Page 4 of 5 FirstFirst ... 2345 LastLast
Results 31 to 40 of 47

Thread: Low voltage on the 12V rail in a brand new Cx450

  1. #31
    Join Date
    Jul 2007
    Location
    Sweden
    Posts
    348
    Thanks Thanks Given 
    14
    Thanks Thanks Received 
    3
    Thanked in
    3 Posts

    Default

    There is a sense wire (mandatory for 3.3v, optional for EPS 12v) that is used for feedback to the PSU to compensate for voltage drop in the cable.
    As you have some load on the PSU it will compensate for this loss.
    But since you are measuring on an unloaded molex connector you will not see this drop with your meter.
    Instead what you will basically see is how much the PSU compensates for the voltage drop in the main ATX cable.

    Short version: yes it's normal to see a small increase in voltage on a molex connector as load increases for this reason.
    "The one who says it cannot be done should never interrupt the one who is doing it."

  2. The Following User Says Thank You to Per Hansson For This Useful Post:

    ssm002 (12-15-2017)

  3. #32
    Join Date
    Dec 2017
    Location
    india
    Posts
    30
    Thanks Thanks Given 
    11
    Thanks Thanks Received 
    0
    Thanked in
    0 Posts

    Default

    Thanks ,so can it be assumed that 12.3v is the actual voltage that's being supplied to the motherboard by the psu under load,or is it an adjusted figure that does not truly reflect what the voltage under load really is?

    If its not totally accurate,then how much of a drop in voltage can there really be from 12.27V under load conditions?Can it possibly go down to as low as 11.66v as hwinfo had been showing?

    Also is the reading of 12.27v when idle accurate at all and truly shows the actual voltage that the motherboard is receiving from the psu?Is testing a molex connector a reliable way to measure the voltage on the 12v rail?

  4. #33
    Join Date
    Nov 2006
    Location
    Slovenia
    Posts
    91
    Thanks Thanks Given 
    8
    Thanks Thanks Received 
    2
    Thanked in
    2 Posts

    Default

    It's possible to get the DMM probes into connectors from the same side as the cables go, so you can measure the actual voltage under load.
    R7 1700x @3,95
    Geil EVO x 16gb ddr4 3200
    Asus Prime x370 pro
    Vega 64
    500GB Samsung 970 pro
    Corsair RM1000i
    EKWB watercooling (cpu+gpu)

  5. #34
    Join Date
    Sep 2017
    Location
    Romania
    Posts
    226
    Thanks Thanks Given 
    95
    Thanks Thanks Received 
    0
    Thanked in
    0 Posts

    Default

    i was running out of patience,so i decided to test the psu today and i've got some rather surprising(or should i say shocking?!)results-i tested the 12v rail by connecting the leads of the multimeter to the 12v and ground wires of a spare molex connector-while the pc was idling,hwinfo displayed 11.95/11.88V whereas the reading on my multimeter was 12.27V,and when the pc was being subjected to load(i ran a couple of games and 3d mark benchmarks for a few minutes)hwinfo reported 11.73/11.66V whereas my multimeter gave a reading of 12.29-12.30V(the voltage seemed to have increased slightly rather than decrease,unlike what was being shown in hwinfo)!
    This is another proof that the voltages on the HWinfo reading are just shitty. In your case the difference is higher than in both my cases. Kinda 0.6volts diference! and that's huge.

    So are the readings taken using the multimeter the real deal?If yes,then it can be safely concluded that hwinfo was reporting completely bogus and inaccurate values.Is there much of a discrepancy between the 12v output of a molex connector and the 12v that goes into the motherboard(via the 24 pin atx connector)?Are these voltages fairly similar to each other or can they differ to some extent?If not,then i suppose my psu is fine,right?
    Now we are on the other side of things.
    The value you taken with the multimeter at molex connector is in the ATX specifications and are normal voltages. Like Per Hannson noted, at the molex connector the voltage will always have a little increase when the PSU is on load.

    Thanks ,so can it be assumed that 12.3v is the actual voltage that's being supplied to the motherboard by the psu under load,or is it an adjusted figure that does not truly reflect what the voltage under load really is?
    If its not totally accurate,then how much of a drop in voltage can there really be from 12.27V under load conditions?Can it possibly go down to as low as 11.66v as hwinfo had been showing?
    For measure the voltage at the motherboard you need to measure the 12v rail voltages at the 24 pin connector and CPU connector. To be sure about other cables take a measurment at the PCI-E connector and SATA one also. In this way you will have a complete set of real values at 12V rail, and, if the voltages are in the range 11.4 - 12.6 is mean that your PSU meet the ATX specifications with the 12V rail.

  6. The Following User Says Thank You to Al3xand3r For This Useful Post:

    ssm002 (12-16-2017)

  7. #35
    Join Date
    Dec 2017
    Location
    india
    Posts
    30
    Thanks Thanks Given 
    11
    Thanks Thanks Received 
    0
    Thanked in
    0 Posts

    Default

    Given that the voltage from the molex connector turned out to be 12.27V,can it be safe to assume that the voltage on the other cables(such as sata,24pin connector,EPS etc)is also around 12.27V?

    Is it possible that there might be a large degree of variation between the voltage on the 12v rail that is being provided by the psu via different connectors or is it fairly at a constant level for all of them?

    Also i used a generic multimeter to take the readings-could it possibly have affected the testing in anyway or resulted in values that are far from accurate?

  8. #36
    Join Date
    Nov 2017
    Location
    Somewhere in the world...
    Posts
    51
    Thanks Thanks Given 
    6
    Thanks Thanks Received 
    0
    Thanked in
    0 Posts

    Default

    Quote Originally Posted by ssm002 View Post
    Also i used a generic multimeter to take the readings-could it possibly have affected the testing in anyway or resulted in values that are far from accurate?
    If yours have user manual, you can look for the accuracy values.

  9. #37
    Join Date
    Jul 2007
    Location
    Sweden
    Posts
    348
    Thanks Thanks Given 
    14
    Thanks Thanks Received 
    3
    Thanked in
    3 Posts

    Default

    Quote Originally Posted by Overclocked View Post
    If yours have user manual, you can look for the accuracy values.
    Honestly for a meter like the OP showed those numbers are as credible as your run of the mill Deer PSU that costs $7 and delivers 750w according to it's specifications.
    "The one who says it cannot be done should never interrupt the one who is doing it."

  10. #38
    Join Date
    Dec 2017
    Location
    india
    Posts
    30
    Thanks Thanks Given 
    11
    Thanks Thanks Received 
    0
    Thanked in
    0 Posts

    Default

    Greetings and a very happy new year to my fellow members!

    I tested my cx 450 with a better quality mastech multimeter today and got some rather startling results.From a spare molex connector,the voltage readout that i got on idle was around 11.71v and under load it increased to around 11.74V.

    I also tested a pcie connector and found it to be around 11.68v on idle,which rose to around 11.71v on load.

    Does it mean my cx 450 psu is faulty?if cybernetics' test report on this psu is to be trusted,a good cx 450 isn't supposed to output such low voltages-even under load,the voltage on the 12v rail of their test unit didnt drop below 11.9v

    to confirm that the low readings weren't being caused by the meter itself,i tested the 12v output from a molex connector of another psu thats also in my possession(seasonic s12ii 620)-it was 12.07v on idle which dropped slightly to around 12.05v on load.

    If i were to get rid of this psu and purchase a different one,which would be a much better(and more reliable)alternative to my cx 450?are corsair's newer tx550m & 650m (gold certified)psus any good??

  11. #39
    Join Date
    Mar 2015
    Posts
    260
    Thanks Thanks Given 
    88
    Thanks Thanks Received 
    3
    Thanked in
    3 Posts

    Default

    The ATX allow +-5% from the main rail.
    It means 11.4 to 12.6V (line regulation).
    Most PSU reviews check this and load regulation.
    Load regulation it's the change in load, normally it should not go more than 5%, I didn't found that it written.
    Your line regulation it's 2.5%.
    Your load regulation to the max load you tested was around 0.3%.
    In the old days 3% in both line, load regulation was good.
    Now days......
    In other word you got O.K. line regulation and very good load regulation by absolute resault.
    If that not normal to this PSU that is another story.

  12. #40
    Join Date
    Oct 2017
    Location
    Grand Junction, CO
    Posts
    1,964
    Thanks Thanks Given 
    114
    Thanks Thanks Received 
    55
    Thanked in
    48 Posts

    Default

    Here is a cool voltage drop calculator

    https://www.supercircuits.com/resour...rop-calculator

    One can drop 0.32V running 20 amp on 2 ft of 16 gauge cable
    Last edited by ashiekh; 01-02-2018 at 01:25 PM.

Similar Threads

  1. Seasonic X-750, Low 3.3V rail voltage, should i be worried?
    By trumee in forum PC Power Supply Discussion
    Replies: 2
    Last Post: 09-03-2018, 01:07 PM
  2. Voltage stability on the +5VSB rail chart
    By xpwner96 in forum PC Power Supply Discussion
    Replies: 2
    Last Post: 12-15-2017, 06:31 AM
  3. Replies: 6
    Last Post: 10-27-2017, 09:19 AM
  4. PSU voltage rail discussion
    By HOOfan_1 in forum PC Power Supply Discussion
    Replies: 20
    Last Post: 07-15-2008, 02:28 PM
  5. Failing voltage rail tests
    By jcgamo88 in forum PC Power Supply Discussion
    Replies: 1
    Last Post: 08-11-2007, 09:42 AM

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •