Page 7 of 8 FirstFirst ... 5678 LastLast
Results 61 to 70 of 78

Thread: DEEPCOOL DQ 850-M-V2L & CWT GPX

  1. #61
    Join Date
    May 2020
    Location
    bandung indonesia
    Posts
    17
    Thanks Thanks Given 
    7
    Thanks Thanks Received 
    1
    Thanked in
    1 Post

    Default

    Well, I am no expert either to these stuff so I can't give any extra comment on the very issue. But if it really is the PFC Mosfet that is causing this problem it doesn't make any sense to me. The MOSFETs are actually commonly used in other mid to high end PSUs, an example would be the GPU based DQ-M (v1), so maybe it's something else.
    Last edited by efficient; 07-12-2020 at 06:11 AM.

  2. #62
    Join Date
    Oct 2017
    Location
    Grand Junction, CO
    Posts
    2,964
    Thanks Thanks Given 
    207
    Thanks Thanks Received 
    116
    Thanked in
    105 Posts

    Default

    It's actually the PWM FET that fails (but your point remains valid)


    Quote Originally Posted by Jon Gerow View Post
    PWM. I only mention PFC because usually that goes out in low voltage conditions when they're anemic. I was quite surprised when Q2 blew. Now I have to read this: https://www.onsemi.com/pub/Collateral/AN-9067.pdf.pdf

    But that's what's weird, right? PWM should be working at bus voltage. So why would the PWM FET blow if mains voltage is low???
    Last edited by ashiekh; 07-12-2020 at 11:36 AM.

  3. #63
    Join Date
    Oct 2017
    Location
    Grand Junction, CO
    Posts
    2,964
    Thanks Thanks Given 
    207
    Thanks Thanks Received 
    116
    Thanked in
    105 Posts

    Default

    Quote Originally Posted by Jon Gerow View Post
    But that's what's weird, right? PWM should be working at bus voltage. So why would the PWM FET blow if mains voltage is low??? So weird.
    Recall
    https://digilander.libero.it/hibone/...est_Report.pdf
    where the bus voltage is anything but constant

    Perhaps something of the sort is happening here; regardless (as the Guru points out), a power supply should not self destruct when the voltage is low.

    Quote Originally Posted by Jon Gerow View Post
    Even if spec is 100V-240V, a PSU shouldn't blow up. It should at least turn off.
    Last edited by ashiekh; 07-12-2020 at 12:14 PM.

  4. #64
    Join Date
    Oct 2006
    Location
    jonnyGURU forums, of course!
    Posts
    17,382
    Thanks Thanks Given 
    581
    Thanks Thanks Received 
    667
    Thanked in
    489 Posts

    Default

    Quote Originally Posted by efficient View Post
    Well, I am no expert either to these stuff so I can't give any extra comment on the very issue. But if it really is the PFC Mosfet that is causing this problem it doesn't make any sense to me. The MOSFETs are actually commonly used in other mid to high end PSUs, an example would be the GPU based DQ-M (v1), so maybe it's something else.
    That was already brought up in this thread. But we can go over it again and word it a different way....

    It's not just the quality of the MOSFET, but the capacity. If you're using a cheaper MOSFET that doesn't have any margin for corner cases, you either beef up that part or use a better part. To your example, if you're using said cheap MOSFET in a 650W product, why would I expect to see the same cheap MOSFET in the same quantity in an 850W product? Answer: I shouldn't. So while it's ok in many cases to use a cheaper part, at least make sure the part is the right size for the job. If you're unsure, use a better part.

    Now, the later part might sound like "if you have to use a better part because of better margins, then your design is flawed." Well... a lot of times that's true! When problems crop up, it's nice to be able to figure out the route cause. But that takes time and money. Too often I go down a rabbit hole too deep only to have someone pull me out and say "how much money do you think we have?"

    Example: The DC to DC on the CX750M and CX850M had some MOSFET failures. We switched the current part (can't remember the brand) with an Infineon part of the same value and the failures went away. But if the original part were right for the job in the first place, they shouldn't have failed. So where was the mistake made? Too late... move on.

    Very early on, the VS Series (the current one) had bulk cap failures. We replaced the Teapo bulk cap with an Elite of the same exact specs. The failures went away. Why? There were four or five of us trying to figure that one out and even got the engineers at Teapo involved and we couldn't figure it out. But all we knew was the Elite cap solved the problem so we moved on.

    At the end of the day, if you want less problems you buy better parts. You can, on paper, figure out all the parts you'll need to build your PSU and then just go out and source the cheapest of each part (cheapest caps, cheapest MOSFETs, cheapest diodes, etc.) and it will work and work to >>a<< specification, but will either not last long or will fail when hit with a corner case scenario.

  5. The Following 3 Users Say Thank You to Jon Gerow For This Useful Post:

    efficient (07-13-2020), JCabanes (07-13-2020), Philipus II (07-13-2020)

  6. #65
    Join Date
    Oct 2017
    Location
    Grand Junction, CO
    Posts
    2,964
    Thanks Thanks Given 
    207
    Thanks Thanks Received 
    116
    Thanked in
    105 Posts

    Default

    Which begs the question: what do other supplies do at out-of-spec under voltage? hopefully not self destruct.
    Last edited by ashiekh; 07-12-2020 at 03:09 PM.

  7. #66
    Join Date
    Oct 2006
    Location
    jonnyGURU forums, of course!
    Posts
    17,382
    Thanks Thanks Given 
    581
    Thanks Thanks Received 
    667
    Thanked in
    489 Posts

    Default

    Quote Originally Posted by ashiekh View Post
    Which begs the question: what do other supplies do at out-of-spec under voltage? hopefully not self destruct.
    Apparently, they do.

  8. The Following User Says Thank You to Jon Gerow For This Useful Post:

    ashiekh (07-12-2020)

  9. #67
    Join Date
    Oct 2017
    Location
    Grand Junction, CO
    Posts
    2,964
    Thanks Thanks Given 
    207
    Thanks Thanks Received 
    116
    Thanked in
    105 Posts

    Default

    You are scaring me!

  10. #68
    Join Date
    May 2020
    Location
    bandung indonesia
    Posts
    17
    Thanks Thanks Given 
    7
    Thanks Thanks Received 
    1
    Thanked in
    1 Post

    Default

    So can you exactly say that even the GPU-based Deepcool DQ-M (v1) will blow up as well when subjected to your tests because it has the same cheap components inside?

  11. #69
    Join Date
    Oct 2006
    Location
    jonnyGURU forums, of course!
    Posts
    17,382
    Thanks Thanks Given 
    581
    Thanks Thanks Received 
    667
    Thanked in
    489 Posts

    Default

    Quote Originally Posted by efficient View Post
    So can you exactly say that even the GPU-based Deepcool DQ-M (v1) will blow up as well when subjected to your tests because it has the same cheap components inside?
    I have not seen that any GPU based PSU uses the same components as GPX. Can you show me that? Because that would be very odd because then it wouldn't be a GPU.

  12. #70
    Join Date
    Oct 2006
    Location
    jonnyGURU forums, of course!
    Posts
    17,382
    Thanks Thanks Given 
    581
    Thanks Thanks Received 
    667
    Thanked in
    489 Posts

    Default

    Looking at this: https://www.kitguru.net/components/p...-psu-review/3/

    It seems that the V1 was essentially a cost down GPU that was still half-bridge. Maybe the made a V2 that was full bridge because they ran into too many failures.

Posting Permissions

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