Hi,
This wasn't an easy repair so I thought it would be interesting to share it, also it's not quite common to see this kind of stuff being worked on.
So here we have a Cisco Catalyst 3750 with a no power issue.
The PSU (Phihong PSM200-120(MC)) was obviously not giving any voltage on its 12V line.
After a bit of basic troubleshooting, the issue was found to be on the primary side: the VCC (pin 13) of the SMPS controller (ML4824) was at around 5V when it should have been around 13.5V. As they often go bad together, I checked the switching MOSFETs, and they were partially shorted. I replaced both the controller and the 2 MOSFETs (IRFP460, good habits to replace all of them in one go if you don't want to see fireworks) and the PSU is good to go. 12V is back and PGOOD signal is 4.6V.
But that wasn't enough to resurrect it: the motherboard would still not give any sign of life. Checked the voltage on the coils on the left side, 3.3V and 5V were present, but no 1.5V, 1.8V and 2.5V (thanks Cisco for labelling those). After some probing, I found that 1.5V and 2.5V controller (ISL6558 on the back side) weren't even powered, but the 1.8V had only around 1V on its VCC when it should have been 5V.
Traced the line back to the 0 ohm resistor near the PIC, itself near the power connector. After removing that resistor, one side had 0V (the one going to the ISL), the other had 5.4V. The path to the ISL wasn't shorted, so I injected 5V and the switch turned on. This line sinked in 60mA.
Ok, so there is an issue on the supply side. Sure enough, the second transistor above the PIC had 12V on its collector, 6V on its base, but the emitter was 1V when the resistor was connected, and 5.4V when it wasn't.
After replacing it with whatever NPN I had lying around (C6144, see the standing transistor near the power connector on the picture), the switch is back in working order.
This wasn't an easy repair so I thought it would be interesting to share it, also it's not quite common to see this kind of stuff being worked on.
So here we have a Cisco Catalyst 3750 with a no power issue.
The PSU (Phihong PSM200-120(MC)) was obviously not giving any voltage on its 12V line.
After a bit of basic troubleshooting, the issue was found to be on the primary side: the VCC (pin 13) of the SMPS controller (ML4824) was at around 5V when it should have been around 13.5V. As they often go bad together, I checked the switching MOSFETs, and they were partially shorted. I replaced both the controller and the 2 MOSFETs (IRFP460, good habits to replace all of them in one go if you don't want to see fireworks) and the PSU is good to go. 12V is back and PGOOD signal is 4.6V.
But that wasn't enough to resurrect it: the motherboard would still not give any sign of life. Checked the voltage on the coils on the left side, 3.3V and 5V were present, but no 1.5V, 1.8V and 2.5V (thanks Cisco for labelling those). After some probing, I found that 1.5V and 2.5V controller (ISL6558 on the back side) weren't even powered, but the 1.8V had only around 1V on its VCC when it should have been 5V.
Traced the line back to the 0 ohm resistor near the PIC, itself near the power connector. After removing that resistor, one side had 0V (the one going to the ISL), the other had 5.4V. The path to the ISL wasn't shorted, so I injected 5V and the switch turned on. This line sinked in 60mA.
Ok, so there is an issue on the supply side. Sure enough, the second transistor above the PIC had 12V on its collector, 6V on its base, but the emitter was 1V when the resistor was connected, and 5.4V when it wasn't.
After replacing it with whatever NPN I had lying around (C6144, see the standing transistor near the power connector on the picture), the switch is back in working order.
Attached Files
Comment