Power Supplies Output Fails

[cmckenzie1635]

I have a WS-4506-e that has two 6000w power supplies that both went into Output Fail mode at the same time. The four input power sources are on 4 separate circuits and my power guys have assured me that we did not experience any spikes. The power setting on the switch is set to the default redundant and the load on the switch is far from maxed out.

Here is what is strange. After the failures, when I turn off one of the power input toggle switches on a power supply, doesn't matter which one, it will then begin to come back up. I can turn that input power back on and the power supply continues to work. Both power supplies will work for a period of time then both will fail again. I swapped out just one of the power supplies and reset just to see what would happen. When the original power supply failed this time around, the new one remained up so I guess that its safe to say that my power supplies are fried.

My question; Has anyone seen this behavior before and does anyone have any idea what might have caused this? Thanks.

[Leo Laohoo]

I have never seen two power supplies, from the same chassis, that are fried.  

Get the PS RMA-ed.  If the problem persist, I don't think it's a PS problem.

[cmckenzie1635]

I replaced both power supplies and the problem went away with the original Chassis. I then took one of them to our lab and installed them in a brand new 4506-e chassis along with a same Sup Mod and a few 48 port blades. I didn't put it on the network or put any load on it. After a period of time the Power Supplies failed just as they have been doing in the production switch. It appears that both of these have malfunctioned in the same manner at the same time. I have never seen this before either. I was hoping someone had and had some input on this situation. I was thinking that maybe one failure cascaded to the other but these are behaving so strangely that  I don't know what to think.

[gfield]

Experiencing the same exact problem.

[Arcshaw]

So I had the same problem and found that the supervisor had intentionally put the output side of each power supply into an error-disabled state because it detected a failure in the chassis fan tray. The nature of the fault in the fan tray saw it only failing intermittently too.