XPost: rec.electronics, sci.electronics, sci.electronics.basic   
   XPost: sci.electronics.basics, sci.electronics.equipment   
   From: CFKinsey@military.org.jp   
      
   On Wed, 19 Feb 2020 19:37:06 -0000, default  wrote:   
      
   > On Wed, 19 Feb 2020 18:55:27 -0000, "Commander Kinsey"   
   >  wrote:   
   >   
   >> On Wed, 19 Feb 2020 18:45:55 -0000, default  wrote:   
   >>   
   >>> On Wed, 19 Feb 2020 15:43:37 -0000, "Commander Kinsey"   
   >>>  wrote:   
   >>>   
   >>>> Why do (cheap? expensive ones may be better) PC ATX power supplies need   
   current drawn from the 5V line to make the 12V line work correctly?   
   >>>>   
   >>>> I have a PC with 3 graphics cards running scientific applications.  I   
   acquired three old graphics cards that take about 300W each, and have loads of   
   cheap (CIT) PSUs that are rated at 650W on the 12V line, which is what those   
   cards use.  So I run    
   each card off its own supply.  But the 12V line at no load, or even at 300W,   
   is only giving out 10 to 10.5V.  If I attach a small dummy load of an amp or   
   so to the 5V line, the 12V line suddenly becomes 12V.   
   >>>>   
   >>>> Why are the two lines related in any way?   
   >>>>   
   >>>> Sorry for the crosspost, I'm not sure which of these groups are active.   
   >>> \   
   >>> There's often a good reason for it.  The 5 volt supply is regulated   
   >>> and the others are not (generally speaking).  The feedback path is   
   >>> from the 5VDC output back to the mains side of the controller.   
   >>>   
   >>> The others get line (but not load) regulation via the 5V supply   
   >>> because they share a common transformer.  They are also switching   
   >>> supplies that work at a high frequency so the transformers have fewer   
   >>> turns of wire and more volts per turn which results in excellent   
   >>> transformer "regulation."   
   >>   
   >> So on a cheap shit supply, the 5V is guaranteed to be very close to 5V, but   
   the 12V will drop under heavy load?   
   >> And on a decent supply like Corsair, they must regulate both seperately?   
   >> I still don't understand why the regulation goes to pot when under 1.5A is   
   taken from 5V.  It still regulates that 5V perfectly with no load, but the 12V   
   goes wildly wrong.  Why does the regulation need  current to be flowing   
   through 5V?   
   >   
   > I suppose there may be some that regulate the 12 for both line and   
   > load but I haven't seen it.  Not having separate regulation for the   
   > coarse 12 supply isn't necessarily a design flaw.  The 12V is less   
   > critical,   
      
   If you look at the 12V line on a Corsair supply under any load, it will always   
   be between 11.8 and 12.  A cheap PSU like CIT, with the 5V loaded normally,   
   the 12V can be 11V to 12V depending on its own load.  Buy an Alpine supply and   
   it will literally go    
   bang if you exceed 50% load for more than an hour.  I went through 10 Alpine   
   supplies, costing the shop a fortune, before I told them enough was enough and   
   got my money back.   
      
   > Some (particularly older) designs do require some minimal loading of   
   > the 5V to keep it from drifting above 5, or above 5 AND causing the   
   > crowbar over-voltage protection from kicking in and shorting it to   
   > ~1V.   
      
   The supplies I'm having bother with are not that old, probably 5 years.  But   
   they were the second cheapest.  They also lie on their specs.  They're sold as   
   850W supplies, but you can only draw 650W of that on the 12V line, which is   
   where 99% of the power    
   goes in a modern PC.   
      
   > 5V is more critical since a lot of components start smoking at 5.2V or   
   > higher and those are the logic components.   
      
   Nowadays, aren't all the chips running at about 1V and powered by their own   
   VRMs, fed off the 12V line?   
      
   > The 12V runs fans, platter motors in CDROM, disk drive motors, etc.   
   > and may be used for RS232 protocols, and it can be regulated on the   
   > board close to the load if the load is critical.  The -12V is often   
   > pretty wimpy power-wise and it's function is for bias and/or   
   > communications protocols.   
   >   
   > The PS should work for the purpose intended, but if you are trying to   
   > use it as a general bench power supply you may encounter issues with   
   > poor regulation.   
      
   I am using it for PC components, but only for one component, the graphics   
   card, which only has a 12V input.   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)