amilo-forum.com

Everything you need to know about Amilo and Lifebook laptops by Fujitsu

D 1845 CPU disassemble issue

Everything around CPU, RAM or cooling of notebook components.

D 1845 CPU disassemble issue

Postby gagarin » Wed Mar 07, 2007 12:24 am

Hi, guys. I have been trying the get the CPU heatsink out of a D 1845 for the last 6 hours. THe idea is to squeeze some thermal paste in order to cure an overheating ploblem. However getting to the bleeding CPU is harder than I thought. Got the Sercive manual for 1840 but the fan there is different. I got the bottom plasic cover off and there are 5 screws on springs holding the whole module. I undid them but it won't move. Suppose there are screws which have to be reached from the top but I can't get the top plastics off so I stuck :(
Can anyone suggest a solution?
gagarin
 
Posts: 2
Joined: Tue Mar 06, 2007 11:50 pm
Location: london

Postby NickS » Wed Mar 07, 2007 8:46 am

The Uniwill 258SA0 service manual says there are 6 screws - did you miss one?
NickS
 
Posts: 24
Joined: Thu Nov 09, 2006 2:14 pm
Location: Nr. Basingstoke, UK
Notebook:
  • Amilo D1845
  • Lifebook S6120D
  • Toshiba Satellite Pro 4200 series

Done

Postby gagarin » Wed Mar 07, 2007 7:33 pm

The Uniwill 258SA0 service manual says there are 6 screws - did you miss one?


No, I didn't miss a screw (d 1845 got 5 screws!). What I missed is that someone has tried to cure the problem before me but he/she has put too much paste (it has gone over the edges, very shabby). So because the whole contact surface was pasted the vacuum between the heatsink and the CPU was holding them firmly together and as I was trying to remove the heatsink the whole motherboard slightly bent. Just like if there are unodone screws. But after 6 hours I decided to go brutal and forced it. The best way to get rid of the vacuum is to twist the cooling module. However, there is a risk of scratching the surfaces.
It took good 7-8 brutal twists before the two modules separated and both CPU and heatsink were badly scratched. For now I just put new paste and closed it back but I consider using t-cut (car body polisher) to get rid of the scratches. It works for now but here is the other problem.
When I use any of the hardware monitor tools the check my CPU temperature the reading is a constant 75 degrees C. I googled this issue and found out that I am not the only person who has experienced it but I never found a solution. Will try to update the BIOS today but I doubt this will fix it. Any suggestions?
gagarin
 
Posts: 2
Joined: Tue Mar 06, 2007 11:50 pm
Location: london

Re: Done

Postby NickS » Thu Mar 08, 2007 8:48 am

gagarin wrote:...here is the other problem.
When I use any of the hardware monitor tools the check my CPU temperature the reading is a constant 75 degrees C. I googled this issue and found out that I am not the only person who has experienced it but I never found a solution. Will try to update the BIOS today but I doubt this will fix it. Any suggestions?
No, I used the latest FSC BIOS and it still shows CPU at 75. Others have tried the 285SAx BIOSes at the Uniwill website but I would not recommend this - for example 285SA5US.109 ("EC fixed bugs: 1.Fix therma IC read byte issue.") has caused at least one other user's D1845 to die. Also
Re: Averatec 6110 or 6100A - The 6100 series is built by Uniwill model 258SA2. With the difference being video ram. DO NOT LOAD THE UNIWILL BIOS. WILL LOSE VIDEO IF YOU DO.
NickS
 
Posts: 24
Joined: Thu Nov 09, 2006 2:14 pm
Location: Nr. Basingstoke, UK
Notebook:
  • Amilo D1845
  • Lifebook S6120D
  • Toshiba Satellite Pro 4200 series

Yep

Postby Fezar » Thu Mar 08, 2007 4:47 pm

I have the same laptop as you and tried to reflash the BIOS to the uniwill one for better thermal control. This wrecked my board so don't use anything but a FSC OEM bios. its taking the combined efforts of NickS and the folks at wims bios to get it going again.

If you want a service guide I've put one here:
http://tom.l.ferris.googlepages.com/d1845

Documentation for our mobo is very poor, but I have been given a datasheet for the 258SA0, similar to the mobo in my 1845, the 257SA0 and the CPU thermal sensor is stated as a ADM 1032. If your into linux I suggest lm-sensors to check this out, the lm90 kernel driver may work for this, see what reading you get from it if it works... but I am conscious that I am giving advice that I have not tested myself, but will when I get it working again.
Fezar
 
Posts: 25
Joined: Wed Feb 14, 2007 12:18 pm
Location: Birmingham, UK

Re: Yep

Postby NickS » Fri Mar 09, 2007 9:37 am

Tom, my Adobe Reader says that file's broken.
NickS
 
Posts: 24
Joined: Thu Nov 09, 2006 2:14 pm
Location: Nr. Basingstoke, UK
Notebook:
  • Amilo D1845
  • Lifebook S6120D
  • Toshiba Satellite Pro 4200 series

DisassemblyManual

Postby Fezar » Fri Mar 09, 2007 12:04 pm

Thats unusual, works fine when I load the link in the browser or download it. I'm using Adobe reader 8 and it opens fine

http://tom.l.ferris.googlepages.com/Sys ... sembly.pdf

Try it out and if its still no good I'll email it to you

Tom
Fezar
 
Posts: 25
Joined: Wed Feb 14, 2007 12:18 pm
Location: Birmingham, UK

Postby Fezar » Tue Apr 17, 2007 5:07 pm

Ignore the advice about lm_sensors, it appears that the d1845 has absolutely no readable sensor chips, and the values for CPU thermal and CPU core voltage are read from unchanging values from ACPI data

There appears to me to be no workaround for his, I've been trying on and off with different programs for the last year to no avail until I tried the latest version of PC Wizard...
And since one needs to be a service partner to even look at hardware flow charts and documentation I cannot be 100% sure, but it seems bleak.
PC Wizard will give you some (ACPI) sensor readings, but they are not accurate.

At least there is another fellow sufferer out there :wink:
Fezar
 
Posts: 25
Joined: Wed Feb 14, 2007 12:18 pm
Location: Birmingham, UK


Return to CPU, RAM, fan