Printable View
Yup I've just taken a look and on revision 2.01g of the P5LD2 there's no other similar rectangular chip. The only exception is if it's under the "Asus Fanless Design" heatsink, which is possible but then again I don't recall anyone else mentioning this. The P5WD2 also uses the same heatsink and yet the VRM chip is visible.
Strange...
What counts is not the shape of the chip (i.e. rectangular etc.), but the numbers printed ontop of it. Nearly all chips today are being produced using several package types (shapes). So, there can be different types of ADP3181s for example, that are electrically identical, but look completely different.Quote:
Originally Posted by GFORCE100
Thanks. You may just be right. I'll take a closer look with a magnifying glass. I remember that on the P5WD2 it was a square chip, or maybe this was the P5W DH Deluxe. Either way I did the pencil mod back in 2005/2006 on whatever of two boards I had then.Quote:
Originally Posted by celemine1Gig
I found a large close-up of revision 2.01g at the link at http://www.nix.ru/autocatalog/mother.../54540_top.jpg
I have a feeling it's the square chip next to CPU FAN #2. Will verify shortly.
Yup. It's the square one next to CPU FAN #2 but it says ADP3198 and not ADP3181.
Now the only question is are these two models the same and only differ in the packaging or are they totally opposite in pinout (i.e. are pins #17 and #18 still the ones to pencil mod).
I can try taking a photo of it though it's going to be tricky.
According to the PDF datasheet for ADP3198 located at http://www.datasheet4u.com/download.php?id=560535
it would appear pin #17 and #18 are CSCCOMP and GND respectively.
CSCCOMP = Current Sense Compensation Point. A resistor and capacitor from this pin to CSSUM determines the gain of the current sense amplifier and the positioning loop response time.
GND = Ground. All internal biasing and the logic output signals of the device are referenced to this ground.
On the PDF datasheet for ADP3181 located at http://www.datasheet4u.com/download.php?id=516795 there's a difference in that pin #17 and #18 are CSSUM and CSCOMP respectively. GND is pin #19 on the ADP3181.
Given the above if my logic is correct, this would suggest that on the ADP3198 one should pencil in pins #16 and #17 as oppose to #17 and #18 on the ADP3181. Would like a second opinion though. My justification for this is that one should pencil in CSSUM and CSCOMP irrespective of the model of ADP chip...right?
I've uploaded some pin-out diagrams of both these chips and marked the pins I think are valid in red.
http://img693.imageshack.us/img693/1071/adp3181.png
http://img121.imageshack.us/img121/4862/adp3198.png
Just an update to say I followed my above line of contemplation and it works on the ADP3198, just the pins are different to the ADP3181.
Tested it with a rare Q650 ES CPU at 3.73GHz. Prior to the mod the Vdroop was bad going down to 1.520V, now it doesn't go down at all but up slightly from 1.616V idle to 1.624V load with the seldom spike to 1.632V.
Used a HB pencil rather than a 2B.
Thanks for your help celemine1Gig.
Your welcome. But you might want to consider redoing the mod. As I already said, you should leave a little droop. If the voltage actually increases on full load that probably isn't exactly good for your CPU. This most likely causes voltage spikes on load changes.
I'll redo it with a 2B pencil once I get one. The pins on the ADP3198 are smaller and closer together than on the ADP3181 so it's somewhat tricky.Quote:
Originally Posted by celemine1Gig
Just get yourself a digital multimeter and measure the resistance between the pins. When redoing the mod you would want the resistance to be a bit higher than it is now. ;)
Thanks again. Will investigate the matter. Oh how I love small pins :)Quote:
Originally Posted by celemine1Gig