Testing Motherboards??

Do you have a question? Post it now! No Registration Necessary.  Now with pictures!

Threaded View

I have 3 motherboards none of which work. I guess they went on strike ;-)
None of them look like they were fried.

Is there a reputable outfit around that could hook up these boards and run
diagnostic tests? and do they charge more than the value of the boards for
the testing and/or  replacement of a capacitor or two? Or are there any
computer progs that could help out?

I already have a collection of CPUs, fans, memory chips etc, etc. Matching
them up and getting them to run again should keep me out of mischief for a
while  :-)


Re: Testing Motherboards??

RF wrote:
Quoted text here. Click to load it

I just bought a few *new* motherboards for about $50 each,
so there is no way you'd find anyone to repair them economically.

However, if you have CPU's and RAM for them...
might as well try re-setting the bios...
that sometimes gets "dead" mobo's working again

Re: Testing Motherboards??

philo wrote:
Quoted text here. Click to load it

Check the CMOS batteries, which will also reset the CMOSes.

Test the boards without RAM first. If one complains then
it might be OK, otherwise the board's probably toast.

-- Marten Kemp
(Fix name and ISP to reply)

Re: Testing Motherboards??

Quoted text here. Click to load it
Quoted text here. Click to load it

Thanks Marten for your suggestion. The computer clock is keeping accurate
time so I expect that the battery is ok, but I will check that later and
will also remove the memory chips in the test. I have other interesting news
to post - below.

Re: Testing Motherboards??


Quoted text here. Click to load it

No, their value is far lower than diagnostics fees,
particularly anyone competent enough to find more difficult
falures.  IE - they may easily take the money and either say
they can't find the fault or make up something infeasible to
repair (they may believe the diagnosis but it could take
many hours to find some kinds of faults).

However there are a few things you can try yourself.  Start
out with known constants, all parts known or at least
believed to be compatible, and known to be in working order
besides the motherboards.

Take the battery out, check it's voltage (some boards will
not post without high enough battery voltage), clear CMOS,
inspect capacitors (on the board itself it is possible but
rare that a capacitor fails and yet doesn't vent, or explode
leaving an empty spot with partial leads remaining), cards,
slots, cables.  Ideally a board is tried outside of the case
on a non-conductive (not ESD dissipative) surface.  It still
won't be worth the time, but as a hobby or learning
experience, or if you are just a glutton for punishment you
may get a few of them working.

Beyond a POST card, no there are no programs to help you
out, at least not on an individual level though I am sure
there is a jig and a program the manufacturers themselves
use to test in-house, one that they are not going to share
with the outside world.

If you just need some spare old motherboards to complete a
few systems with other parts you have lying around, make
friends with the grunts at a local computer shop.  They'll
often have excess old boards from systems that they parted
out because a customer brought one in and it wasn't worth
the cost to repair it or even salvage data and reinstall the
OS and apps... not when a new low end Dell/HP/etc can be had
for a hundred or two more and will not only run circles
aroud an old system, it'll also have more of it's viable
lifespan remaining, come with a modern OS (Win2k and older
based system owners are now finding more and more popular
software like Turbotax, AVG, etc, will not be supported on
win2k anymore and in some cases that means there are
functionality bugs that they are not going to fix even if in
theory apps that are bug-free and work on xp ought to run on
2k too, until a developer depends on features not
back-ported to 2k at least).

Anyway I used to sell a fair amount of bulk supplies to
local PC shops in a metropolitan area.  They'd give me a box
of boards for the asking sometimes, other times at $5 each
or in trade for something trivial.  A good 2/3rds of them
(besides the ones that had burst capacitors) were still
good, or good again after a battery swap and CMOS clear).
That's a good way to get cheap OEM cases too, if you don't
mind the hassle of having to trace the pinout for the front
panel LEDs and switches and have a spare connector or two
lying around.

Re: Testing Motherboards??


Quoted text here. Click to load it

You might also get old boards rather cheap if not the cost
of shipping alone on a FS/FT web forum like the following
(or one local to you if not in the continental US), or of
course an auction 'site like ebay.  At least with some of
these options there is the possiblity of finding something
the owner claims, and hopefully guarantees, works still.



Re: Testing Motherboards??

----- Original Message -----
Newsgroups: alt.comp.hardware
Sent: Wednesday, May 13, 2009 6:39 PM
Subject: Re: Testing Motherboards??

Quoted text here. Click to load it

Something wierd is going on here. I am using OE (last gasp choice!) in the
old box for this communication and it seems to have lost some of my
responses to this group.
It is unlikely to be a problem with the hard disk on this old box - it has
been tested and cleaned up and I have SP4 installed. Oh, well - just another
gremlin ;-)

Thank you Kony and Philo. I am now splitting the time between getting a new
motherboard and satisfying my curiousity. I did some things along the lines
you both suggested. First I stripped out all the unnecessary components and
then switched everything off and shorted the BIOS - nothing new discovered.
While doing that I thought about a test prog that I used way back - I wanted
some info about the CPU but I couldn't remember the name of the prog and
spent a while browsing a disk that was on the newer box when the crash came
and is now on the old. Finally I found it - SiSoftSandra. I copied the .exe
to the newer box where I had a test Win2K installation running - minus SP4.

I ran SiSoft on the CPU and the results blew my mind. Here is the report:

Model : Intel(R) Pentium(R) 4 CPU 2.26GHz
Co-Processor (FPU) : Built-in
Speed : 2.27GHz
Performance Rating : PR2494 (estimated)
Type : Standard
Package : FC PGA478
Rated Speed/FSB : 2260MHz / 4x 133MHz
Multiplier : 17x
Generation : 7 (7x86)
Name : P4N (Northwood) Pentium 4 1.6-3.4G 1.5-1.6V
Revision/Stepping : 2 / 9 (9)
Stepping Mask : D1
Core Voltage Rating : 1.550V
Part Number : To Be Filled By O.E.M.
Asset Tag : To Be Filled By O.E.M.
Serial Number : To Be Filled By O.E.M.

Processor Cache(s)
Internal Data Cache : 8kB Synchronous Write-Thru (4-way sectored, 64 byte
line size)
Internal Trace Cache : 12kops Synchronous Write-Thru (8-way, 64 byte line
L2 On-board Cache : 512kB ECC Synchronous ATC (8-way sectored, 64 byte line
L2 Cache Multiplier : 1/1x  (2267MHz)

Host Interface(s)
Front Side Bus Speed : 4x 133MHz (532MHz data rate)

Socket/Slot : J2E1
Upgrade Interface : Socket 478
Supported Speed(s) : 3.06GHz+

Environment Monitor 1
Model : AD ADM1027 SMB
Mainboard Specific Support : No

Power Rating(s)
CPU Core Power : 57W (estimated)

CPU Temperature : 25.0C / 77.0F td
Auto Fan Speed Control : Yes
CPU Fan Speed : 917rpm
CPU Voltage : 1.51V

FPU - Co-Processor Built-in : Yes
VME - Virtual Mode Extensions : Yes
DE - Debugging Extension : Yes
PSE - Page Size Extension : Yes
TSC - Time Stamp Counter : Yes
MSR - Model Specific Registers : Yes
PAE - Physical Address Extension : Yes
MCE - Machine Check Exception : Yes
CX8 - Compare & Exchange Instruction : Yes
APIC - Local APIC Built-in : Yes
SEP - Fast System Call : Yes
MTRR - Memory Type Range Registers : Yes
PGE - Page Global Enable : Yes
MCA - Machine Check Architecture : Yes
CMOV - Conditional Move Instruction : Yes
PAT - Page Attribute Table : Yes
PSE36 - 36-bit Page Size Extension : Yes
PSN - Unique Serial Number : No
CLF - Cache Line Flush Support : Yes
DS - Debug Trace & EMON Store : Yes
ACPI - Software Clock Control : Yes
MMX Technology : Yes
FXSR - Fast Float Save & Restore : Yes
SSE Technology : Yes
SSE2 Technology : Yes
SS - Self Snoop : Yes
HTT - Hyper-Threading Technology : No
TM - Thermal Monitor : Yes
SBF - Signal Break on FERR : Yes
IA-64 Technology : No
SSE3 Technology : No
MON - Monitor/MWait : No
DSCPL - CPL qualified Debug Store : No
TM2 - Thermal Monitor 2 : No
CID - Context ID : Yes
DAZ - Denormals Are Zero : Yes

Advanced Settings
Data Error Checking : No
Fast Strings : Yes
x86 FPU Compatibility Mode : No
Prefetch Queue : Yes
Branch Trace Storage : Yes
Data Cache Active Mode : Yes
IO Queue Depth : 12 request(s)
Thermal Monitor Enabled : Yes

Machine Check Architecture Settings
Number of Reporting Banks : 4 bank(s)
Extended Machine Check Support : Yes
Number of Extended Reporting Banks : 12 bank(s)

Variable Range MTRR Settings
MTRR 0 : 00000000-3FFFFFFF (0MB-1024MB) WB

Fixed Range MTRR Settings
MTRR 0 Range 0 : 00000000-0000FFFF (0kB-64kB) WB
MTRR 0 Range 1 : 00010000-0001FFFF (64kB-128kB) WB
Quoted text here. Click to load it
.MTRR 10 Range 6 : 000FE000-000FEFFF (1016kB-1020kB) WB
.MTRR 10 Range 7 : 000FF000-000FFFFF (1020kB-1024kB) WB

PAT Settings
PAT 0 : WB
PAT 1 : WC
PAT 2 : UC-
PAT 3 : UC
PAT 4 : WB
PAT 5 : WC
PAT 6 : UC-
PAT 7 : UC
I had a tip and a warning:

Tip T210 - The main board supports faster CPUs so you don't need to upgrade
the main board when you decide to install a new CPU. However, a new main
board will yield faster performance with a new CPU simply because it will
use a new chipset.
Fix: Nothing to fix.

The Warning 216 was:  Warning W216 - The CPU fan has failed. This should
appear only if the CPU has a fan with tachometer installed, connected
correctly to the main board data capturing chip.
Fix: Power down your system immediately. Open the case and check that the
CPU fan is working and the heat sink is not unusually hot. Make sure the CPU
has good all-round ventilation.
SiSoftware Sandra Help File

Back to me :-)

The fan ran hot for a very short while - possibly 15 mins. Fortunately I was
able to shut down quickly just when the fan was barely able to rotate. The
fan hub was very hot - the aluminum heat sink was not very hot. I checked
all the capacitors on the board and there were no bulges in any. There were
no traces of damage on the board around the CPU - the thermal compound
looked soft and very normal.  The air flow around the CPU had no
impediments - the fan problem undoubtedly was due to lack of lubrication -
the need for which was completely hidden by the Spire company. The time of
the fan failure was not given, and I am astonished at how SiSoft could tell
now that a crash happened a week ago - could there be some data cache on the
motherboard?  It is a SiSoft 2003 version.

The fact that the OS acted up after that incident may have been due to an
unrelated cause.
I still need to do some more checking on the newer motherboard and will
check for some Intel test progs and possibly a BIOS upgrade.

Thanks to you all for your interest and help.

Site Timeline