Take a look at our
ThinkPads.com HOME PAGE
For those who might want to contribute to the blog, start here: Editors Alley Topic
Then contact Bill with a Private Message

i1500 2621-540 running at 216Mhz (33x6,5). Is this correct?

Older ThinkPads from the 300, 500, 600, 700 Series, iSeries, Transnote etc.
Post Reply
Message
Author
sergiodevega
Posts: 4
Joined: Mon Nov 29, 2010 1:42 pm
Location: Avila, Spain

i1500 2621-540 running at 216Mhz (33x6,5). Is this correct?

#1 Post by sergiodevega » Tue Nov 30, 2010 5:50 am

Hello:

I have a i1500 2621-540 that i was given long time ago. I replaced the faulty board and put 2 new Kingston 128MB PC100 RAM modules. Now it works fine, but somehow slow, and I think there´s something wrong with the RAM, FSB or CORE SPEED.

This is what CPU-Z says:
Intel Mobile Celeron 433 Mhz <---------
Code Name: Dixon
Core speed: 216 Mhz <----------
Multiplier: x 6.5
Bus Speed: 33.3 Mhz

Everest says it is a Mobile Intel Celeron-A Mendocino and shows the same speeds as CPU-Z.

Are theese speeds corrects? Shouldn´t CORE SPEED be 433mhz 66x6.5?

I believe that the BUS SPEED for this sistem should be 66. ¿Why is it only 33?

I have tried different RAM molules, but the RAM clock always stays at 33, getting a ratio 1:1.

Can you please tell me if this sound normal to you?

I suspect that my sistem is running at only 50% of it´s factory speed. Am i wright or wrong?


Thank you very much.

Sergio.

(Excuse my language, as english not my mother tongue)

sergiodevega
Posts: 4
Joined: Mon Nov 29, 2010 1:42 pm
Location: Avila, Spain

Re: i1500 2621-540 running at 216Mhz (33x6,5). Is this correct?

#2 Post by sergiodevega » Fri May 20, 2011 1:06 pm

Hello.

I have some more info here: this afternoon i have dissasembled the laptop and found out that there was a very thick layer of thermal paste on top of the CPU (i put there last time... one of my first times doing it).

Now that i am a bit more experienced using the thermal paste, i know that it shouldn´t be like that. Searching around i came to the conclusion that the heatsink is designed to use with tape not paste. It has the rectangle to fit the CPU deeper than the rest of the surface.

I have no thermal tape at hand but i am ansious to know if that could be the reason for the sistem (fsb and memory) running half speed (33mhz instead of 66mhz).

What do you think? Have any of you experienced something like this?

would it be possible to put away (cut) the four metal pieces in the corners of the heatsink that make it stand higher, and use thermal paste making sure that it makes pressure over the CPU. Or would i risk to make a short circuit?

Thanks in advance for any opinion.

ozzymud
Senior Member
Senior Member
Posts: 590
Joined: Sun Apr 03, 2011 3:38 pm
Location: Klamath Falls, OR

Re: i1500 2621-540 running at 216Mhz (33x6,5). Is this correct?

#3 Post by ozzymud » Fri May 20, 2011 7:32 pm

The thermal tape is probably not really tape but a square pad of "grease", I know it comes on stock AMD retail processors... but I always scrape it off and use Arctic Silver (1 pea sized ball)...

As far as the standoffs, i would definately NOT remove them, you risk crushing the actual chip, which is just the tiny die in the middle of the whole chip.

As to speed, if it is overheating, yes, the cpu will reduce in speed to prevent damage... does your bios report temperatures? If not there are windows apps available, or use lmsensors in linux.

The AMD cpu with the 4 standoffs... http://members.arstechnica.com/x/tangle ... ropped.JPG note the actual chip is only the black part in the center... thus the standoffs

The "tape" i normally see on the bottom is actually thermal compound...
http://www.elitebastards.com/hanners/am ... r_rear.jpg

could you post an image of your cpu and heatsink so that we can see and better advise?
(2)701C,(1)760EL,(6)760XL,(1)760XD
(4)CD Drives (5)int floppies (3)ext floppy (4)2.1GB
(10)CF/IDE w/2 or 4GB 133x CF (1)760XL restore CD
(1)Belkin USB 2.0 32bit Cardbus (2)WPC54G(S) Wifi Cardbus
(1)Belkin F5D5020 NIC (1)Giga-Byte GN-WLM01 Wifi
(1)Backpack CD (1) Xircom REM56G-10 + misc

sergiodevega
Posts: 4
Joined: Mon Nov 29, 2010 1:42 pm
Location: Avila, Spain

Re: i1500 2621-540 running at 216Mhz (33x6,5). Is this correct?

#4 Post by sergiodevega » Sat May 21, 2011 7:10 am

Hello ozzymud, and thanks for your help.

Here is a picture of the CPU and the heatsink.

Image

Image

When the heatsink is in place, the heatsink doesn´t touch the CPU. It´s difficult to messure, but i guess (because of the thick layer of compound i found when removing it) that there is room to fill of about 0,5mm.

AFAIK, the thermal paste is to be aplied in a very thin layer, and with some pressure over the CPU. So if i put thermal paste again, it´s not going to be that tight. That´s why i think i should apply 0,5mm thick tape. Or the rude way... cutting the four metal pieces of the corners and tightening the screws carefully, not all the way down, but enough to make the heatsink touch the CPU.

What do you think?

By the way, do you know what this DIP SWITCH called SWZ1 shown in the following picture is for?

Image

Thanks a lot.

ozzymud
Senior Member
Senior Member
Posts: 590
Joined: Sun Apr 03, 2011 3:38 pm
Location: Klamath Falls, OR

Re: i1500 2621-540 running at 216Mhz (33x6,5). Is this correct?

#5 Post by ozzymud » Sun May 22, 2011 3:35 am

Yea, your correct on thermal paste usage... it is only intended to fill minor imperfections in the heatsinks surface, when I "lap" (wet sand on glass surface from 400 to 2000 grit) my desktop heatsinks then apply arctic silver, then later remove the heatsink, you can barely tell there is any compund in between.

Thermal tape is generally used for like ram chips on video cards, northbridge chips, etc... i.e. cases where you won't normally remove the heatsink, same goes for thermal adhesive.

I would definately not try shortening the standoffs on the heatsink/motherboard... you definately run the risk of crushing the actual cpu (the blue center part).

From all that i have read, the intel heatsinks come with that same "pre-installed" thermal paste applied to the heatsink. I know it it a decently thick layer, but in my experience (AMD guy here on desktops), it pretty much "squishes" completely down.

What does it look like if you apply a pea sized ball of compound to the cpu, then attach and then remove it? ---> O <-- a ball this tiny


Here is what the factory versus arctic silver looks like on an old AMD Duron that had factory and I just removed it to replace with arctic silver... (Also notice older AMD chips were like your Intel, but the "standoffs" were foam circles ON the cpu...

Original compound on heatsink & cpu:
Image Image
The "pea sized" gob of arctic silver (this was even TOO much :P ):
Image
Heatsink & cpu after removing with arctic silver (notice all the extra compound):
Image Image

Heh, this heatsink needs "lapping" :O Not enough bare spots in the compound for my liking :P

BTW, this is all besides the point if you weren't overheating, did you ever get some thermal readings from either the BIOS or from software?
(2)701C,(1)760EL,(6)760XL,(1)760XD
(4)CD Drives (5)int floppies (3)ext floppy (4)2.1GB
(10)CF/IDE w/2 or 4GB 133x CF (1)760XL restore CD
(1)Belkin USB 2.0 32bit Cardbus (2)WPC54G(S) Wifi Cardbus
(1)Belkin F5D5020 NIC (1)Giga-Byte GN-WLM01 Wifi
(1)Backpack CD (1) Xircom REM56G-10 + misc

sergiodevega
Posts: 4
Joined: Mon Nov 29, 2010 1:42 pm
Location: Avila, Spain

Re: i1500 2621-540 running at 216Mhz (33x6,5). Is this correct?

#6 Post by sergiodevega » Wed May 25, 2011 5:32 pm

Hi. Back to the subject.

I have taken some more pictures.

Here i aply a lot of paste
Image

Put the heatsink back on place and tighten screws
Image

take it apart again and see that the paste is still there :(
ImageImage


The CPU temp right after booting is about 55º. After some usage it goes up to 73 and if i don´t use a fan refrigerated laptop base... it normally hangs after a while.

After this testing... i was too ansious to find out if it would run at FSB 66 with the proper heatsing alignment... and cut the standoffs and put new screws.

It runs cooler now (fortunately did´t crush the CPU) and don´t need the fan base... even stressing the system with everest it doesn´t hangs. The CPU now runs about 5 degrees cooler... but the FSB is still 33.

I´ve done some research, and found out that the faulty original system i was given was a Thinkpad iSeries 1400 2621/489 and the motherboard i bought (including the body) was the iSeries 1500 2621/540. All the parts from the first one fitted properly into the new one, and then added the two memory modules. MAYBE THIS IS THE CLUE FOR THE WRONG SPEED. don´t know. I´m lost here.

I´ve downloaded the bios update disk utility from lenovo and i get the error message that the utility doesn´t recognize the system and then it says something like "pci device not found" and ask me to press any key to reboot.

Well this is an old laptop and i think i´ve invested too much time with this. I am giving up. It runs very slow but will try to use it for some light applications as i don´t have XP MODE in my other win7 laptop.

If you can give me any last minute advise i will give it a try... otherwise that´s all.

THANK YOU.

Sergio

ozzymud
Senior Member
Senior Member
Posts: 590
Joined: Sun Apr 03, 2011 3:38 pm
Location: Klamath Falls, OR

Re: i1500 2621-540 running at 216Mhz (33x6,5). Is this correct?

#7 Post by ozzymud » Wed May 25, 2011 9:03 pm

Sounds like the seller of your replacement mobo/body had pieced together a working set, but might have fitted the cpu with thewron heatsink for the motherboard... even thermal tape is not designed to fill a gap that big... 3M's tape comes in 0.125 mm, 0.25 mm, .375 mm, with the FATTEST being 0.50 mm...


Anyhow, the SW1 on the 1400 is listed as "unconfigured" in the hardware maintenence manual with the default setting being off for both switches...

Also SW2 =
  • 1 = unconfigured (OFF)
  • 2 = password bypass - bypass it (ON) check password (OFF)
  • 3 & 4 = Language -
    • Japanese = 3(ON) - 4(OFF)
    • U.S. = 3(OFF) - 4(OFF)
    • European = 3(OFF) - 4(ON)
So, anyhow, 50C and 73C is well within the 100C max for this chip... so after some digging with google, i found this on these forums...

http://www.thinkpads.com/forum/viewtopi ... 36&start=0

Aboout a 600E with a Celeron WITHOUT Speedstep, if "Max Battery" is selected in Windows 2000, then the FSB is halved, to solve the user always selects "Minimum Power Management"

Try switching XP to "Minimum Power Management" and see what CPU-ID says then (or after a reboot if no change)
(2)701C,(1)760EL,(6)760XL,(1)760XD
(4)CD Drives (5)int floppies (3)ext floppy (4)2.1GB
(10)CF/IDE w/2 or 4GB 133x CF (1)760XL restore CD
(1)Belkin USB 2.0 32bit Cardbus (2)WPC54G(S) Wifi Cardbus
(1)Belkin F5D5020 NIC (1)Giga-Byte GN-WLM01 Wifi
(1)Backpack CD (1) Xircom REM56G-10 + misc

Post Reply

Return to “ThinkPad Legacy Hardware”

Who is online

Users browsing this forum: No registered users and 17 guests