• Please review our updated Terms and Rules here

IBM Thinkpad 850

Can anybody locate me the redbook and the schematic for the motherboard.

I am trying to repair my 7249-851 and cannot find either !
 
if you still want a thinkpad 850 i have one no battery but fully working complete with floppy cd and user manuals
 
thinkpad 850

thinkpad 850

NeXT said:
BuMP

Token Ring is a complete YES as I found the card with A token Ring card and my assumption is YES too for ethernet.
I tried a copy of OS/2 for Power PC and it seems it's more designed for the desktop RS/6000 systems and not the 850. :(
In the meantime, I have managed to wrangle back the hard drive so it's installed again and I raided RE-PC Seattle for a floppy drive and walked out with a 2.88mb drive for $3.

I still however can't find the restore CD's for the system so for a lot of the time it's been sitting as a paperweight...
I wish I knew someone who I could borrow their discs.
Click to expand...

I have a cd that came with the 850 with various programs on
If you still need any more info let me know
I also have the user manual, and maintanance manual

MikeS
 
I have the maintanance manual which I have scanned if you would like a copy 30H2383 also the user manual 79H3648 and the safety booklet 45H3699
I have not yet scanned these but if there is call for them the winter is coming so I may have time

MikeS
 
Mike Senior said:
I have the maintanance manual which I have scanned if you would like a copy 30H2383 also the user manual 79H3648 and the safety booklet 45H3699
I have not yet scanned these but if there is call for them the winter is coming so I may have time

MikeS
Click to expand...

Noted that IBM (http://www-01.ibm.com/common/ssi/pr...mon/ssi/rep_ca/8/877/ENUSZG96-0188/index.html) quotes as HMM for the 820 as 30H-2388 and for the 820 as 30H-2383. However your copy of the HMM is 30H-2383 and headed 850...odd
 
aberco said:
The bottom cover removed. A square cutout is made in the board under the CPU to connect the ceramic substrate to the metallic cover for heat-spreading.

View attachment 996817

The CPU board with its two connectors. 33.33MHz frequency from oscillator is tripled for 100MHz system clock. A metallic bar connects to the top of the CPU to transfer heat away.

View attachment 996818

Here's what everyone wants to see!
The PowerPC 603e CPU @ 100Mhz. It comes in a flip chip QFP208 package potted with a blue compound. This package was used for the PowerPC 601, on some early IBM versions of the PowerPC 604, but rarely on the 603e which also came in a full ceramic QFP208 package.

View attachment 996819

:arrow: Disassembling the display does not require to take the frame out of the unit. 6 screws covered by adhesive pads holds the cover. The two upper screws are made of white plastic.
The LCD is an active matrix 800x600 24bit color module (PN 84G6835). Very interestingly, the inverter and CCFL lamp (PN 85G0531 - FRU 85G1031) are mounted on a removable plastic frame meant to be replaced in case of failure. It is the very first time I see a LCD made to be serviced without taking everything apart.

View attachment 996820

My unit also came with a PCMCIA ethernet card, without the dongle. I am wondering if this is a regular unit or one specifically designed for RS/6000 systems.

View attachment 996821

Now will be the time to do more troubleshooting in order to repair the backlight, since I doubt the display assembly was used in another IBM laptop. The replaceable CCFL seems to be specifically designed for the ThinkPad 850.
Click to expand...

Great teardown, thanks a lot for doing this. I have dumped the firmware bios revisions for the 850/851 Thinkpads, the chip under the label is a 29F040A from Fujitsu. Both bios versions can be found here: https://github.com/ArcadeHacker/Dumps/tree/main/Computers/IBM/Laptops/IBM ThinkPad 850

Revision 40H5218 is dated 08 12 1996
Revision 91G0610 is dated 07 03 1995

At some point I will also try to dump the HD6473388 MCU code.

Regards.
 
ClassicHasClass said:
Has anyone found where the SCSI fuse is on these? I have a 6020 that blew its SCSI fuse with what I suspect was a defective SCSI2SD but I can't figure out what I need to bypass.
Click to expand...

A good assumption on the 7249-851 would be the white fuse (F3) next to the SCSI connector (top of the board).

1725023660494.png
A
 
Ohhhh, that does look like a very good suspect. I'll get it on the workbench next time I have a moment and see if that makes it happy.

Fortunately my AIX laptop jones is being met by my 860, but the 6020 was my first and a definitely rarer bird (tip of the hat to @bear ), so I'd love to make it fly again.
 
edcross said:
Great teardown, thanks a lot for doing this. I have dumped the firmware bios revisions for the 850/851 Thinkpads, the chip under the label is a 29F040A from Fujitsu. Both bios versions can be found here: https://github.com/ArcadeHacker/Dumps/tree/main/Computers/IBM/Laptops/IBM ThinkPad 850

Revision 40H5218 is dated 08 12 1996
Revision 91G0610 is dated 07 03 1995

At some point I will also try to dump the HD6473388 MCU code.

Regards.
Click to expand...

I made a mistake, bios revision 40H5218 is for the 830/850 Power pc desktops, not Thinkpads. The available BIOSes for the Thinkpad Power Series 850 are as follows:

Revision 91G1671 dated 09 10 1996 (Dakota bios v1.01)
Revision 91G0610 dated 07 03 1995 (Dakota bios v1.00)


Change log fr 91G1671 / v1.01

This release of microcode for the 7249/6042 Series of ThinkPad Power Series is version 1.01.

Year 2000 Support
For model 851 of machine types 7249 and 6042, system microcode must be updated to be Year 2000 ready*. Note that this statement does not apply to the product 7249 Model 860 whose system microcode is Year 2000 ready* as shipped.

You will want to upgrade your microcode to version 1.01 to ensure that your system is Year 2000 ready*. This update provides other functional enhancements. See the next section.

* Please consult the IBM Year 2000 web page at http://www.ibm.com/IBM/year2000 for IBM's definition of "Year 2000 ready" and the latest information of the status of IBM products and the Year 2000.
Enhanced Function
It contains all of the features of the previous release, 1.0, plus the following enhancements and fixes:

ROM code updates RTC Date registers improperly.
PCIC registers are not initialized on warm boot.
Flash ROM write-protect is not enabled when OS boot.
NT Shutdown & restart causes ROM hang.
ROM code updates RTC Date registers intermittently.
Windows NT ARC reports an internal error.
MPC105 REV2.4 Errata workaround should be applied.233Mhz Processor support.
 
So I got into my "ThinkPad 800" and disassembled down to the system board, but both F2 and F3 show continuity and no resistance. I bodged F2 just to make sure, and while I didn't kill it, it still comes up in the config panel with the dreaded SCSI 00370830.

Technically the IBM service docs call it a SCSI "circuit breaker." I suppose I need to be looking for one of those next. Any guesses as to which chip? There's a UCC3916 that TI seems to have been a second source for that seems to act as a solid state cutoff for termination power, and that would make sense if the SCSI2SD that zapped it pulled too much, but I can't find other such chips with a cursory Google. I should think shorting Vin to Vout would be enough to bypass it.
 
I actually got the system board out to look at it. There is a blown fuse, on the underside, at F6. This fuse is next to the hard disk connector, so this makes sense. Unfortunately bodging that did give me continuity but did not fix the SCSI problem. I'm starting to think that the bad SCSI2SD blew the fuse and took the SCSI controller with it. I couldn't find any chip that looks like a CMOS circuit breaker, but the Type 6020 has some important differences from the 850, so it's possible they didn't use one on this model.

Anyone ever tried booting these things from PCMCIA? I should find it a floppy drive.
 
I'm starting to think that the bad SCSI2SD blew the fuse and took the SCSI controller with it.
Click to expand...
This reinforces my worry that most of the SCSI emulators out there are slow, expensive, inefficient and will ruin your day if you don't configure it correctly.
 
It breaks my heart, because the "800"/6020 is a really rare bird, and it was working perfectly beforehand. But I was concerned about its hard disk and wanted to make sure that was properly preserved, and then boom.

The hard disk of course is still working.

At least the 860 works great, but I think I'll just wait and see on replacing the hard disk in that.
 
NeXT said:
This reinforces my worry that most of the SCSI emulators out there are slow, expensive, inefficient and will ruin your day if you don't configure it correctly.
Click to expand...
I'm sorry your experiences have been poor. I own ten ZuluSCSI devices (of various sizes and models) and they are in regular use on both DEC and Apple systems. I find them somewhat faster than the spinning platter drives they replaced, reasonably priced (particularly the Zulu Pico Slim if you solder it together yourself) and cannot theorize how one could possibly damage the host system.
 
As long as I have actual SCSI disks I will continue to use them until entropy and magnetic fade runs them into the ground. There's no point switching to emulated solutions when they are as I've complained above and I still have enough disks to last me another 25 years, hopefully.
The problem is the 800 series needs abnormally BIG 2.5" SCSI disks. 1gb is and was at the time massive. That's not something you were pulling out of a Powerbook at the time.
 
You must log in or register to reply here.
Back
Top