Page 1 of 2 12 LastLast
Results 1 to 10 of 20

Thread: Question about SF750 hold up time and using a UPS

  1. #1
    Join Date
    May 2020
    Location
    Italy
    Posts
    21
    Thanks Thanks Given 
    13
    Thanks Thanks Received 
    0
    Thanked in
    0 Posts

    Default Question about SF750 hold up time and using a UPS

    Thank you. Interesting review.

    I have deepen the study about the most important Cons that the review expresses: Low hold-up time

    In the review they stand this:
    "This platform is made by Great Wall and, as you can see in the table above, it uses high-quality components. However, it would be even better if a higher-capacity bulk cap was installed in the APFC converter, allowing for more than 17ms hold-up time."

    According to this article (https://blog.fsp-group.com/en/the-im...-hold-up-time/), this is what they say regarding hold-up time:
    "As set by Intel, the holding time must be greater than 16ms. First, we must understand the AC mains power in the world has two AC input frequency specifications, which are 60Hz, primarily used in the US, and 50Hz, used in EU. If a AC sine-wave is calculated with these two frequencies, the period would be 16ms and 20ms respectively."
    The article also explains why the use of 16ms and 20ms, but lets focus on the number of the review.

    I immagine that the review has been made in the US, where 110V is the main power source. In that case they should have been in the case of considering the reference number of 16ms.
    As the SF750 was identified in having a 17ms hold-up time, when in the review they say that the number is "Low", is it because of the 1ms higher value than the minimum one acceptable?

    For me that I live in Europe and have 230V as main power, I imagine that I'll probably rather reach 21ms or maybe 22ms. In such situation (if it will be like that), will the hold-up time be considered too low as well, even if there will probably be a bigger gap between the psu hold-up time at 230V and the minimum value per specification?


    Said so...is this all bad? I have to mention that I have my full computer behind UPS (read the article up hear for understand the importance of this detail when speaking about hold-up time)

    Thank you all.

  2. #2
    Join Date
    Oct 2006
    Location
    jonnyGURU forums, of course!
    Posts
    16,837
    Thanks Thanks Given 
    568
    Thanks Thanks Received 
    499
    Thanked in
    371 Posts

    Default

    So, your post is quite long. But I think I can summarize with this:

    An SFX power supply will have limited hold up time due to limited size.

    Hold up time is determined at low mains voltage, which means at higher mains voltage, the hold up time will be longer.

    And finally, if you have the PC on a UPS, hold up time makes no difference. The switching from mains to battery takes less time then any PSU's hold up time.

  3. #3
    Join Date
    May 2020
    Location
    Italy
    Posts
    21
    Thanks Thanks Given 
    13
    Thanks Thanks Received 
    0
    Thanked in
    0 Posts

    Default

    Quote Originally Posted by Jon Gerow View Post
    So, your post is quite long. But I think I can summarize with this:

    An SFX power supply will have limited hold up time due to limited size.

    Hold up time is determined at low mains voltage, which means at higher mains voltage, the hold up time will be longer.

    And finally, if you have the PC on a UPS, hold up time makes no difference. The switching from mains to battery takes less time then any PSU's hold up time.
    Sorry for the lengths, but I preferred to be precise for better understanding.

    Yes, the UPS is quicker, so the PSU hold-up time should just be greater just to cover that switching time.

    But I was indeed mentioning this because the time stated by the review is quite just a bit more than the standard transition time for UPS, so I was wondering if with this psu I might be in a case of time limit for some scenarios where it might not be enough.

    And yes, in my case I'll have higher mains voltage where the hold-up time will be longer...and was asking confirmation if I should consider my self OK or not if using an UPS. I thought I was still near the time limit so to not have a safe scenario.

    From your reply, I imagine that I could consider my self OK.


    So in the review they mark the hold-up time as a bad thing only because (as it is determined at low mains voltage) at low mains voltage it is nominally at time limit, and of course it doesn't take in consideration any use of UPS (that can help in having more margin of time).


    Thank you very much.

  4. #4
    Join Date
    Oct 2006
    Location
    jonnyGURU forums, of course!
    Posts
    16,837
    Thanks Thanks Given 
    568
    Thanks Thanks Received 
    499
    Thanked in
    371 Posts

    Default

    UPS's have a transfer time of around 8ms.

    Most PSU's won't have an issue with that.

  5. #5
    Join Date
    Mar 2015
    Posts
    297
    Thanks Thanks Given 
    91
    Thanks Thanks Received 
    7
    Thanked in
    6 Posts

    Default

    The hold up time measured at full load, when the load decrease the hold up time increase, IMO its not a deal breaker, in bad line its not going to help you anyway, in good line the psu will not use it, there are cases that the main will stop for that period, its not that common, in most designs increasing the hold up time will increase the start up load.

  6. #6
    Join Date
    May 2020
    Location
    Italy
    Posts
    21
    Thanks Thanks Given 
    13
    Thanks Thanks Received 
    0
    Thanked in
    0 Posts

    Default

    Quote Originally Posted by Jon Gerow View Post
    UPS's have a transfer time of around 8ms.

    Most PSU's won't have an issue with that.
    Ah! I though more, as stated in the article I've linked:

    "The transition time of UPSs of different levels is different. For the lowest on line level UPS, this specification indicator is set to not exceed 20ms, which is the period of an AN sine-wave in a 50Hz region. 16ms, if in a 60Hz region."

    That is the main reason of my questioning..

    Thank you for the clarification.

  7. #7
    Join Date
    Oct 2017
    Location
    Grand Junction, CO
    Posts
    2,559
    Thanks Thanks Given 
    173
    Thanks Thanks Received 
    93
    Thanked in
    84 Posts

    Default

    Quote Originally Posted by PIRATA! View Post
    And yes, in my case I'll have higher mains voltage where the hold-up time will be longer...

    Not sure it works like that as the bulk capacitors will have the same voltage no matter 110 or 220V

  8. #8
    Join Date
    Oct 2006
    Location
    jonnyGURU forums, of course!
    Posts
    16,837
    Thanks Thanks Given 
    568
    Thanks Thanks Received 
    499
    Thanked in
    371 Posts

    Default

    Quote Originally Posted by ashiekh View Post
    Not sure it works like that as the bulk capacitors will have the same voltage no matter 110 or 220V
    I was keeping it simple. The bulk cap voltage doesn't change, but a cycle on most (all?) high line mains is 50 Hz, while it's only 60 Hz in most (all?) low line regions.

    I also don't think FSP is saying that UPSs tend to have 20ms switch over times. They're saying that the switch over time CAN BE up to 20ms; which is equal to 50 Hz.

    A UPS with a 20ms switch over time would be complete garbage. Almost useless.

    And also, none77 is correct. Hold up time is measured when the PSU is under 80% to 100% load, which causes a bulk cap without mains power to drain faster. If you're only at 10%, 20% or 50% of the PSU's capacity, the bulk cap will drain much slower.

  9. #9
    Join Date
    May 2020
    Location
    Italy
    Posts
    21
    Thanks Thanks Given 
    13
    Thanks Thanks Received 
    0
    Thanked in
    0 Posts

    Default

    Well, I shall try to unplug the power to be sure then

    I assume that FSP is saying what the specs documentations says about switching time in UPS speaking about possibilities. Is like speaking about ATX volts specs and saying that the hardware shall tollerate any min and max values of the ATX specs documentation.

    If we follow the specs and the documentation, in the worst scenario the PSU should handle the 16ms for the lower mains voltage / 60Hz and 20ms for higher mains voltage / 50Hz.

    But I read from you guys that this is not a problem as for the reasons above, specially if using a good UPS.

    If not using any UPS, I immagine that if the system loads the PSU always between the 80% and 100%, it might be the case of having the time at limit. But in that case I think the culprit is more having chosen a PSU low in W for that hardware.

  10. #10
    Join Date
    Oct 2017
    Location
    Grand Junction, CO
    Posts
    2,559
    Thanks Thanks Given 
    173
    Thanks Thanks Received 
    93
    Thanked in
    84 Posts

    Default

    Quote Originally Posted by PIRATA! View Post
    But I read from you guys that this is not a problem as for the reasons above, specially if using a good UPS.

    Did you have a sine-wave UPS in mind? there are also zero switch over time (on-line) UPS units.

    For me the problem is the recurring cost of batteries every few years; one thing I tried was to run a UPS on super capacitors
    https://www.instructables.com/id/Con...er-Capacitors/

    As well as an attempt to extend the life of lead acid batteries
    https://www.instructables.com/id/Rev...cid-Batteries/

    One might consider using lithium batteries instead of lead acid

    In the end I just invested in a lot of premium surge protectors.
    Attached Images Attached Images
    Last edited by ashiekh; 05-17-2020 at 10:07 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
  •