There are some rumurs around the web that ATX power supplies should not be turned on with no load. I have a partially unstable PSU that I want to turn it on out of PC case and check the voltages with multimeter.

Can you tell me having no load is dangerous or not? is it a myth?

  • Have you tried reading the documentation for your power supply? – Phil Frost Feb 15 '13 at 14:28
  • 6
    No they won't explode. Most ATX PSU just won't power up without a load and start making a clicking sound, caused by an internal overvoltage safety tripping. There is a lot of information on how to run a ATX PSU outside the PC on the Internet, that is a fact. I believe it usually includes connecting a power resistor between correct pins. Voltages inside a PSU can be lethal, so be very careful if you're planning to open it up. – jippie Feb 15 '13 at 14:45
  • 3
    The ATX standard has required power supplies to be safe under no load condition for many years now. For example, in 2004, the spec said, "No damage or hazardous condition should occur with all the DC output connectors disconnected from the load. The power supply may latch into the shutdown state." – David Schwartz Feb 15 '13 at 15:11
  • 1
    Well there is no documentation for the PSU and generally they dont provide documents for consumer PSUs – Sean87 Feb 15 '13 at 15:19
  • They generally do state what standards they comply with. For example, I just randomly clicked on the top power supply link on NewEgg and it says, "ATX12V 2.3 / EPS12V 2.91". ATX12V v2.3 has a section on no load operation. – David Schwartz Feb 15 '13 at 15:26

There should be no danger in turning on an ATX power supply at zero load.

That being said, the outputs may be out of regulation, hunting (overshooting and undershooting their setpoints), or absent. None of these things are 'dangerous' per the definitions of safety used by regulatory agencies like UL.

What this means is unless you apply some load, you won't be able to determine anything useful by probing it while disconnected from the PC. How much "some load" is will likely vary from model to model, but if you don't want to shell out for a full-blown PSU tester as Kruug suggested, try loading each rail with power resistors to 10% of its label-rated maximum, then test with your multimeter. If the behaviour is still bad, you may indeed have a defective power supply.

| improve this answer | |

Not true. Switch mode power supplies (typical PC PSU) will blow up with no load. The fan counts as a load. Take out the fan and watch the voltage go up until it smokes. I know, I fix the things and have blown up many with no load. Switch mode means it takes the AC in, converts it to DC, then switches it to high voltage, high frequency, then steps it back down again, then converts back to DC. This is far more efficient as transformers don't perform as well at lower frequencies and lower voltages.

It's not like mains power or a battery which can sit around doing nothing. I would say from experience that one in four can be blown up in this way, most likely not much to do with cheapness or quality of design as just about every PSU brand (nearly every brand is different) has a different circuit for it, and even the same brands scarcely keep the same circuit for long as they always source the cheapest components available at the time.

The underlying technology has hardly changed since the the days of XT computers, nothing really "modern" about it. You won't find any more modern fancy components in newer PSU's.

| improve this answer | |
  • If you ever return here: what is not true? It seems to me your answer applies to ATX power supplies, right? – Arjan Aug 30 '17 at 17:33

It is not a myth. In older/cheaper units the same transformer is used to provide power for both rails - 3.3/5V and 12V. The load on those 2 rails had to be balanced. Otherwise the voltage regulator could not guarantee that the voltage would to stay within the standard limits for both rails. People were not in any real danger but devices connected to such supplies were. Modern power supplies (but the cheapest ones you don't want to use anyway) do not have this problem anymore.

| improve this answer | |
  • 1
    I would expect that by the time the ATX standard was developed, power-supply designs had evolved to the point that the issues you cite are not a factor. I think the biggest issue wouldn't have been that things be nearly balanced, but rather that the imbalance was in the proper direction, since the ground rail may be designed to source current, or to sink current, but may not be able to do both. – supercat Feb 15 '13 at 17:21
  • 1
    @supercat: Wrong. I have just fried my 2001 Chinese model with a paperclip-startup + no load. – Sz. Jun 25 '18 at 17:38

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.