Anyone knows will the power supply turns on (eg fan starts to rotate) if I just attach to the power by itself, without to the motherboard? I wanted to test whether this new power supply I have just bought is working?? Thankz..
It is a switching PSU. That is the technical reason (if you know about switching PSUs - which I don't, but I got it from someone who does) I don't know the technical details... hang on, lemme look it up...
Here is a short and not technical version. If you want a more technical version I can try to dig one up for you.
The ATX design guide (here's on, its on pg 29) says that the PSU must allow for the condition without damage, but that doesn't mean it is done well, or that they test for it... (i.e., most psu's are crap, and even if it does have that safety feature its better safe than sorry).
Yeah I do know a little about them. Part of my job is designing them into the products my company sells.
I have a couple of products we sell piles of every day which are standalone switching supplies. Both are perfectly happy sitting unloaded for all time maintaining specified output voltage and continuing to work as specified when a load is applied. I can tell you with no amount of uncertainty not all switching supplies fail when there is no load applied.
I don't know what is going on with that link you supplied. Either those guys are smoking the good stuff or there is some arcane ATX spec I am unaware of which requires/allows this behavior. If it is the spec thing then that certainly doesn't mean all switchers can't maintain good behavior unloaded.
In my opinion supplies that don't behave well under all loading conditions shouldn't even get out the door. That is just bad design practice. I am shocked to consider the possibility that there are companies who would even offer such crap to their customers and they have to give them money for it.