Page 1 of 2 12 LastLast
Results 1 to 10 of 16

Thread: Advanced transient testing

  1. #1
    Join Date
    Jan 2010
    Location
    Greece
    Posts
    1,528
    Thanks Thanks Given 
    185
    Thanks Thanks Received 
    7
    Thanked in
    4 Posts

    Default Advanced transient testing

    I wanted to add some more advanced trancient tests in my reviews so after a little research (and examining HOCP's excellent transient tests) I did the following.

    I programmed my loader to add about (because my loader uses resistors the actual load depends on the voltage rails) 10A at 12V, 5A at 5V and 2.2A at 3.3V, all at the same time and for 50ms. The above Amps are added when the PSU already runs at 20% and 50% of full load (two different transient tests). Through the labjack (which is able to log 50ks/sec) I log all voltage fluctuations at 12V, 5V, 3.3V and 5VSB (although I do not add any load at 5VSB during testing).

    One of my recently reviewed samples dropped the 3.3V rail to 3.071V during the above test with 50% load.

  2. #2
    Join Date
    Oct 2006
    Location
    Moderator
    Posts
    4,126
    Thanks Thanks Given 
    19
    Thanks Thanks Received 
    14
    Thanked in
    7 Posts

    Default

    Quote Originally Posted by crmaris View Post
    One of my recently reviewed samples dropped the 3.3V rail to 3.071V during the above test with 50% load.
    That is certainly possible since your total load is considerably heavier than mine and it was at 50% already. The big thing to watch is if you review stuff in the ~400W-500W range not to overload the units. You will have enough that fail even without doing that

  3. #3
    Join Date
    Jan 2010
    Location
    Greece
    Posts
    1,528
    Thanks Thanks Given 
    185
    Thanks Thanks Received 
    7
    Thanked in
    4 Posts

    Default

    in small units I may ran the test only at 20% load, although the ATX specs (2.2) at section 3.2.7 give pretty big transient load steps (**For example, for a rated +5 VDC output of 18 A, the transient step would be 30% × 18 A = 5.4 A).

    Also in the same section says "Load-changing repetition rate of 50 Hz to 10 kHz"

    I could program my loader to go between test#1 (20% load) through test#6 (100% load) at 100-200 Hz and monitor all rails, but i seriously doubt if any PSU would pass this test. So I will leave it as it is (I don't want also to stress the relays too much)

  4. #4
    Join Date
    Jan 2010
    Location
    SoCal
    Posts
    1,038
    Thanks Thanks Given 
    2
    Thanks Thanks Received 
    0
    Thanked in
    0 Posts

    Default

    If you ever do rapid, repeated load changes like that, make sure to get the fireworks on video.
    It's my PSU in a box!
    Ooo-ooh,
    My PSU in a box, baby!

  5. #5
    Join Date
    Jan 2010
    Location
    Greece
    Posts
    1,528
    Thanks Thanks Given 
    185
    Thanks Thanks Received 
    7
    Thanked in
    4 Posts

    Default

    thats the main reason that I always test psus inside the hotbox

  6. #6
    Join Date
    Jan 2010
    Location
    Greece
    Posts
    1,528
    Thanks Thanks Given 
    185
    Thanks Thanks Received 
    7
    Thanked in
    4 Posts

    Default

    some early results

  7. #7
    Join Date
    Oct 2006
    Location
    Moderator
    Posts
    4,126
    Thanks Thanks Given 
    19
    Thanks Thanks Received 
    14
    Thanked in
    7 Posts

    Default

    I would use the second kind of chart not the first...then again that is they way I make mine...so I may be biased...but I really do think it is the best way to show the data.

  8. #8
    Join Date
    Jan 2010
    Location
    Greece
    Posts
    1,528
    Thanks Thanks Given 
    185
    Thanks Thanks Received 
    7
    Thanked in
    4 Posts

    Default

    thanks Spectre Actually I am thinking of using both types (In the end my review will be full of charts )

  9. #9
    Join Date
    Jan 2010
    Location
    Greece
    Posts
    1,528
    Thanks Thanks Given 
    185
    Thanks Thanks Received 
    7
    Thanked in
    4 Posts

    Default

    Quote Originally Posted by Spectre View Post
    I would use the second kind of chart not the first...then again that is they way I make mine...so I may be biased...but I really do think it is the best way to show the data.
    I suspect that the labjack, due to software limitations, cannot catch the right peak voltage drop. So, I tried to measure it with the Stingray but I run onto some problems. In some rails I see the voltage drop but mainly at 12V I catch only a raise and not a drop

    How do you set up the stingray for your transient measurements? You use AC coupling with positive trigger at some mV, right? Also shouldn't the trigger be negative, since you want to catch the voltage drop?

  10. #10
    Join Date
    Oct 2006
    Location
    Moderator
    Posts
    4,126
    Thanks Thanks Given 
    19
    Thanks Thanks Received 
    14
    Thanked in
    7 Posts

    Default

    I am actually away from home right now but I will get you the setup when I get back.

Similar Threads

  1. Will We Eventually Be Seeing More Complicated Transient Response Testing?
    By turkey3_scratch in forum PC Power Supply Discussion
    Replies: 3
    Last Post: 07-19-2016, 08:57 PM
  2. NSC: Load Transient Testing Simplified
    By Travis in forum Testing Methodology Discussion
    Replies: 8
    Last Post: 09-16-2010, 11:37 PM
  3. Overshoot Transient testing query
    By crmaris in forum Testing Methodology Discussion
    Replies: 2
    Last Post: 03-11-2010, 01:02 PM
  4. Advanced Robot.
    By yakko in forum Off Topic
    Replies: 4
    Last Post: 03-03-2007, 07:58 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
  •