XPost: rec.electronics, sci.electronics, sci.electronics.basic   
   XPost: sci.electronics.basics, sci.electronics.equipment   
   From: CFKinsey@military.org.jp   
      
   On Thu, 20 Feb 2020 01:29:02 -0000, Jasen Betts  wrote:   
      
   > On 2020-02-20, Commander Kinsey  wrote:   
   >> On Wed, 19 Feb 2020 23:46:45 -0000, Jasen Betts  wrote:   
   >>   
   >>> On 2020-02-19, Commander Kinsey  wrote:   
   >>>> On Wed, 19 Feb 2020 18:54:56 -0000, Phil Hobbs  wrote:   
   >>>>   
   >>>>> On 2020-02-19 10:43, 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.   
   >>>>>   
   >>>>> A lot of cheap supplies regulate only one output, and rely on   
   >>>>> cross-regulation via the transformer to control the others.  If the   
   >>>>> regulated output isn't loaded, it rises out of spec and so do the others.   
   >>>>   
   >>>> Strangely, with no 5V load, I get 5.2V and 10.5V.  A small rise and a   
   large drop.   
   >>>>   
   >>>> I can't understand why the following happens: No 5V load, 12V is out.    
   Small (2A) 5V load, 12V is ok.  Yet if I draw 30A from 5V, the 12V is still   
   ok?  How can zero load upset it, but 2A or 30A (big difference) both be ok?   
   >>>   
   >>> The difference between nothing and 2A is a factor of infinity   
   >>> The difference between 2A and 20A is only a factor of 10   
   >>>   
   >>> Actually the fisrt is more like the difference betweem 2mA and 2A   
   >>> because the internal feedback takes avout 2mA to run the LM431 and the   
   >>> optocoupler. So going to 2A loads the 5V output by 1000 times more, better   
   >>> than infinity. but not by much.   
   >>   
   >> If it's going to be so shit, they could have added a dummy load inside the   
   PSU to make it work properly.   
   >   
   > They could, but it was already working propperly with their existing   
   > external loads.   
      
   They assume all computers take a couple of amps at 5V.  With modern machines   
   that may not be true.  Almost everything takes 12V now.   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)