Results 1 to 9 of 9

Thread: Corsair CX transients

  1. #1
    Join Date
    Aug 2019
    Location
    Somewhere
    Posts
    123
    Thanks Thanks Given 
    38
    Thanks Thanks Received 
    13
    Thanked in
    11 Posts

    Default Corsair CX transients

    https://www.tomshardware.com/reviews...su,5678-8.html

    Is there something I am missing, or does the transients for the CX seem really bad, doubly so for an LLC resonant PSU?

  2. #2
    Join Date
    Oct 2006
    Location
    jonnyGURU forums, of course!
    Posts
    16,318
    Thanks Thanks Given 
    545
    Thanks Thanks Received 
    339
    Thanked in
    247 Posts

    Default

    Quote Originally Posted by hello_there_123 View Post
    https://www.tomshardware.com/reviews...su,5678-8.html

    Is there something I am missing, or does the transients for the CX seem really bad, doubly so for an LLC resonant PSU?
    How do you figure?

    They're perfectly fine except for the +3.3V and +5V, which run off a DC to DC circuit. Has nothing to do with the +12V being LLC or DF or whatever. But regardless.... you're not going to have transient loads like that on the +3.3V and +5V.

  3. #3
    Join Date
    Aug 2019
    Location
    Somewhere
    Posts
    123
    Thanks Thanks Given 
    38
    Thanks Thanks Received 
    13
    Thanked in
    11 Posts

    Default

    Quote Originally Posted by Jon Gerow View Post
    How do you figure?
    Has nothing to do with the +12V being LLC or DF or whatever.
    Wait. So, you're telling me that LLC vs DF does not matter for the +12V transients? Unless I am mistaken in the interpretation, you've said that LLC used in the CX results in better handling of GPU transient peaks compared to DF of CXm...

    Also, CXm transients: https://www.tomshardware.com/reviews...su,4770-8.html

    DF CXm seems to have way better +12V transients than the GW CX...

  4. #4
    Join Date
    Oct 2006
    Location
    jonnyGURU forums, of course!
    Posts
    16,318
    Thanks Thanks Given 
    545
    Thanks Thanks Received 
    339
    Thanked in
    247 Posts

    Default

    Quote Originally Posted by hello_there_123 View Post
    Wait. So, you're telling me that LLC vs DF does not matter for the +12V transients? Unless I am mistaken in the interpretation, you've said that LLC used in the CX results in better handling of GPU transient peaks compared to DF of CXm...
    I didn't say that at all.

    The poor transient response is coming from the DC to DC (the +3.3V and +5V). Poor transient response from the DC to DC has nothing to do with the PSU's topology. It has to do with the DC to DC.

    The +12V transient response is good. As it should be as the PSU uses an LLC toplogy.

    Also, topology alone is not going to improve transient response.

    And you're also quoting me out of context when you say I said that LLC generally handles transients better. What I had said is that the hard switching of DF puts a lot of strain on the main transformer when there are heavy transient spikes. If anything, you would see a larger dip in output voltage with LLC because it takes more time for the converter to respond.

    Your real benefits to LLC is lower switching losses, which results in better efficiency, and lower EMI.
    Last edited by Jon Gerow; 12-04-2019 at 10:44 AM.

  5. The Following 2 Users Say Thank You to Jon Gerow For This Useful Post:

    hello_there_123 (12-04-2019), rgbodosk (12-04-2019)

  6. #5
    Join Date
    Aug 2018
    Location
    Korea
    Posts
    322
    Thanks Thanks Given 
    150
    Thanks Thanks Received 
    21
    Thanked in
    16 Posts

    Default

    Does transient response matters with psu capacity? In Aris reviews: " We call these tests "Advanced Transient Response Tests," and they are designed to be very tough to master, especially for a PSU with a capacity of less than 500W. "

  7. The Following User Says Thank You to rgbodosk For This Useful Post:

    Mezoxin (12-04-2019)

  8. #6
    Join Date
    Oct 2017
    Location
    Grand Junction, CO
    Posts
    2,134
    Thanks Thanks Given 
    119
    Thanks Thanks Received 
    70
    Thanked in
    63 Posts

    Default

    Quote Originally Posted by hello_there_123 View Post
    https://www.tomshardware.com/reviews...su,5678-8.html

    Is there something I am missing, or does the transients for the CX seem really bad, doubly so for an LLC resonant PSU?

    Perhaps this is where a small by-pass capacitor in parallel with the main smoothing capacitor might help.

    One might think a larger smoothing capacitor would help, but this would actually have the opposite effect for short transients due to stray inductance. Two 'half' capacitors in parallel are sometimes used instead of one to achieve the same capacitance with reduced inductance (capacitors in parallel lead to greater overall capacitance, inductors in parallel lead to lesser overall inductance).


    However what is seen is problems at the kHz level, when the stray inductance is probably not an issue.
    Last edited by ashiekh; 12-05-2019 at 01:57 AM.

  9. The Following User Says Thank You to ashiekh For This Useful Post:

    hello_there_123 (12-04-2019)

  10. #7
    Join Date
    Aug 2019
    Location
    Somewhere
    Posts
    123
    Thanks Thanks Given 
    38
    Thanks Thanks Received 
    13
    Thanked in
    11 Posts

    Default

    Quote Originally Posted by Jon Gerow View Post
    I didn't say that at all.

    The poor transient response is coming from the DC to DC (the +3.3V and +5V). Poor transient response from the DC to DC has nothing to do with the PSU's topology. It has to do with the DC to DC.

    The +12V transient response is good. As it should be as the PSU uses an LLC toplogy.
    Ah, I see. I should've clarified that I was only talking about the +12V rail. Sorry about that.

    However are the +12V transients really that good? I'm seeing even the Corsair VS450 do way better per Tom's Hardware on the +12V transients, and the EVGA BT doing somewhat similarly... What am I misunderstanding? Is it one of those things where it doesn't matter as long as it's within ATX spec or another certain defined range?

  11. #8
    Join Date
    Sep 2010
    Location
    New Jersey, USA
    Posts
    1,398
    Thanks Thanks Given 
    914
    Thanks Thanks Received 
    21
    Thanked in
    20 Posts

    Default

    Isn't the CX line a budget line?
    Core i7 4770K @4.3GHz + iGPU | GA-Z87X-UD4H | 2x8GB Crucial Ballistix DDR3@1600MHz
    Prolimatech Armageddon heatsink w/ 2 x 140mm NF-A14 PWM fans
    Samsung 840 Pro 512GB SSD | WD Black 4TB file disk | WD Red 3TB backup disk
    Fractal Design Define R5 case w/2 x NF-A14 ULN front fans | no exhaust fan; grill removed
    PSU Seasonic X-650 Gold

  12. #9
    Join Date
    Oct 2006
    Location
    jonnyGURU forums, of course!
    Posts
    16,318
    Thanks Thanks Given 
    545
    Thanks Thanks Received 
    339
    Thanked in
    247 Posts

    Default

    Quote Originally Posted by ehume View Post
    Isn't the CX line a budget line?
    It is. And transient response doesn't even play a part on how it can handle high transient loads. It's how you have the OCP and/or OPP set for those big spikes, because the voltages drop isn't enough to trigger UVP. But you can tweak OCP and OPP to not only trip at a certain wattage, but offer a delay for when it should actually trip.

    Believe it or not, Corsair actually tests all of their power supplies with real PCs before they even launch. And the most grueling PC is still the dual R9 295x2 and the program that will likely cause a reboot is still Catzilla. We do capture the transient spike on an oscilloscope, but we look more at the amperage of the load and not so much drop of the voltage (though it is measured... it's always within spec).

  13. The Following 2 Users Say Thank You to Jon Gerow For This Useful Post:

    hello_there_123 (12-05-2019), rgbodosk (12-05-2019)

Posting Permissions

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