Author Topic: PC inrush current reduction  (Read 10945 times)

0 Members and 1 Guest are viewing this topic.

Offline djsbTopic starter

  • Frequent Contributor
  • **
  • Posts: 890
  • Country: gb
PC inrush current reduction
« on: October 17, 2011, 06:38:12 pm »
I have had to replace 2 of my PC's power supplies under guarantee because of inrush current problems. I have looked at plugging the PC into a SurgeX AC power conditioner but they are mainly for using with power amplifiers.
As a PC is a more complex load does anyone have any suggestions on what I can use to soften/damp the inrush current? Would the delayed switching of damping resistors in series with the supply work or is it more complex than that?
Thanks for any assistance.

David.
David
Hertfordshire,UK
University Electronics Technician, London PIC,CCS C,Arduino,Kicad, Altium Designer,LPKF S103,S62 Operator, Electronics instructor. Give a man a fish and you feed him for a day. Teach a man to fish and you feed him for a lifetime. Credited Kicad French to English translator.
 

Offline Lightages

  • Supporter
  • ****
  • Posts: 4313
  • Country: ca
  • Canadian po
Re: PC inrush current reduction
« Reply #1 on: October 17, 2011, 06:48:38 pm »
It is not clear what your problem is. Are you saying that the power supplies are self destructing because of the sudden demand of your system when you start your computer? You have systems that draw that much power at startup? If this is what you are saying then there isn't a solution to limit the inrush current by limiting the current to the power supplies from the AC lines. If you try to limit this you can fry more things or the system will not start properly. What you need is bigger and better quality power supplies.

If you are saying the opposite, as in the inrush current that the power supplies are trying to grab from your AC lines and they are frying from not getting sufficient current during startup, then you need something more like a dedicated feed from the distribution box with larger wires, or a local full time high current UPS that can kick in when the current demand causes the voltage drop.
 

Offline djsbTopic starter

  • Frequent Contributor
  • **
  • Posts: 890
  • Country: gb
Re: PC inrush current reduction
« Reply #2 on: October 17, 2011, 09:08:46 pm »
To clarify this is one computer with a Corsair 650W power supply (HX650W). The power supply has a seven year guarantee so it is a good quality one. However they keep failing after around a year or so. This mainly seems to happen when I switch the PSU on with it's on/off switch at the back. The switch goes POP and then a little while afterwards the supply stops working. The supply is limited to 13 amps but for a very short time the inrush current exceeds this as is usual with switch mode supplies. Any ideas?

David.
David
Hertfordshire,UK
University Electronics Technician, London PIC,CCS C,Arduino,Kicad, Altium Designer,LPKF S103,S62 Operator, Electronics instructor. Give a man a fish and you feed him for a day. Teach a man to fish and you feed him for a lifetime. Credited Kicad French to English translator.
 

Offline ejeffrey

  • Super Contributor
  • ***
  • Posts: 3709
  • Country: us
Re: PC inrush current reduction
« Reply #3 on: October 17, 2011, 09:36:26 pm »
To clarify this is one computer with a Corsair 650W power supply (HX650W). The power supply has a seven year guarantee so it is a good quality one.  However they keep failing after around a year or so.

The inrush current is a design parameter of the PSU.  It should be brief and of limited total energy.  If the supply does not limit the inrush to a level it can handle, the supply is not high quality regardless of the length of the guarantee.  The other possibility is something wrong with your mains supply: the voltage is too high or you have lots of surges.  If you have a poor line quality a filter might help, but it will do so not by limiting the inrush current, but by filtering out spikes and surges.

There is an outside possibility that the supply is just overloaded.  That could potentially lead to reduced lifetime although I find it highly suspect.  It is likely that the computer would become unstable and unusable due to brown-outs well before the power supply started failing from overload.  In any case, a higher capacity supply would solve that problem.
 

Offline IanB

  • Super Contributor
  • ***
  • Posts: 11849
  • Country: us
Re: PC inrush current reduction
« Reply #4 on: October 17, 2011, 10:04:57 pm »
To clarify this is one computer with a Corsair 650W power supply (HX650W). The power supply has a seven year guarantee so it is a good quality one. However they keep failing after around a year or so. This mainly seems to happen when I switch the PSU on with it's on/off switch at the back. The switch goes POP and then a little while afterwards the supply stops working. The supply is limited to 13 amps but for a very short time the inrush current exceeds this as is usual with switch mode supplies. Any ideas?
How often do you switch the computer on and off like that? It's a rare thing that I power cycle a computer at the mains switch, probably not more than once every few months. I always use the soft switch on the front of the computer and leave it in standby mode when not in use. It would not surprise me if that is the expected usage and is factored into the design of the power supply.

One other thing I have noticed is that if you switch off a power supply at the mains, you should let it sit for at least five minutes before powering it up again. If you switch it off and on within seconds you seem to get a much larger power surge than otherwise.
 

Offline amspire

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: PC inrush current reduction
« Reply #5 on: October 17, 2011, 11:55:29 pm »
Some of the new generation PC power supplies have a power factor correction regulator on the input. An example is the Antec Green Earthwatch series of supplies.

It would be interesting to find out what chips they are using, and does it mean that they have some degree of inrush current limiting as well?

Richard.
 

alm

  • Guest
Re: PC inrush current reduction
« Reply #6 on: October 18, 2011, 12:03:36 am »
Some power supplies? Active PFC has been mandatory (over 100W or so) in the EU for at least a number of years, so unless they did a separate design for the rest of the world, I would expect all brand-name PSUs to feature active PFC.
 

Offline amspire

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: PC inrush current reduction
« Reply #7 on: October 18, 2011, 12:19:41 am »
Some power supplies? Active PFC has been mandatory (over 100W or so) in the EU for at least a number of years, so unless they did a separate design for the rest of the world, I would expect all brand-name PSUs to feature active PFC.

You are right.  It looks liek the HX650W has it too, so I guess that answers the question - PFC doesn't stop inrush current.

Richard
 

Offline rr100

  • Frequent Contributor
  • **
  • Posts: 339
Re: PC inrush current reduction
« Reply #8 on: October 29, 2011, 12:27:57 pm »
I always killed the power at the power stripe for desktops (and for everything around them) for more than 15 years already. And I have quite a few PCs and some are really beefy (think 100W CPU plus dual slot video card plus 8 hdds, by the way that one has a 550W corsair - CMPSU-550VX and it's quite old). And no problem.
Most likely there's something else at play here, do you have tens of drives or some ridiculous video card? Is the mains power stable for you (some nasty stuff on the AC might damage the PS in such a way that it breaks just as you start it)?
 

Online Psi

  • Super Contributor
  • ***
  • Posts: 9925
  • Country: nz
Re: PC inrush current reduction
« Reply #9 on: October 29, 2011, 12:48:01 pm »
yeah, switching your pc on/off every day using the mains switch on the back of the powersupply maybe wearing out the switch and eventually causing arcing.
I know the mains switches on all the cheap power supplies aren't designed for many cycles but i would have expected corsair to use a decent quality switch.

Probably a good idea to buy a mains surge protector for your PC, just to eliminate power spikes as the cause. Ya don't have to get an expensive one, just $10-20.

If you have lots of drives you could build a relay board with timer to power each drive up in sequence 100ms apart.
That would definitely reduce some of the start-up load.


« Last Edit: October 29, 2011, 12:52:46 pm by Psi »
Greek letter 'Psi' (not Pounds per Square Inch)
 

Online NiHaoMike

  • Super Contributor
  • ***
  • Posts: 9004
  • Country: us
  • "Don't turn it on - Take it apart!"
    • Facebook Page
Re: PC inrush current reduction
« Reply #10 on: October 29, 2011, 12:56:08 pm »
I would recommend a good quality power supply without PFC for a home PC. Much fewer things to go wrong that way.

In my experience, the most common cause of power supply failures are heat related. Add some nice high power Deltas to your case.
Cryptocurrency has taught me to love math and at the same time be baffled by it.

Cryptocurrency lesson 0: Altcoins and Bitcoin are not the same thing.
 

alm

  • Guest
Re: PC inrush current reduction
« Reply #11 on: October 29, 2011, 01:05:04 pm »
Is it still legal to sell >80W power supplies without active PFC in the UK?
 

Online NiHaoMike

  • Super Contributor
  • ***
  • Posts: 9004
  • Country: us
  • "Don't turn it on - Take it apart!"
    • Facebook Page
Re: PC inrush current reduction
« Reply #12 on: October 29, 2011, 01:28:44 pm »
http://dansdata.com/gz028.htm
If they require PFC for home use equipment, it's a waste since it adds cost and decreases efficiency for no real benefit. It makes perfect sense for industrial and commercial use, however.
Cryptocurrency has taught me to love math and at the same time be baffled by it.

Cryptocurrency lesson 0: Altcoins and Bitcoin are not the same thing.
 

alm

  • Guest
Re: PC inrush current reduction
« Reply #13 on: October 29, 2011, 02:57:59 pm »
The benefit is for the mains supply, since reactive power causes I2R losses in the cables, and the weird current waveforms distort the mains supply (which is why mains is often not perfectly sinusoidal). The fact that it doesn't save the consumer any money is a good reason to require it by law, since otherwise nobody would bother.
 

Offline Zero999

  • Super Contributor
  • ***
  • Posts: 19479
  • Country: gb
  • 0999
Re: PC inrush current reduction
« Reply #14 on: October 29, 2011, 04:55:24 pm »
http://dansdata.com/gz028.htm
If they require PFC for home use equipment, it's a waste since it adds cost and decreases efficiency for no real benefit. It makes perfect sense for industrial and commercial use, however.

Quote
Active PFC can iron out lousy power factor better, but it's less efficient, not more; an active PFC circuit will waste some power (at least 10%, in this case) as heat, just like every other circuit in the world.
That's no longer the case, PSUs with active power factor correction are now hardly any less efficient than those with it.

Most modern PSUs also use active power factor correction which is just a boost converter connected before the filter capacitor. A controller ensures the input voltage is in phase with the output voltage.

In the EU all larger SMPSes have to have power factor correction anyway because it reduces power transmission losses and increases the overall efficiency of the electricity grid.
 

Offline PetrosA

  • Frequent Contributor
  • **
  • Posts: 625
  • Country: us
Re: PC inrush current reduction
« Reply #15 on: October 29, 2011, 10:48:22 pm »
I'm also thinking that the switch is the culprit. I don't think those PSU rockers are rated for switch use in the sense of using them repeatedly for turning the computer on and off. A switch rated rocker would be a lot beefier to handle larger inrush currents and the arcs they produce.
I miss my home I miss my porch, porch
 

Offline orbiter

  • Frequent Contributor
  • **
  • Posts: 619
  • Country: gb
  • -0 Resistance is Futile
Re: PC inrush current reduction
« Reply #16 on: October 31, 2011, 03:46:43 pm »
Firstly you could try putting all your computer specs into this online calculator. It gives a reasonable idea of the amount of power a system may require.

Even though you have a good quality Corsair PSU.. at a rated 650W it still could still be overloaded depending on what it's running. For instance with my system running 2x GTX295's, each of which will pull 280W under full load. Add to that a 4.2Ghz overclock, 5 HDD's, 8 fans, sound, tv cards etc, a pretty beefy supply is required. At present I'm using the 1200W Enermax PSU for that lot.

Also as mentioned I don't think turning it on and off all the time with the main switch will be doing it any good, although that still shouldn't be casing such a good brand PSU to fail.
 

alm

  • Guest
Re: PC inrush current reduction
« Reply #17 on: October 31, 2011, 05:26:00 pm »
I'm not familiar with this calculator, but the majority of those calculators produce inflated numbers, possibly to compensate for the fact that many cheap power supplies are unable to deliver their rated power. A multi GPU setup is about the only thing to push a system beyond say 400W. Did you record peak and average power draw for your system to see if you actually exceed 650W? Many systems, especially those with entry level/integrated graphics cards, are in the 100W ballpark, so even 300W is overkill in that case.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf