Cisco Switch No Power

[Bluestreak66]

I have a Cisco 2960 that appears to have went for a swim before it got to me. The power supply was completely blown and looks like it had sat with water in it for a while powered on. I cleaned up the circuit board and checked it for physical damage but there is nothing visibly wrong with it. I pulled a power supply out of another 2960 I have and put power to this one, the fan comes on but that's it.

The power supply outputs 12v to the board which then goes into (from what I can tell) 4 DC to DC converters that output 1.5v, 2.5v, 3.3v, and 1.2v. There are test points on the board for all the voltages but no voltage to any except the 12v coming into the board. The dc to dc converter seem fairly straight forward two ISL6534 Synchronous-Rectified Buck Controllers which drive 8 N-Channel MOSFETs, 3 IRF7828 and 5 PSMN005-30K. Is there a reason all 8 are not the same?

I'm really interested in how this circuit design works and if someone could give and explanation of whats happening here, particularly in the circuit at the fan. There are two power tracks, one leading to the diodes labeled CR34 and CR37 which the go to surfacemont diodes CR35 and CR38, after that ther are what looks like 2 transistors and another diode across the two tracks (voltage clamping?). This seems quite complicated to power a fan?
 Here is a picture of that section:


I'm not really going for fully functional here if I could get dc to dc converter section powered up and blinking lights I'd be happy, but if It powered up to a console that would be even better. Any thoughts of where to start are appreciated. I have a reflow station and can replace SMD chips, but wanted to get some thoughts before just replacing chips. I pulled off the power connector when cleaning the board and it broke so I just soldered wires on for tesing. The four white wires are +12v and four black wires for ground. The last four appear to be some kind of sensing wires as two of them go to resistors and two go to capacitors with the components the grounded.

[HalfSpace]

Hi Bluestreak66,

Here is a link to the data sheet for the ISL6534 http://www.intersil.com/content/dam/Intersil/documents/isl6/isl6534.pdf Check out the typical application circuits for this device, it will give some idea as to what signals and voltages that should be present on the pins of the ISL6534. Check that the supply voltages (VCC) are present and that the enable lines are enabled.

HalfSpace 

[Bluestreak66]

Thanks for the reply I had pulled up the data sheets for all the chips on the board. Also checked for power at on the chips and there is indeed +5v on both chips. Is there anything that would stop these from operating? It seems there needs to be 12v applied to boot 1 and boot 2 through a diode as well, is this correct?

The intersil mentioned above:
http://www.intersil.com/content/dam/Intersil/documents/isl6/isl6534.pdf
International Rectifier MOSFET:
http://www.irf.com/product-info/datasheets/data/irf7828.pdf
NXP MOSFET:
http://www.nxp.com/documents/data_sheet/PSMN005-30K.pdf
There is also a Quad Comparator as well:
https://www.fairchildsemi.com/datasheets/LM/LM2901.pdf
If I had to guess it probably has somthing to do with the fan?

[justanothercanuck]

I would start by re-soldering everything.  The pads on Q5 and Q6 look a little dirty, but overall, there could be issues with part connectivity, especially if it's been sitting in water (or leaky capacitors / batteries).

[bktemp]

The dc to dc converter seem fairly straight forward two ISL6534 Synchronous-Rectified Buck Controllers which drive 8 N-Channel MOSFETs, 3 IRF7828 and 5 PSMN005-30K. Is there a reason all 8 are not the same?

The PSMN005 have a lower RDSon. These are used for the synchronous rectifier. Since the duty cycle is quite low (e.g. 10% for 1.2V), the synchronous rectifier mosfets are switched on for a much longer period than the highside mosfets. Therefore their RDSon is more important.

[HalfSpace]

Hi Bluestreak66,

The IC needs +5V for the VCC pin and +12V for the VCC12 pin and +12V for the diodes connected to the boot pins. If these supply rails are not correct the IC’s internal reset circuit will hold the three enable pins EN1, EN2 and EN3 low which will disable all three regulators in the IC. The enable pins should be above 1V for normal operation.

HalfSpace