UPS. That said, I have an old APC UPS that is pushing 20 years old. But I also recent tossed an APC UPS that gave out after 6.
Hmm, interesting. I've utilized a variety of UPS's over the last decade, and based on that I'd rate the APC in question as very good. It kept the original batteries working reasonably well up to five years, although the next set only lasted about three. The unit was from 2003 I believe. It began acting like the latest batteries were in need of replacing after just two years, so I replaced them again but the new batteries only worked for a few weeks, then the unit began failing them in its self-tests as well.

Long story short as I was trying to figure out whether to toss it or risk forking another $50 for batteries, the thing stopped turning on entirely when plugged into the wall. Yet it would turn on when attached to the old, original batteries it had been failing. I took that as my answer and tossed it. A few weeks later I randomly noticed the new SLA batteries I bought for it had begun to bulge and swell extremely rapidly with cracks suddenly visible (Which was scary as the APC used a pair of large 12Ahr size batts). Went from perfectly fine to dangerous looking inside weeks while just sitting in a corner, they looked worse than even a few really really old SLAs I've recycled. It was random chance I noticed them at all. So I was out the UPS, the batts, and still had to buy a new UPS.

The second, seperate APC UPS unit I've been running alongside the first only had its batteries last 2, maybe 3 years before it's own self-check failed them. As I was replacing them I found out those had cracked as well while inside the unit. I was not pleased, although that UPS always was warm to the touch so heat probably helped with that.

Regarding AVR you don't need to sell me on it, I've been sold for awhile! Since that first UPS I bought its become a defacto feature on most line-interactive UPS's I noticed. But I do wonder often about the quality of that AVR between brands/models...

---

Separately, while on the topic of UPS systems & build quality though... it seems like the built quality has dropped a fair bit since that 2003 APC unit. It was a huge unit, plenty of space inside and so it never got warm like any modern UPS does. But then I realized something that seems more damning...

APC 1100 BackUPS Pro = 1000VA/670watts rating. Uses two 12Ahr batteries
APC XS 1500 LCD = 1500VA/865watts rating. Uses two 9Ahr batteries

So APC is putting physically smaller, lower rated batteries in units that it is rating for even higher ratings/draw than my original 2003 UPS. Not only that, but the build quality is lower on the 1500 as it is designed to self-heat the batteries.

Last edited by Kougar; 03-20-2013 at 06:12 PM.
Reply With Quote
  #4  
Old 03-21-2013
Digerati's Avatar
Digerati Digerati is offline
Built, Broke, Fixed, Learned.
 
Join Date: Jan 2011
Location: Nebraska, USA
Posts: 149
Thanks: 0
Thanked 5 Times in 5 Posts
Default

3 years is normal for SLA batteries so 5 years is great. Sadly, about the only sure way to ensure the batteries still have enough oomph is to put a load on them and then yank the power cord from the wall and see if your computer comes crashing down. I have a couple 150W light bulbs I use for testing, but it is a pain to tear down my system just to test with a couple lamps. Still, that is better than a catastrophic power outage taking out my data drives.
Quote:
A few weeks later I randomly noticed the new SLA batteries I bought for it had begun to bulge and swell extremely rapidly with cracks suddenly visible (Which was scary
Yeah, that's a bad sign and would be a bit scary - especially if they were not connected to anything as that would indicate a possible internal short.

I wonder if they were bounced around a bit too much during shipping.

Quote:
But I do wonder often about the quality of that AVR between brands/models...
And you should. This is why I keep saying, "good" UPS with AVR. You don't have to spend $400 for a "good" UPS (though you easily can) but you should avoid the budget/entry level models.

To be fair, battery technologies have advanced in the last 10 years so they do provide more punch, or longer usable output than batteries of days past. The battery ratings are not a sign of quality.

Also, the electronics in UPS today compared to UPS of yesteryear have advanced significantly too. If you tore apart an UPS from 10 years ago, you would see a large circuit board with many large, heavy, discrete components. An equivalent UPS today would have a much smaller board with many fewer, integrated circuits (ICs) instead of discrete components.

Designed to self-heat the batteries??? Umm, I don't think so. You don't want your batteries to freeze, but they are not designed to be intentionally heated either - since these UPS are designed for indoor use only, they should never be exposed to extreme temperatures - in storage, or in use.

BTW - Radio Shack will take your old batteries off your hands for recycling and keep them out of landfills. And not just SLA batteries, but lithium (CMOS) type batteries too. Kudos to RS for that.
__________________
Bill (AFE7Ret)
Freedom is NOT Free!
MS MVP Windows Expert-IT Pro 2007 - 2015

Heat is the bane of all electronics!

───────────────────────────
Reply With Quote
  #5  
Old 03-21-2013
westom westom is offline
200W User
 
Join Date: Sep 2010
Posts: 26
Thanks: 0
Thanked 4 Times in 3 Posts
Default

Quote:
Originally Posted by Kougar View Post
But what would generally be considered average, good, and max life for the UPS electronics, completely ignoring the batteries? Tried asking a local electronics/computer shop that sells a lot of 'em, but got some pretty general estimates as they didn't know, most sure answer I found was "10 years+ usually".
How can they provide numbers when they don't have any numbers? Wild speculation.

Electronics should be good for many decades. Why would one UPS fail? Most common source is manufacturing defects. Even electrolytic capacitors manufactured with a defect may fail after seven years. These failure rates are statistically predictable. And not provided to retail people.

Another and significantly less often reason for failure is a destructive transient. One typically occurs every seven years. You had a UPS that worked only fifteen years? Well, how often do you have a potentially harmful spike?

Protection inside a UPS is typically equal or inferior to what is already inside computers. But again, view the numbers. How many joules does that UPS claim to absorb? Read its specifications. 300? Destructive surges are hundreds of thousands of joules. By installing near zero protection, then the naive will insist it does 100% protection. Because the naive ignore numbers.

Reliable numbers are not possible from people who ignore numbers. And who answer technical questions without numbers.

Expect UPS electronics to work for more than 20 years - like any properly designed circuit. That would mean six battery replacements. Since batteries cost almost as much as some UPSes, many just replace the entire UPS. So many UPSes may 'cheat' or 'cost control' their designs since batteries will only last 3 years. Since almost no one is replacing batteries.


Do you need AVR? Well how often do your incandescent bulbs dim? Normal voltage for any computer is even when an incandescent bulb dims to 40% intensity. Better designed AVR is already inside every computer (even long before the IBM PC existed). Are your light dimming even to 70% intensity? If not, then what is AVR doing inside a UPS? Wasting money.

How often does you UPS switch to batteries? Again, control circuits are often so crude that even mild line noise gets confused as a blackout. Observe how often you need that UPS. How often do your incandescent bulbs dim to 50% intensity? Never? Then a UPS is not doing anything useful.

A number that helps identify who really needs a UPS: light dims to 40% intensity.
Reply With Quote
  #6  
Old 03-22-2013
bobnova's Avatar
bobnova bobnova is offline
1kW User
 
Join Date: Jun 2012
Posts: 377
Thanks: 5
Thanked 44 Times in 28 Posts
Default

Hey look it's the Spam About Surges guy!
Not all destructive surges are a bazillion joules. Perhaps you'd like to talk to the surge protector that was added (nothing else was) to my ancient apple II+, and stopped the string of blown to hell PSUs due to bad power.
Reply With Quote
  #7  
Old 03-22-2013
Digerati's Avatar
Digerati Digerati is offline
Built, Broke, Fixed, Learned.
 
Join Date: Jan 2011
Location: Nebraska, USA
Posts: 149
Thanks: 0
Thanked 5 Times in 5 Posts
Default

Quote:
Originally Posted by westom
Protection inside a UPS is typically equal or inferior to what is already inside computers.

Better designed AVR is already inside every computer (even long before the IBM PC existed).
Inferior??? Long before the IBM PC??? This is the same feculent blather you have posted on multiple forums and each time you are shown this is nothing but BS!

You pull numbers out of no where then criticize every one for ignoring your fabricated numbers. You NEVER provide any applicable, supporting documentation for your claims, and you totally ignore all the facts!

And now you are at it again on this site where many if not most of the regular members actually know something about electronics!

Better AVR since long before the IBM PC even existed!!! That's total hogwash!

The ATX12V Form Factor PSU Design Guide requires ATX PSUs maintain a voltage uptime of just 17ms! 17 milliseconds (Page 25, Section 3.2.11)!!!! That is much faster than you can blink. That is MUCH faster than the human eye can detect and it means a power interruption of just 20ms will bring the computer to a screeching halt!

Most movies are recorded at 24 FPS (frames per second) because that is plenty fast for most humans to enjoy flicker-free viewing. The absolute best human eyes are unable to detect any flicker at 30FPS. That equates to 1 frame per 33ms. Plenty of time for a PSU to halt BEFORE any human sees any flicker!

20, 50, 70% intensity??? More hogwash! A noticeable "flicker" is a long power outage when it comes to high speed digital electronics and switching ATX PSUs.

No way a $20 PSU made in some obscure factory in the backwoods of China using parts from their sister factory upriver will have better regulation than a decent UPS.

No way a $15 1500W hair dryer made in a similar 3rd world factory has better (or anything near equal) regulation than a computer power supply - even a budget PSU.

And for the umpteenth time I will point out to you that a PC PSU does not power the monitor or network equipment - which an UPS can easily do (assuming LCD and not CRT monitor).

And before you spew your regular feculent blather about whole-house suppressors, don't! As noted many times before, they do NOTHING for anomalies originating from inside the house or small office - such as from a cheap hair dryer or aging refrigerator compressor, or faulty microwave oven. They are great for suppressing high-voltage anomalies coming off the grid, but that's it.

Quote:
control circuits are often so crude that even mild line noise gets confused as a blackout.
Another pointless obfuscation. (1) Only you are talking about "crude" designs and (2) so what? If the line noise is too dirty, it should switch to batteries.

You are still living in the past, westom. It is time to accept that times have changed and so must you to get current. Your experiences from decades ago working in telco facilities do not apply to today's home and small office environments so stop trying to convince everyone it does.

Quote:
Originally Posted by bobnova
Not all destructive surges are a bazillion joules.
Exactly! In fact, the vast majority of surges and spikes are not destructive at all - but are common, every day (hourly!) occurrences and are not problems. Those are not the ones we worry about.

As I have told westom multiple times, a "good" UPS with AVR is like car insurance. You hope you never need it - but when you do, you want good insurance.
__________________
Bill (AFE7Ret)
Freedom is NOT Free!
MS MVP Windows Expert-IT Pro 2007 - 2015

Heat is the bane of all electronics!

───────────────────────────
Reply With Quote
  #8  
Old 03-22-2013
bobnova's Avatar
bobnova bobnova is offline
1kW User
 
Join Date: Jun 2012
Posts: 377
Thanks: 5
Thanked 44 Times in 28 Posts
Default

The 30FPS thing has been disproved, but on all other points I agree.
Anything less than around 4-6kHz PWM frequency on LEDs drives me nuts.
Reply With Quote
  #9  
Old 03-22-2013
Digerati's Avatar
Digerati Digerati is offline
Built, Broke, Fixed, Learned.
 
Join Date: Jan 2011
Location: Nebraska, USA
Posts: 149
Thanks: 0
Thanked 5 Times in 5 Posts
Default

Quote:
The 30FPS thing has been disproved
I far as I know, it has been contested, perhaps, but not disproved. But that's really a moot point as again, movies, as shown via projectors at movie theaters, are typically at 24FPS. I note Peter Jackson had to jump through hoops to get The Hobbit filmed at 48FPS so 3D would be blur free.

Nevertheless, 48FPS means 1 frame lasts just 20.1ms - still much faster than your eyes can detect or the "flicker" of an incandescent light bulb - and longer than ATX PSUs are required to hold output.
__________________
Bill (AFE7Ret)
Freedom is NOT Free!
MS MVP Windows Expert-IT Pro 2007 - 2015

Heat is the bane of all electronics!

───────────────────────────
Reply With Quote
  #10  
Old 03-22-2013
bobnova's Avatar
bobnova bobnova is offline
1kW User
 
Join Date: Jun 2012
Posts: 377
Thanks: 5
Thanked 44 Times in 28 Posts
Default

Your eyes, maybe. Have you watched The Hobbit in 3d? Motion blur and flicker like whoa. 24FPS/eye is not nearly enough.
For giggles, I built a quick test setup using an Arduino.
It turns an LED on, waits half a second, turns it off for an adjustable amount of time, and then turns it back on and waits some more.

A 20ms off is blindingly obvious to me. I mean OBVIOUS LIKE WHOA. Even in peripheral vision it's blindingly obvious.
A 10ms off is still very obvious.
A 5ms off I have to be looking right at the LED to notice it.
A 2ms off I can just barely see it if I focus on the LED.
A 1ms off is harder still, but visible if I focus on the LED.
A 500Ás off I cannot see even if I focus closely on it. If the LED is moving the streak it leaves behind on my eye has a dim spot, but never goes completely out.

Try it, it's a simple cheap experiment. Write something up in processing to display something and drop the picture for one frame on your monitor. That should be 16ms or so if you're running 60Hz.

Here's some decent reading on the subject:
http://www.100fps.com/how_many_frame...humans_see.htm

One thing that is worth noting is that I'm not perhaps the best human to test on. As a fencer and a racer I'm rather highly tuned towards tiny, brief, visual and auditory things.

Still, if you have a MCU and an LED and a resistor, play with your eyes a bit, it'll change your tune rather seriously.



EDIT:
I'll test my 4 year old and 8 year old when they get home from school.

2.EDIT:
Want to know why they still use 24FPS in theaters? Higher FPS looks too realistic, people get seasick. The slight jerkiness of 24FPS helps prevent that.

3.EDIT:
Tested my 4 year old, he can see a 3ms dropout but not a 2ms dropout.
8 year old isn't home yet.

4.EDIT:
8yo sees 3ms every time, 2ms sometimes, and 1.5ms he sees one off per ten or so. 1ms he can't see.

Last edited by bobnova; 03-22-2013 at 08:22 PM.
Reply With Quote
Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT -4. The time now is 12:18 PM.


Powered by vBulletin® Version 3.8.1
Copyright ©2000 - 2014, Jelsoft Enterprises Ltd.