what to do next?

AHAA, looks like your suggestion found something: CPU Core (1.3-1.4V) measured 1.137 V DDR Core (2.5V) measured 1.137 V DDR Termination (1.25V) measured 1.138 V Chipset Core (could be 1.2, 1.5, 2.5 or other voltage) measured

4.980 V

have no schematic, i would welcome suggestions.

Reply to
nucleus
Loading thread data ...

Different CPUs require different voltages. The CPU tells those voltage regulators what voltage it needs. Consult numeric spec sheets for your processor to know what voltages should be for that particular processor.

If a CPU volts should be between 1.2 and 2.5 AND if the regulator is outputting 5 volts, then CPU damage would exist. Better is to confirm your voltage readings are correct. If that voltage is too high, then is it the regulator or CPU that is wrong. For example, a CPU could be telling that regulator to output a higher voltage. Best is to first confirm all those numbers from the CPU datasheet; confirm using a primary (reliable) source.

Reply to
w_tom

Yes, it found that I gave you some flawed advice. Read on...

What I forgot is that the CPU Core supply can be multiphase (wrt the PWM controller), so all three FET/inductor/cap sets below the CPU are supplying the same rail. That's why they're the same voltage.

But is 1.137V enough for the CPU core? You'd have to check the AMD spec to be sure, but some Athlons have a start up core voltage (different than the nominal core voltage range specified above) as low as 1.1V, so it might be OK.

What about the DDR voltages? Well, they're obviously generated somewhere else. It's hard to tell from the low res images available online, but it appears that there are a few supplies to the right of the DIMM sockets. Check the caps in that area for DDR Core and Termination voltages, or on the DIMM socket itself (see e.g.

formatting link
for pinout; VREF should be 1.25V).

I think you're measuring the input voltage to that supply. I couldn't find a spec for the Nvidia C51G core voltage (which is why I hedged above), but it's unlikely to be 5V. Check some other caps in that area, particularly ones that are connected (positive terminal, of course) to an inductor nearby.

TM

Reply to
tonym924

e.g.http://www.technick.net/public/code/cp_dpage.php?aiocp_dp=pinconmem_d...

i removed the heat sinks from north bridge and south bridge. south bridge looked normal but north bridge had at some time gotten hot enough to melt the rectangular white (perhaps conductive) material, and it had flowed down the base.

formatting link
shows the bottom of the heatsink.

formatting link
shows what i interpret as the north bridge chip, after cleaning most of the melted material that had flowed down the base.

all caps in the vicinity of north bridge are not shorted, as measured with power removed.

is there a possibility that the north bridge is still functional? or toast?

Reply to
nucleus

That's just heatsink compound, why did you remove the heatsinks from those chips? It's very easy to damage the exposed die.

Reply to
James Sweet

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.