Chieftec APS-650C owners beware

Collapse
X
 
  • Time
  • Show
Clear All
new posts
  • mitchw
    Senior Member
    • Dec 2016
    • 136
    • Slovenia

    #1

    Chieftec APS-650C owners beware

    My trusty old Chieftec APS-650C decided to emit a pop and some smell few weeks ago. This happened when I pressed the power on button on the case.

    Upon examination I discovered:
    - No mains fuse blown, still drawing some little current from the mains
    - TNY278PN MOSFET almost shorted (about 90 ohms drain to source)
    - Primary of the 5VSB transformer open
    - 470uF 400V TEAPO mains smoothing capacitor reading only 5nF

    I managed to repair it. First I repaired the blown transformer, the tiny wire coming out of the winding to the pin on the transformer was blown. I was able to solder some new wire to it. Next, I replaced the TNY278, and the last, the 470uF 400V cap.

    I cracked open the 470uF cap and one of the internal connections failed.

    As I read from some foreign forums, this failure on these power supplies is quite common. The 470uF cap goes open, and then usually the TNY278 fails short which either blows the primary of the standby transformer or it blows the 0.22ohm resistor next to it. Sometimes it can also blow more stuff.

    So if you are an owner of a PSU like this, and it still works, you better replace the 470uF 400V cap before it fails.
    Attached Files
  • stj
    Great Sage 齊天大聖
    • Dec 2009
    • 31044
    • Albion

    #2
    Re: Chieftec APS-650C owners beware

    the cap was probably destroyed by the APFC circuit - that's starting to become a pattern in all modern psu's!
    it's only there to please the power company to make up for deficiencies in their networks!

    Comment

    • mitchw
      Senior Member
      • Dec 2016
      • 136
      • Slovenia

      #3
      Re: Chieftec APS-650C owners beware

      I am not sure why it failed. Either the APFC switching action combined with slowly increasing cap ESR due to age. Or it failed just from old age, as there is a trace eaten away. The cap wasn't bulging. I had old 1960s caps from vacuum tube era fail the same way. They were measuring okay in the early 00s, but 15 years later they were O/L, despite sitting not used the whole time.

      Testing the capacitor directly revealed it still had decent capacitance, but ESR was a bit high - could be that my leads were not touching properly and this tester is not to be trusted 100% really.
      Attached Files

      Comment

      • ChaosLegionnaire
        HC Overclocker
        • Jul 2012
        • 3264
        • Singapore

        #4
        Re: Chieftec APS-650C owners beware

        the use of a 400v cap is the issue on apfc circuits. 400v is too close to the working voltage of the apfc circuit of 390v. sometimes due to spikes or brownouts on the power grid, it can even overshoot the 400v rating of the primary cap, causing it to fail.

        to prevent this from happening again, uprate the voltage of the primary cap to 450v or 420v if there are space constraints. this should prevent it from happening again.

        teapo is considered a junk cap brand here anyways, so im not surprised it failed, much less last as long as it did.
        Last edited by ChaosLegionnaire; 04-14-2021, 06:12 PM.

        Comment

        • mitchw
          Senior Member
          • Dec 2016
          • 136
          • Slovenia

          #5
          Re: Chieftec APS-650C owners beware

          I temporarily replaced it with a lightly used Hitachi HU series, which has the same ratings.

          It lasted almost 11 years (from June 2010 to Mar 2021), so this isn't that bad for a cap running at almost 100% of its voltage rating

          Comment

          • PeteS in CA
            Badcaps Legend
            • Aug 2005
            • 3581
            • USA, Unsure of Planet

            #6
            Re: Chieftec APS-650C owners beware

            Originally posted by ChaosLegionnaire
            the use of a 400v cap is the issue on apfc circuits. 400v is too close to the working voltage of the apfc circuit of 390v. sometimes due to spikes or brownouts on the power grid, it can even overshoot the 400v rating of the primary cap, causing it to fail.

            to prevent this from happening again, uprate the voltage of the primary cap to 450v or 420v if there are space constraints. this should prevent it from happening again.

            teapo is considered a junk cap brand here anyways, so im not surprised it failed, much less last as long as it did.
            Does your experience suggest that the reliability issue is with running the caps close to their rated voltage? Could it be that these types of parts were originally developed for 50/60 Hz usage and handle higher switch frequency ripple much less well? Obviously the answer could be "Both".
            PeteS in CA

            Power Supplies should be boring: No loud noises, no bright flashes, and no bad smells.
            ****************************
            To kill personal responsibility, initiative or success, punish it by taxing it. To encourage irresponsibility, improvidence, dependence and failure, reward it by subsidizing it.
            ****************************

            Comment

            • momaka
              master hoarder
              • May 2008
              • 12175
              • Bulgaria

              #7
              Re: Chieftec APS-650C owners beware

              Originally posted by stj
              the cap was probably destroyed by the APFC circuit - that's starting to become a pattern in all modern psu's!
              The first time I saw that failure, I knew those APFC circuits are up to no good. Surely some last longer than others, especially the ones build with plenty of caps (or just big enough caps) to handle the higher ripple current. But in the end, I don't think we will see any APFC circuit last as long as the oldschool method of caps straight-up filtering 50/60 Hz mains.

              Originally posted by mitchw
              It lasted almost 11 years (from June 2010 to Mar 2021), so this isn't that bad for a cap running at almost 100% of its voltage rating
              True.
              But the problem is, I think manufacturers/designers of these PSUs have now well understood exactly how long those parts are going to last. And to me at least, this just seems like another form of planned obsolescence. Yes, 10 years is a pretty good run for a PSU... but the fact that I have 20-year-old Deer PSUs with completely junk brands on their primaries, like JEE, CapXon, Viva, and etc, and none of those yet failed... that should tell you something about how unreliable APFC circuits have made things.

              Originally posted by mitchw
              I had old 1960s caps from vacuum tube era fail the same way. They were measuring okay in the early 00s, but 15 years later they were O/L, despite sitting not used the whole time.
              Caps not sitting used at all is about the 2nd worst thing to do to them after caps being abused with too much heat, ripple current, high voltage (or a combo of all three.) Electrolytic capacitors will last the longest when used on a stable voltage supply (lower ripple) that's not too far below their rating (2/3 of their rated voltage tends to be optimal) and in a device where temperature swings are fairly controlled (temperature not too high or too low.) The further you move from these, the less the cap will last - especially with the last one: that is, e-caps certainly don't do well with high heat.

              Originally posted by mitchw
              ITesting the capacitor directly revealed it still had decent capacitance, but ESR was a bit high - could be that my leads were not touching properly and this tester is not to be trusted 100% really.
              Perhaps there was flux on the leads? Or perhaps one of the leads inside the cap got snapped due to chemical reactions with the flux or improper assembly/design at the factory. I suspect the cap did go open-circuit, though, because otherwise the TNYswitch likely wouldn't/shouldn't have blown on its own.

              Originally posted by PeteS in CA
              Obviously the answer could be "Both".
              Indeed.
              I think the answer is that both the high running voltage and high ripple current are contributing factors to the failures.
              Last edited by momaka; 04-15-2021, 11:15 PM.

              Comment

              Related Topics

              Collapse

              • harp
                Learning with ATX CHIEFTEC 450W PSU
                by harp
                I have a little question regardless a way to repair ATX power supply CHIEFTEC MODEL: GPS-450AA-101 A. It is 450w power for computer.

                The psu has initial short on +5VSB (purple), and I found and replace shorted diode SB340. After replace, unpowered measurement of all outputs seems fine. All other big double diode on heatsink are ok in secondary side. I connect psu with 60w bulb in serial, and bulb shortley flash as it would be. There is steady 5.2v on +5VSB and PS-ON, on any other is nothing, like 0.00v. So the primary is ok, right?

                But, when I PS-ON (green) connect, bulb...
                06-17-2022, 05:04 PM
              • pingaroon
                Question for Dell Latitude 7420 owners
                by pingaroon
                Hi,
                Would somebody be able to confirm if this model will power on normally with the battery disconnected. I have a buyer who tells me this is normal behaviour and the board is designed to only switch on with both charger and battery attached. Since I've not seen that in a Dell board (except with Chromebooks), I'm skeptical.

                Thanks!
                01-10-2023, 04:06 PM
              • mo000
                470uf tantalum caps replacement
                by mo000
                hi
                motherboard toshiba C50 PT10F

                i found 2 bad 470uf tantalum in cpu sircuit making short .

                but i only have 380uf tantalum can i replace them with electrolytic caps .
                05-11-2021, 10:44 AM
              • elzorroazul777
                Owners of X1 Extreme Gen 2 w/ i9 Processor I Need BIOS Dump
                by elzorroazul777
                My X1EG2 has been bricked due to failed firmware update. I have searched all over the internet looking for a BIOS dump and cannot find it for the life of me. I plead to the old gods and new; is there anyone with the X1 Extreme Gen 2 w/ i9 Processor willing to create a BIOS dump and share it with me? A tool can be found here: https://www.bios-mods.com/forum/Thre...Megatrends-Inc and I found that tool by searching reddit.

                Thank you!
                10-10-2022, 11:44 AM
              • zunasthegreat
                Chieftec GPS-550AB-A (Delta OEM)
                by zunasthegreat
                Hello again and I really hope somebody can give a heads up so I can find the issue.
                This PSU has busted Ltec caps, the 12V and 5VSB ones are sooo bulged and the reading is 28pF... The 5V + 3.3V ones show way above rated capacity and I guess they are approaching their death.

                Stock CAPS
                12V -> Ltec LTG 2200uf 16V X2 -> Nichicon HV 1500 16V + HD 2200 16V
                5V -> Ltec LZG 3300uf 10V -> Nichicon HD 3300uf 10V
                3.3V -> Ltec LZG 3300uf 6.3V X2 -> NCC KY 2200uf 10V X2
                5VSB -> Ltec LTG 1000uf 10V + 1500uf 10V -> NCC KY 1000uf 16V X2
                ...
                05-09-2021, 10:28 AM
              • Loading...
              • No more items.
              Working...