Dear members! I have a blog about these craps and finally i feel the time right to post here too.
The story and possible background is described in details in hungarian. If needed, i will translate all of them. The basics in short:
So many FSP and OEM power supplies built around the CM6800 controller, commonly named as "green line", "Bluestorm II" and "Epsilon" suffering this. Symptoms are clear: PSU failing with low load, or even no load, buzzing noise. These are the least worries, but some of them failing with the PFC coil totally burnt without failing the PFC FETs, or even the FETs are in short. After the significant rise of these unusual problems with FSP PSUs, it's clear that the primary capacitor has some kind of faliure. But this not as straight, as these caps may show exactly good parameters with a multimeter or even an ESR meter! Only a part of them may show no capacity at all, or even very high ESR. So afterall, i decided to take a closer look inside.
Devastating. The anode conductor has totally pulverized, some kind of corrosion is in effect and contact is loose, unstable, sometimes shows spikes have arisen.
Collected data show these commons:
-All Capxon LPs are affected with date 2005-2006, seems that 2006.10 is already OK
-The affected voltage rating is 400V. 450V models seem OK, but sample is too low.
-180uF..330uF values are surely all affected.
-There are some of these replaced as a preventive action, showing nothing unusual in use, but disassembled and clearly affected (that makes the problem even risky)
-No bulging, no vent
With conjunction of some collagues, we believe that a failing manufacturing process may be the primary reason, but the use of active PFC may also helps, as with the use of active PFC, the ESR value of the primary capacitor surely gets into brighter sight.
The blog is here: http://logout.hu/cikk/kapacitasproblemak/teljes.html
We (the prohardver forum) have also contacted FSP about this issue and awaiting answer from them.
The story and possible background is described in details in hungarian. If needed, i will translate all of them. The basics in short:
So many FSP and OEM power supplies built around the CM6800 controller, commonly named as "green line", "Bluestorm II" and "Epsilon" suffering this. Symptoms are clear: PSU failing with low load, or even no load, buzzing noise. These are the least worries, but some of them failing with the PFC coil totally burnt without failing the PFC FETs, or even the FETs are in short. After the significant rise of these unusual problems with FSP PSUs, it's clear that the primary capacitor has some kind of faliure. But this not as straight, as these caps may show exactly good parameters with a multimeter or even an ESR meter! Only a part of them may show no capacity at all, or even very high ESR. So afterall, i decided to take a closer look inside.
Devastating. The anode conductor has totally pulverized, some kind of corrosion is in effect and contact is loose, unstable, sometimes shows spikes have arisen.
Collected data show these commons:
-All Capxon LPs are affected with date 2005-2006, seems that 2006.10 is already OK
-The affected voltage rating is 400V. 450V models seem OK, but sample is too low.
-180uF..330uF values are surely all affected.
-There are some of these replaced as a preventive action, showing nothing unusual in use, but disassembled and clearly affected (that makes the problem even risky)
-No bulging, no vent
With conjunction of some collagues, we believe that a failing manufacturing process may be the primary reason, but the use of active PFC may also helps, as with the use of active PFC, the ESR value of the primary capacitor surely gets into brighter sight.
The blog is here: http://logout.hu/cikk/kapacitasproblemak/teljes.html
We (the prohardver forum) have also contacted FSP about this issue and awaiting answer from them.
Attached Files
Comment