IBM Thinkpad 600E bios mod for processor update

Don't ask how to hack password. (BIOS Passwords)
Katch
BIOS Rookie
Posts: 40
Joined: Wed Feb 02, 2005 11:07 pm

Post a nice Hi Res photo of your module, front and back and we'll show you what to do. You need to cut 1 track from the speedstep controller and solder in a 2.2kohm resistor to complete the mod.
Thinkpad T20 - 900 MHz (Looking for 1 GHz) 512 Mb - 40Gb 5200rpm - Mini PCI wireless Mod installed behind LCD - (Looking for Slim DVDRW)
Dr. Morg
BIOS Rookie
Posts: 44
Joined: Sat Apr 16, 2005 10:14 pm
Location: Russia
Contact:

Also, I need to lower somehow a voltage on the processor.
The current voltage: 1.8 volts;
The current power: ~18 watts;
100% CPU loading power: ~28 watts;
The current temperature: 41 C;
100% CPU loading temperature: 75 C (!!!).

What it is necessary to make, that it to lower?

Thanks!
IBM ThinkPad 600E (2645-4AU) PIII 600 @ 598.2MHz /13.3"/CD-ROM/160Mb/40Gb/56k
@ [FSB_mod] => (Current ~99.7MHz)
@ [SS_mod] => Enable Intel SpeedStep technology
@ [DS_mod] => Start SpeedStep
@ [Volt_mod] => Takes 0.20 volts off Vcore (Current 1.35V)
Dr. Morg
BIOS Rookie
Posts: 44
Joined: Sat Apr 16, 2005 10:14 pm
Location: Russia
Contact:

Katch wrote:Post a nice Hi Res photo of your module, front and back and we'll show you what to do. You need to cut 1 track from the speedstep controller and solder in a 2.2kohm resistor to complete the mod.
This is my CPU:
Image
Click picture to enlarge [1.66Mb]

Thanks for your answers!
IBM ThinkPad 600E (2645-4AU) PIII 600 @ 598.2MHz /13.3"/CD-ROM/160Mb/40Gb/56k
@ [FSB_mod] => (Current ~99.7MHz)
@ [SS_mod] => Enable Intel SpeedStep technology
@ [DS_mod] => Start SpeedStep
@ [Volt_mod] => Takes 0.20 volts off Vcore (Current 1.35V)
Katch
BIOS Rookie
Posts: 40
Joined: Wed Feb 02, 2005 11:07 pm

Image

Now post me a nice HI RES of the other side of the CPU and I'll show you where to solder the 2.2k ohm resistor.
Thinkpad T20 - 900 MHz (Looking for 1 GHz) 512 Mb - 40Gb 5200rpm - Mini PCI wireless Mod installed behind LCD - (Looking for Slim DVDRW)
Sharedoc
Notebook Genius
Posts: 679
Joined: Mon Aug 18, 2003 8:46 pm
Location: Finland

Dr Morg,

600MHz PIII seems to use ADP3420 Voltage regulator for Vcore.

It is using so called Geyserville binary coding for voltage setting. The binary code is set by resistors R3, R5 and R15.

Resistors on the upper side of the mmc2 printed circuit board have effects (as far as I remember)

R3 has effect of 0,05V
R15 has effect of 0,10V
R5 has effect of 0,20V.

Example: if you remove R15, you Vcore should drop 0,10V.

You can measure Vcore from left end of condenser C17.


Warning: Please understand that stability of the processor may be compromised if you reduce Vcore. According to my experiences with Intel mobile processors 0,10V Vcore reduction does not do any harm. 0,20 volts may still be ok, but if you go below that you may start having trouble. Please make a copy of the hard disk before you boot with substantially lowered Vcore, since you may corrupt your hard disk if the processor is not stabile. You never know for sure.
Dr. Morg
BIOS Rookie
Posts: 44
Joined: Sat Apr 16, 2005 10:14 pm
Location: Russia
Contact:

Katch wrote: Now post me a nice HI RES of the other side of the CPU and I'll show you where to solder the 2.2k ohm resistor.
This is an other side of my CPU:
Image

Don't frightened a black square! ;)
It's only an additional radiator which I have pasted on thermal glue that the FPU was heated less.

Thanks for your answers!
IBM ThinkPad 600E (2645-4AU) PIII 600 @ 598.2MHz /13.3"/CD-ROM/160Mb/40Gb/56k
@ [FSB_mod] => (Current ~99.7MHz)
@ [SS_mod] => Enable Intel SpeedStep technology
@ [DS_mod] => Start SpeedStep
@ [Volt_mod] => Takes 0.20 volts off Vcore (Current 1.35V)
Dr. Morg
BIOS Rookie
Posts: 44
Joined: Sat Apr 16, 2005 10:14 pm
Location: Russia
Contact:

Sharedoc wrote:Dr Morg,

600MHz PIII seems to use ADP3420 Voltage regulator for Vcore.

It is using so called Geyserville binary coding for voltage setting. The binary code is set by resistors R3, R5 and R15.

Resistors on the upper side of the mmc2 printed circuit board have effects (as far as I remember)

R3 has effect of 0,05V
R15 has effect of 0,10V
R5 has effect of 0,20V.

Example: if you remove R15, you Vcore should drop 0,10V.

You can measure Vcore from left end of condenser C17.


Warning: Please understand that stability of the processor may be compromised if you reduce Vcore. According to my experiences with Intel mobile processors 0,10V Vcore reduction does not do any harm. 0,20 volts may still be ok, but if you go below that you may start having trouble. Please make a copy of the hard disk before you boot with substantially lowered Vcore, since you may corrupt your hard disk if the processor is not stabile. You never know for sure.
Please, can you post there some photos on which the arrangement of these resistors is precisely visible?

I'd like to drop about 0.20 volts.


Thanks!
IBM ThinkPad 600E (2645-4AU) PIII 600 @ 598.2MHz /13.3"/CD-ROM/160Mb/40Gb/56k
@ [FSB_mod] => (Current ~99.7MHz)
@ [SS_mod] => Enable Intel SpeedStep technology
@ [DS_mod] => Start SpeedStep
@ [Volt_mod] => Takes 0.20 volts off Vcore (Current 1.35V)
Sharedoc
Notebook Genius
Posts: 679
Joined: Mon Aug 18, 2003 8:46 pm
Location: Finland

Vcore resistors seem to be almost the same. Just remove R5 and you get 0,20V off Vcore. R5 is up-middle in the pic.

Removing Resistor R3 takes 0,05 volts off Vcore.

R15 (0,10v) seems to be R14 in your PCB. It is vacant, so adding a resistor there will up Vcore +0,10V.
Dr. Morg
BIOS Rookie
Posts: 44
Joined: Sat Apr 16, 2005 10:14 pm
Location: Russia
Contact:

Sharedoc wrote:Vcore resistors seem to be almost the same. Just remove R5 and you get 0,20V off Vcore. R5 is up-middle in the pic.

Removing Resistor R3 takes 0,05 volts off Vcore.

R15 (0,10v) seems to be R14 in your PCB. It is vacant, so adding a resistor there will up Vcore +0,10V.
Big thanks!

I'll try to make a SpeedStep mod, than, if after that power consumption will not decrease, - I'll reduce a voltage by desoldering R5.

Thanks!
IBM ThinkPad 600E (2645-4AU) PIII 600 @ 598.2MHz /13.3"/CD-ROM/160Mb/40Gb/56k
@ [FSB_mod] => (Current ~99.7MHz)
@ [SS_mod] => Enable Intel SpeedStep technology
@ [DS_mod] => Start SpeedStep
@ [Volt_mod] => Takes 0.20 volts off Vcore (Current 1.35V)
Sharedoc
Notebook Genius
Posts: 679
Joined: Mon Aug 18, 2003 8:46 pm
Location: Finland

SpeedStep mod will increase power consumption because it forces higher Vcore voltage and higher processor speed (if mod works).
Dr. Morg
BIOS Rookie
Posts: 44
Joined: Sat Apr 16, 2005 10:14 pm
Location: Russia
Contact:

Sharedoc wrote:SpeedStep mod will increase power consumption because it forces higher Vcore voltage and higher processor speed (if mod works).
If I'll lower a voltage, desolder resistor R5, whether it will affect negatively characteristics CPU and systems as a whole?
IBM ThinkPad 600E (2645-4AU) PIII 600 @ 598.2MHz /13.3"/CD-ROM/160Mb/40Gb/56k
@ [FSB_mod] => (Current ~99.7MHz)
@ [SS_mod] => Enable Intel SpeedStep technology
@ [DS_mod] => Start SpeedStep
@ [Volt_mod] => Takes 0.20 volts off Vcore (Current 1.35V)
Sharedoc
Notebook Genius
Posts: 679
Joined: Mon Aug 18, 2003 8:46 pm
Location: Finland

In SPeedStep PIII's there are two different Vcore voltages for high-speed and default low speed.

In your processor resistors R3, R5 and R14 set the upper Vcore voltage. They do NOT have effect on the lower Vcore.

If you intent to keep running lower speed there is NO need to change the resistors. IF you remove R5 it does not effect at lower speed.

However, if you do the SpeedStep mod then the higher Vcore is forced on always and resistors R3, R5 and R14 have effect.

I recommend first to test the SpeedStep mod and if it is working (enters higher speed) and is stabile and how much is the temperature under loading. Only after if everything is working fine would I go lowering the Vcore, and I would take a clone of the hard disk before I test lower Vcore settings.
Katch
BIOS Rookie
Posts: 40
Joined: Wed Feb 02, 2005 11:07 pm

Just thought I'd post to confirm a sucessful FSB 8% Overclock mod. Kinda trick to solder in but not too bad (must get a small soldering iron for these things).

Everything seems rock stable and my Celeron 600 is now running at 650MHz.
Thinkpad T20 - 900 MHz (Looking for 1 GHz) 512 Mb - 40Gb 5200rpm - Mini PCI wireless Mod installed behind LCD - (Looking for Slim DVDRW)
Dr. Morg
BIOS Rookie
Posts: 44
Joined: Sat Apr 16, 2005 10:14 pm
Location: Russia
Contact:

Katch wrote:Just thought I'd post to confirm a sucessful FSB 8% Overclock mod. Kinda trick to solder in but not too bad (must get a small soldering iron for these things).

Everything seems rock stable and my Celeron 600 is now running at 650MHz.
It's great!
My congratulations! :P

My 8% mod work perfect too!

Are you remember, that you promised me to post picture with SpeedStep mod? :?: :)

I've already post other side of my CPU.

Thanks!
IBM ThinkPad 600E (2645-4AU) PIII 600 @ 598.2MHz /13.3"/CD-ROM/160Mb/40Gb/56k
@ [FSB_mod] => (Current ~99.7MHz)
@ [SS_mod] => Enable Intel SpeedStep technology
@ [DS_mod] => Start SpeedStep
@ [Volt_mod] => Takes 0.20 volts off Vcore (Current 1.35V)
Katch
BIOS Rookie
Posts: 40
Joined: Wed Feb 02, 2005 11:07 pm

Just working on that with Sharedoc now, as soon as he confirms the pic I'll post it up. I am really only making the pics here, Sharedocs the man with the knowledge.
Thinkpad T20 - 900 MHz (Looking for 1 GHz) 512 Mb - 40Gb 5200rpm - Mini PCI wireless Mod installed behind LCD - (Looking for Slim DVDRW)
Post Reply