Announcement

Collapse
No announcement yet.

My PCP&C Silencer 750 seems to be a bit dodgy

Collapse
X
Collapse
 
  • Page of 1
  • Filter
  • Time
  • Show
Clear All
new posts
Previous template Next
    #1

    My PCP&C Silencer 750 seems to be a bit dodgy

    I've got a comp with an issue....with a tiny GPU, say an 8600GT, it will OC to 4.2GHz. With the 560Ti or a 460, it WILL NOT OC beyond 3.5GHz.

    Now in theory I have enough power from this old Silencer 750 and to rule out caps in this older unit, some new Panasonic FMs were fitted. 3.5GHz became possible after I did this....before 3.4 was all it would do without random crap. (the spontaneous reboot followed by a messed up IOH that found no drives was the most annoying....a hard reset was required to fix it.)

    The scope traces aren't impressive. I'm unsure if its PSU generated noise as I dont have a load tester. Ripple is 170mV....out of spec.

    So is the PSU broken or is it borderline overloaded? The PSU calc says it's not overloaded at 630 watts with the most extreme OC possible with my current cooling.....I should have some headroom here with this unit right?

    Stuff in the rig:
    Gigabyte GA-X58A-UD5
    Core i7 950 (D0 revision)
    2GB x6 OCZ Platinum DDR3-1600 (12GB total, 2T rate)
    4 WD Caviar Black HDDs
    GTX 560 Ti GPU
    SB X-Fi
    ATI Theater 650 TV tuner
    Linksys 802.11g card
    BD-RE drive
    3 big ass fans (200mm?)
    3 120mm fans
    6 USB peripherals in need of power

    In any case it looks like a bigger PSU is in the cards....so what's it gonna be, the Seasonic X850 / Corsair AX850 (same units) or the Corsair AX 1200 (overkill but Flextronics made...looks nice).

    Hopefully I get this sorted before Quakecon....I don't want to be the guy with the slowest computer. (ie, the only un OCd rig in the group)
    Tags: None
    #2
    Re: My PCP&C Silencer 750 seems to be a bit dodgy

    did you do a full or partial recap? if the former, yeah, get a better PSU. if the latter, complete the recap.
    sigpic

    (Insert witty quote here)

    Comment

      #3
      Re: My PCP&C Silencer 750 seems to be a bit dodgy

      Sounds to me like you ignored the ESR specs of the original caps and slapped FM's in there.
      -
      Your ripple is high because your filters are no longer optimized for the frequency of the ripple.
      -
      Mann-Made Global Warming.
      - We should be more concerned about the Intellectual Climate.
      -
      Be who you are and say what you feel, because those who mind don't matter and those who matter don't mind.
      - Dr Seuss
      -
      You can teach a man to fish and feed him for life, but if he can't handle sushi you must also teach him to cook.
      -

      Comment

        #4
        Re: My PCP&C Silencer 750 seems to be a bit dodgy

        ESR specs of the original caps weren't any better....they were UCC KZE. Looking back at the invoices, I used FR, which is a replacement for FM. ESR and ripple ratings are similar, with the Panasonic taking a slight lead on both specs with *much* higher rated life.

        Performance *improved* after the recap, just not by much. I was expecting a repair, what I got was a band aid.

        It was a full recap of the post output rectifier section. This PSU has seen nearly 24/7 use for over 3 years in sometimes less than optimal conditions (hot room) but I was hoping to get a bit more out of it.

        Comment

          #5
          Re: My PCP&C Silencer 750 seems to be a bit dodgy

          I wouldn't call shooting ripple up to 170 mV an improvement of performance or anything else.

          You repeated the same mistake Oklahoma Wolf did on the jonnyguru site a few years ago.

          FR/FM and KZE are significantly different grades.
          Unless you reduced the can size going from KZE to FR they aren't that close in specs.
          Considering the FM and FR tend to be large for their uF [and KZE isn't] that is very unlikely.

          The OP caps in PSUs are part of filters that also include inductors.
          The values of the caps work with the values of the inductors to optimize the filter to remove Ripple for some frequency which is the expected Ripple frequency of that PSU design.

          Your Ripple not only went up, it went completely out of spec, which means you shifted the filters operating curve and it's no longer as effective as it was at the Ripple frequency of concern in that PSU.
          .
          Attached Files
          Last edited by PCBONEZ; 08-02-2011, 10:28 AM.
          Mann-Made Global Warming.
          - We should be more concerned about the Intellectual Climate.
          -
          Be who you are and say what you feel, because those who mind don't matter and those who matter don't mind.
          - Dr Seuss
          -
          You can teach a man to fish and feed him for life, but if he can't handle sushi you must also teach him to cook.
          -

          Comment

            #6
            Re: My PCP&C Silencer 750 seems to be a bit dodgy

            If it's 170mV with the computer as a load you are picking up interference from some other device inside the computer. Unless the load is fully resistive you cannot accurately determine if the noise is coming from the PSU itself or from somewhere else.
            Originally posted by PeteS in CA
            Remember that by the time consequences of a short-sighted decision are experienced, the idiot who made the bad decision may have already been promoted or moved on to a better job at another company.
            A working TV? How boring!

            Comment

              #7
              Re: My PCP&C Silencer 750 seems to be a bit dodgy

              Along that line, how many of those "big ass fans" [120mm or 200mm] connected straight to +12v vice being connected to mobo headers that have EMI filters for motor noise?
              .
              A PSU can normally filter for a few of them directly on +12v but with a slew of them on +12v you may be having issues with motor noise feeding back and bouncing around in you PSU circuits.
              .
              .
              At any rate, your problem is not too few watts on the PSU.
              A 750 watt is probably already over-kill.
              .
              Mann-Made Global Warming.
              - We should be more concerned about the Intellectual Climate.
              -
              Be who you are and say what you feel, because those who mind don't matter and those who matter don't mind.
              - Dr Seuss
              -
              You can teach a man to fish and feed him for life, but if he can't handle sushi you must also teach him to cook.
              -

              Comment

                #8
                Re: My PCP&C Silencer 750 seems to be a bit dodgy

                Only the CPU fan is plugged into a header. The rest are tied onto a separate line from the PSU. I guess it's possible Cooler Master provided crap fans but I kinda doubt it.....this is the ATCS 840 we're talking about, not exactly an "entry level" chassis!

                I'm unsure of why this much ripple is on the line.....it's very unusual. It just started acting funny one day. The configuration worked well when it was first assembled.

                750 should be overkill, I agree. A 600 should run this thing with everything being under load in theory. Idle only lights up one dot on the Smart UPS 1500....and that suggests less than 250W consumption.

                Comment

                  #9
                  Re: My PCP&C Silencer 750 seems to be a bit dodgy

                  You misunderstood me....ripple went DOWN to 170mV.....it was hitting 220mV before hand. Needless to say, my machine was not amused by this.

                  I saw the data sheet for both, in the sizes I speced out, the parameters are very similar. In the smaller can sizes, things look different. This PSU uses large can caps.

                  One of the more common caps in this unit is a 10V 4700uF unit which does filtering for the 3.3 and 5V rail, here are the specs:

                  UCC KZE: Size: 12.5x35mm, Ripple Current: 3400mA, ESR 0.015 ohms
                  Panasonic FR: Size: 12.5x30mm, Ripple Current: 3630mA, ESR: 0.013 ohms

                  For the 3300uF 16V size which filters the 12V rail:
                  UCC KZE: 12.5x35mm, 3400mA, 0.015 ohms
                  Panasonic FR: 12.5x30mm, 3630mA, 0.013 ohms

                  I doubt its these new caps....in the case of these particular units, they are *very* similar. Caps tend to be loose tolerance devices (+/- 20% is typical) and at least on paper, I'm within those specs. The only thing that's vastly different is rated life, with the FR being 10k hours and the KZE being 5k.

                  I may get the scope out and see if those fans are causing the issue....it would be a first but I've seen weird stuff this week. (Don't even get me started on the BIOS that ships with the new Gigabyte Z68 boards.....5 revisions old and buggy as hell!)

                  Comment

                    #10
                    Re: My PCP&C Silencer 750 seems to be a bit dodgy

                    Cooler Master does provide crap fans.
                    Remove the side, unplug them, and see what happens to the ripple.
                    [Read the rest later. I'm cooking dinner....]
                    Mann-Made Global Warming.
                    - We should be more concerned about the Intellectual Climate.
                    -
                    Be who you are and say what you feel, because those who mind don't matter and those who matter don't mind.
                    - Dr Seuss
                    -
                    You can teach a man to fish and feed him for life, but if he can't handle sushi you must also teach him to cook.
                    -

                    Comment

                      #11
                      Re: My PCP&C Silencer 750 seems to be a bit dodgy

                      Yes those units are very well behaved typically you are most likely getting interference. You are suppose to put a .1uf ceramic and 10uf lytic or something like that across the probe's when measuring on a ATX power supply.

                      Comment

                        #12
                        Re: My PCP&C Silencer 750 seems to be a bit dodgy

                        Is the UPS a stepped square-wave UPS or a real sine-wave UPS?
                        .
                        Mann-Made Global Warming.
                        - We should be more concerned about the Intellectual Climate.
                        -
                        Be who you are and say what you feel, because those who mind don't matter and those who matter don't mind.
                        - Dr Seuss
                        -
                        You can teach a man to fish and feed him for life, but if he can't handle sushi you must also teach him to cook.
                        -

                        Comment

                          #13
                          Re: My PCP&C Silencer 750 seems to be a bit dodgy

                          Originally posted by PCBONEZ View Post
                          Is the UPS a stepped square-wave UPS or a real sine-wave UPS?
                          .
                          Sine. It's an APS Smart UPS 1500....excellent units.

                          I'll perform additional tests later....time to RMA some hard disks that are DOA. (65677 bad sectors on one!)

                          Comment

                            #14
                            Re: My PCP&C Silencer 750 seems to be a bit dodgy

                            Originally posted by 370forlife View Post
                            Yes those units are very well behaved typically you are most likely getting interference. You are suppose to put a .1uf ceramic and 10uf lytic or something like that across the probe's when measuring on a ATX power supply.
                            Now that I was unaware of....let me retest using those.

                            Comment

                              #15
                              Re: My PCP&C Silencer 750 seems to be a bit dodgy

                              Is that ripple you posted at the PSU +12v connectors or at Vcore?
                              .
                              Mann-Made Global Warming.
                              - We should be more concerned about the Intellectual Climate.
                              -
                              Be who you are and say what you feel, because those who mind don't matter and those who matter don't mind.
                              - Dr Seuss
                              -
                              You can teach a man to fish and feed him for life, but if he can't handle sushi you must also teach him to cook.
                              -

                              Comment

                                #16
                                Re: My PCP&C Silencer 750 seems to be a bit dodgy

                                How that cap thing is supposed to work:
                                -
                                It is to isolate scope trace to PSU ripple.
                                Use both a .1uf ceramic and a 10uf lytic in parallel with the load.
                                That filters out higher frequency noise [EMI] and the scope shows you what noise the PSU switching is making.
                                -
                                So, if your ripple voltage goes way down with the caps there you know you are having EMI noise problems instead of PSU switch-noise problems.
                                If it doesn't go down then you are having PSU switch-noise problems.
                                .
                                When you are connected to a mobo [vice a test load] the EMI should already be taken care of by small SMD caps or [like] 10uF lytics on the board and those two caps shouldn't really be needed.
                                - If it's not getting taken care of it could be a bad mobo design /or you might have some blown small uF lytic or SMD caps on the mobo /or you have a cubic butt-ton of EMI coming from somewhere.
                                .
                                Last edited by PCBONEZ; 08-03-2011, 07:22 AM.
                                Mann-Made Global Warming.
                                - We should be more concerned about the Intellectual Climate.
                                -
                                Be who you are and say what you feel, because those who mind don't matter and those who matter don't mind.
                                - Dr Seuss
                                -
                                You can teach a man to fish and feed him for life, but if he can't handle sushi you must also teach him to cook.
                                -

                                Comment

                                Previous template Next
                                Working...
                                X