[Search for users] [Overall Top Noters] [List of all Conferences] [Download this site]

Conference noted::ibmpc-95

Title:IBM PCs, clones, DOS, etc.
Notice:Intro in 1-11, Windows stuff in NOTED::MSWINDOWS please
Moderator:TARKIN::LINND
Created:Mon Jan 02 1995
Last Modified:Fri Jun 06 1997
Last Successful Update:Fri Jun 06 1997
Number of topics:3023
Total number of notes:28404

2901.0. "The Route to the Bios" by TAEC::SMITH (Martin Smith, Valbonne. - 828 5128) Fri Jan 31 1997 12:06

    What happens inside a PC, from the moment the Power (On/Start/...) 
    button is pushed, until the BIOS displays its characters on the
    screen? 
    
    The reason I ask, is that I upgraded my children's 386 to a 486/100 
    over the Christmas holidays, and I have not been able to bring it 
    back to life since. (Dad's reputation has suddenly changed!!!)
    
    All that happens is that a few disk seeks can be heard, then silence!
    
    At the same time, I upgraded the cache to 256k. [on hindsight, perhaps
    I should have done one thing at a time!]
    
    As the PC didn't speak to me, I decided to revert back to the 386 
    configuration. The same few disk seeks, then the 'deadness'. :-(
    
    I can confirm that the PC was working correctly minutes before I
    attempted the changes.
    
    Does anyone have any clue as to what could be the reason why the PC
    is still unwell?
    
    Martin (who thought an upgrade was easy!)
T.RTitleUserPersonal
Name
DateLines
2901.1Video card problem???SUBSYS::WARKFri Jan 31 1997 12:1925
	I have seen this problem with some Orchid video cards...
In what I have seen, you don't get BIOS 'beep code' errors,
the drives spin up, no floppy seek...  I've always assumed
the system was getting lost in the video BIOS startup
routines...  The steps I take when it happens are:

	1) Hit the reset button.  Sometimes this will work

	2) Reseat the video card.

	3) Move the video card to another slot (which
		is a more advanced reseat.  With one
		card, moving the slot makes it work
		for a while, then it will stop, but I
		can get it to work back in it's original
		slot, contacts problem, maybe?)

	4) Put in a different video card.  Put old card
		back two days later.

In any event, at least try another video card.

FWIW

Steve Wark
2901.2CAMPY::ADEYIs there a 'Life for Dummies'?Fri Jan 31 1997 12:233
    Sounds to me like it could be a power-to-the-motherboad problem.
    
    Ken....
2901.3floppy driveCSC32::I_WALDOFri Jan 31 1997 13:203
    I had the same problem when I went to a new motherboard.  My old floppy
    drive was killing it.  Got a new floppy drive and cable and didn't look
    back.
2901.4Thanks for your comments.TAEC::SMITHMartin Smith, Valbonne. - 828 5128Mon Feb 03 1997 02:5414
�.1 I have seen this problem with some Orchid video cards...
    
    I don't think it is the video card - there isn't one! ;-) 
    The video chip, and memory, are included on the m/board.
    
�.2 Sounds to me like it could be a power-to-the-motherboad problem.
    
    This had crossed my mind.
    
�.3 My old floppy drive was killing it.
    
    Interesting thought!
    
    Martin.
2901.5May be main memoryRTOMS::dhcp-185-48-51.hao.dec.com::WorkBenchUserMon Feb 03 1997 05:508
I've had that problem going from 486/25 to 486/100 with a new motherboard.

It turned out to be the main memory (I went from Simm to PS/2). Reseating it 
made the problem go away.

The machine is working fine since.

-- Jean
2901.6HELIX::WELLCOMESteve Wellcome SHR3-1/C22 Pole A22Mon Feb 03 1997 10:0515
    I had a similar problem over the weekend when I installed some
    additional memory in my Celebris XL 590.  To get at the memory
    slots, I had to remove the daughtercard with the CPU on it.
    I put in the memory, put everything back, and had the same
    symptoms you described.  I assume that somehow the new memory 
    is causing problems.  I took out the memory, tried again, still nothing.
    I went back, reseated everything I could find (including a
    couple of things on the daughtercard) and it worked.  I put
    the new memory back in, and it still worked.
    
    So...I'd suggest going back and reseating everything in sight
    and seeing what happens.  I assume you took reasonable anti-static
    precautions (touched the case before touching anything else, etc.?)
    
    
2901.7Watch this space!TAEC::SMITHMartin Smith, Valbonne. - 828 5128Mon Feb 03 1997 12:0811
�.6 I went back, reseated everything I could find
    
    I'll try that tonight. I still cannot understand that the addition
    of a processor could cause something, elsewhere on the board, to fail.
    
�.6 I assume you took reasonable anti-static precautions
    
    Oh yes! I didn't want to zap the new chips. I even had an elbow on part  
    of the earthed chassis as I inserted the chip. 
    
    Martin (with crossed fingers!)
2901.8BBQ::WOODWARDC...but words can break my heartMon Feb 03 1997 18:5511
    Hullo,
    
    for those that are delving into the gizzards of PCs often, a good
    investment is an anti-stat wrist-band. After having delved in a few PCs
    over the years using the 'keep one arm attached to the case' method,
    and now having such a beast - I can tell you the added mobility is a
    real plus!
    
    hth,
    
    H
2901.9Clock speed?SALEM::DACUNHATue Feb 04 1997 03:038
    
    
    	Check the clock speed.  Some motherboards and associated sub-
    systems can't quite handle the CPU-MEM transactions at higher speeds.
    
    	You may be limited to to say, 80Mhz.
    
      
2901.10Thanks for your comments.TAEC::SMITHMartin Smith, Valbonne. - 828 5128Tue Feb 04 1997 05:0211
    Well, I removed all memory SIMMs, and replaced them, making sure that 
    they were seated correctly, then switched on - Nothing! :-(
    
�   You may be limited to to say, 80Mhz.
    
    When I contacted the (UK) manufacturer about upgrading the processor
    last year, they suggested the 486/100. 
    
    I think I'll have to contact creator again!
    
    Martin.
2901.11More info pleaseTLE::INGRAMoopsTue Feb 04 1997 11:1831
    
>    The reason I ask, is that I upgraded my children's 386 to a 486/100 

	Is this one of those special 386-to-486 upgrade chips or do you have
	one of those 386/486 motherboards with lots of jumpers?
    
>    All that happens is that a few disk seeks can be heard, then silence!

	What is seeking? The floppy or the hard disk(s)? If the hard disk(s), 
	this would be their own power up re-cal sequence and wouldn't require
	the processor to be working. If it's the floppy, the processor must
	be working at least partially.
    
>    At the same time, I upgraded the cache to 256k. [on hindsight, perhaps
>    I should have done one thing at a time!]
>
>    As the PC didn't speak to me, I decided to revert back to the 386 
    
	If you haven't already, try removing the extra cache. Did you have
	to change jumpers?
    
	Since you're not getting any "beeps" from the speaker, either the
	processor/motherboard is not working at all or the speaker is not
	connected. I'll bet on the motherboard.

	Could you go into a little more detail on exactly what you started
	with, what components you used in the upgrade, what jumpers you
	changed and what components/assemblies you had to remove?

Larry

2901.12CAMPY::ADEYIs there a 'Life for Dummies'?Tue Feb 04 1997 12:386
    re: Note 2901.10 by TAEC::SMITH
    
    Try booting with the absolute minimum memory configuration required
    by the motherboard.
    
    Ken....
2901.13Is there a light at the end of THIS tunnel?TAEC::SMITHMartin Smith, Valbonne. - 828 5128Wed Feb 05 1997 05:3859
    Thanks Larry for your questions. Here are my answers:

�   Is this one of those special 386-to-486 upgrade chips or do you have
�   one of those 386/486 motherboards with lots of jumpers? 

    The chip is an Intel 486/100 Overdrive, and yes, I have a 386/486 
    motherboard with many jumpers.
    
�   What is seeking? The floppy or the hard disk(s)? If the hard disk(s), 
�   this would be their own power up re-cal sequence and wouldn't require
�   the processor to be working. 

    It's the two hard disks, so I guess the noises are the disks going 
    through their power-up sequences.

�   If you haven't already, try removing the extra cache. Did you have
�   to change jumpers? 

    To increase the cache, I had to replace the existing 9 chips (of 
    64Kb) with 9 new chips (giving 256Kb). Two or three jumpers had to 
    be changed. 
    
�   Since you're not getting any "beeps" from the speaker, either the
�   processor/motherboard is not working at all or the speaker is not
�   connected. I'll bet on the motherboard. 

    I tend to agree with you about it being the motherboard, as when I 
    started the PC before 'messing about with it'. the speaker did 
    'beep'.

�   Could you go into a little more detail on exactly what you started
�   with, what components you used in the upgrade, what jumpers you
�   changed and what components/assemblies you had to remove? 

    The motherboard has an AMD 386DX processor soldered onto the mother-
    board, covered by a socket to accommodate a 486 processor. I simply 
    pushed the 486 into this socket, making sure that pin 1 was in the 
    correct position! I then changed the required jumpers (according to 
    the manual and updated spec. sheet (from the manufacturer)). Lastly, 
    I replaced the cache (as descibed above). 

    So, feeling like a kid with a new toy, I switched on. After a few 
    seconds, my confidence and enthusiasm suddenly took a nose-dive! :-(

    The Intel manual that accompanies the new processor states that the 
    old processor must be removed before inserting the new one. In view 
    of the fact that it was soldered onto the motherboard, and when I 
    bought the PC, I was told that to upgrade, I simple insert the new 
    chip into the socket, I left the 386 where it was. [I assume the 
    jumpers take care of the processor type?]

    One thing I have noticed when I have left the PC on for a few 
    minutes - the 486 heatsink begins to get warm. I assume this is normal 
    (otherwise the heatsink wouldn't be there. Ha! Ha!). Therefore, I 
    assume that 'something' knows that the 486 is present?!?! 

    Do you think the motherboard is stuffed? If so, how? 

    Martin.
2901.14TLE::INGRAMoopsWed Feb 05 1997 10:4013
	The heatsink getting warm means the chip is getting power, that's
	about it.

	If you haven't touched the cache since you upgraded it (you didn't
	say if you had in your reply), perhaps you've got a bad cache chip or
	a bent pin? This is the only common thing to both of your processors
	that was also changed. Well, except of course for the jumpers. I'm
	sure you've probably checked and re-checked them several times over
	by now.

Larry

2901.15Could even be the Processor!TAEC::SMITHMartin Smith, Valbonne. - 828 5128Wed Feb 05 1997 12:3211
�   If you haven't touched the cache since you upgraded it (you didn't 
�   say if you had in your reply)
    
    I did - third paragraph. I have (carefully) changed the 256Kb cache 
    with the 64Kb cache chips, making sure that the pins were 'straight' 
    each time.
    
    Thanks for your replies. The problem may have been narrowed down to 
    either the motherboard(?) or the cache. 
    
    Martin.
2901.16go back to square 1WRKSYS::SOVIEPKO3-2/T25Wed Feb 05 1997 16:2614
    
    Did you check that the speaker wires are connected properly?
    
    Maybe the motherboard is beeping and you can't hear it.
    
    Pull the cache chips out and run without external cache. <test>
    
    pull the new processor <test>
    
    put the cache back in and run with the old processor <test>
    
    reinstall the cpu.
    
    /Dean
2901.17TAEC::SMITHMartin Smith, Valbonne. - 828 5128Thu Feb 06 1997 07:1312
�   Did you check that the speaker wires are connected properly?
�   Maybe the motherboard is beeping and you can't hear it.
    
    The speaker is soldered onto the motherboard, and it was working before 
    I opened the box.
    
�   Pull the cache chips out and run without external cache. <test>
    
    I haven't yet tried that, although I did replace the original cache 
    chips - and it still didn't work. I'll try it minus the cache.
    
    Martin.
2901.18Time for a new Motherboard?TAEC::SMITHMartin Smith, Valbonne. - 828 5128Fri Feb 07 1997 03:2211
�   Pull the cache chips out and run without external cache. <test>
    
    Well, I did that last night, and guess what?
    
    
    No Change! BooHoo! :-(
    
    Thanks for all your suggestions. I expect I'll take it back from where I
    bought it, and get them to replace the (???).
    
    Martin.
2901.19CARLSN::GASSERTFri Feb 07 1997 23:203
    Is your 486/100 a 3.3 volt chip or a 5volt overdrive chip? I would
    think you would need the 5 volt overdrive for this to work.
    	kevin
2901.205 Volt.TAEC::SMITHMartin Smith, Valbonne. - 828 5128Mon Feb 10 1997 05:111
    
2901.21How are the kids ?EPS::CONNORSMon Feb 10 1997 12:484
    
    I hope your children have at least 1/2 the patience and
    drive you have !
    
2901.22One of the three R's! :-)TAEC::SMITHMartin Smith, Valbonne. - 828 5128Tue Feb 11 1997 07:436
    Well, one positive thing has come out of this (iro my children): 
    
    As the PC has been 'unwell', they have had to remember how to write 
    again! :-)
    
    Martin.
2901.23The end of this particular story.CHEFS::SMITH_MMartin Smith, Reading. - 830 4544Fri May 09 1997 10:349
    I took the machine back to the manufacturer, who stated that one of the
    jumpers was positioned wrong.
    
    [I lost count at the number of times I checked the jumper positions
     against the data sheet. One of these days I'll check what jumper has
     been changed. Anyway, the machine is functioning fine now - even on
     my 2-PC network!]
    
    Martin.