Undervolting HARDWARE MOD 1810tz SU4100

Hey guys, I am looking to hardware mod to undervolt my 1810TZ. I found this image for the SU3500 processor.

http://i36.tinypic.com/350mrk7.jpg

I was wondering if this would work for me as well. I have VIDs of 0.900V and 1.013V for my CPU.

Based on the documentation, it would appear that it would drop my 0.900V down to 0.850V but it would have no effect on the 1.013V VID right?

http://www.intersil.com/data/fn/fn9251.pdf



EDIT
It appears that if I change VID3 from 0 -> 1 it would change my 1.013V VID to 0.913V and the 0.900V to 0.800V. This seems like a better choice for me and of course I would have to test for stability but could someone please show me how to change VID3 from 0 -> 1?

 

Can anyone show me where the other traces for the intersil chip leads to? I am mainly looking for the VID3 trace. Thanks.

It appears that the pad for PR80 in that picture with the red connection will pull whatever I need to the on state of 1. So would I just measure all the pads near it and check for 0 resistance to pin 31 on the ISL6261 chip. Or can I just solder a wire straight from pin 31 to PR80?

And also do you think the chip will run with 0.800V at 1.2Ghz? ( I know every chip is different but I just want your opinion )

 

Well to update. I identified the pad and it is the one right next to the PR70 pad. It was the PR73 pad at the bottom. And I did the mod accordingly and saw no difference in power drain. I did notice that the PR80 pad lead to the VR_ON pin and after the mod reduced the resistance to ground from 10kOhm to around 6kOhm, which I am sure changed the signal somewhat for the VR_ON pin.

Now since this didn't work I am just going to do the FSB mod and wire it to the bluetooth switch like someone else did.

 

After doing undervolting mod on AS1410 SU3500, the software still reported the old voltage levels, but temps were a bit lower. It might be working, but you cant tell that from software. Maybe you should measure the actual voltage going to the CPU on the motherboard? Not sure how to do that though

 

Well I measured based on the data from readings of power drain from the battery and not the programs that read the data from the CPU. And yeah the idea of measuring the voltage going to the CPU would be ideal but I have no idea where the test points for that would be. I would be happy to do it again if someone could give me the points where I can measure the CPU voltage.