A7N8X-E Deluxe powers off

  • Thread starter Thread starter Hans J. Ude
  • Start date Start date
H

Hans J. Ude

I've bought a brand new A7N8X-E Deluxe witx XP 2800+ CPU and 512 MB
RAM from Empaq (PC 2700, 333Mhz, CL2.5). When I run it with 166 Mhz
clock it always powers down after a while. Sometimes that happens
after minutes, sometimes after hours but it _will_ happen. After that
it can't be restarted with the standby button at the front. I have to
switch main power off and on to get it restarted. Can this be the
power supply? It has 300W by the way. I'm clueless here. Does anyone
know about that problem? There is some extra info from sysinfo below.
Everything works at a clock rate of 133 MHz but I get only the
perfromance of a XP 2200 CPU. Sometimes power off happens even in CMOS
setup, so it can't be an OS issue.

TIA
Hans

Systemhersteller ASUSTeK Computer INC.
Systemmodell A7N8X-E
Systemtyp X86-basierter PC
Prozessor x86 Family 6 Model 10 Stepping 0 AuthenticAMD ~1664
Mhz
BIOS-Version/-Datum Phoenix Technologies, LTD ASUS A7N8X-E Deluxe
ACPI BIOS Rev 1009, 04.02.2004
SMBIOS-Version 2.2
Gesamter realer Speicher 512,00 MB
Verfügbarer realer Speicher 348,77 MB
 
Hans J. Ude wrote:

... Can this be the
power supply? It has 300W by the way....
Click to expand...

Does your PSU fulfill the AMD recommendation, i.e.

+3.3V: 28A, +5V: 30A, +12V: 15/18A?

(Sorry, no direct link to the relevant white paper on www.amd.com at hand)

Roy
 
Hans J. Ude said:
I've bought a brand new A7N8X-E Deluxe witx XP 2800+ CPU and 512 MB
RAM from Empaq (PC 2700, 333Mhz, CL2.5). When I run it with 166 Mhz
clock it always powers down after a while. Sometimes that happens
after minutes, sometimes after hours but it _will_ happen. After that
it can't be restarted with the standby button at the front. I have to
switch main power off and on to get it restarted. Can this be the
power supply? It has 300W by the way. I'm clueless here. Does anyone
know about that problem? There is some extra info from sysinfo below.
Everything works at a clock rate of 133 MHz but I get only the
perfromance of a XP 2200 CPU. Sometimes power off happens even in CMOS
setup, so it can't be an OS issue.

TIA
Hans

Systemhersteller ASUSTeK Computer INC.
Systemmodell A7N8X-E
Systemtyp X86-basierter PC
Prozessor x86 Family 6 Model 10 Stepping 0 AuthenticAMD ~1664
Mhz
BIOS-Version/-Datum Phoenix Technologies, LTD ASUS A7N8X-E Deluxe
ACPI BIOS Rev 1009, 04.02.2004
SMBIOS-Version 2.2
Gesamter realer Speicher 512,00 MB
Verfügbarer realer Speicher 348,77 MB
Click to expand...

Enter the BIOS. Set the FSB to 166MHz, save the setting, and
let the computer POST again. Then, enter the BIOS again and
use the "Power Monitor" page. Watch the voltages and temperatures
and let the machine sit there until it shuts off. If the
CPU temperature is not shooting up, then it is probably the
power supply which is failing. If you see the CPU temperature
shoot up until the machine shuts off, then it could be the
heatsink isn't installed properly. Apparently you can rotate the
heatsink 180 degrees, and that won't work properly. Also, some
people jam a thermal probe under the heatsink, and that is
not a wise thing to do. The die should contact the heatsink
flush. Thermal paste can be used to improve the contact
between the die and the heatsink - use only a thin thin layer,
so it doesn't squirt out.

HTH,
Paul
 
Enter the BIOS. Set the FSB to 166MHz, save the setting, and
let the computer POST again. Then, enter the BIOS again and
use the "Power Monitor" page. Watch the voltages and temperatures
and let the machine sit there until it shuts off. If the
CPU temperature is not shooting up, then it is probably the
power supply which is failing. If you see the CPU temperature
shoot up until the machine shuts off, then it could be the
heatsink isn't installed properly. Apparently you can rotate the
heatsink 180 degrees, and that won't work properly. Also, some
people jam a thermal probe under the heatsink, and that is
not a wise thing to do. The die should contact the heatsink
flush. Thermal paste can be used to improve the contact
between the die and the heatsink - use only a thin thin layer,
so it doesn't squirt out.
Click to expand...

Thanks to all who answered. I've checked the temperatures and they
appear to be OK. CPU has 58°C/136°F. No reason to worry. Perhaps the
next step will be to downgrade from BIOS Rev 1009 back to 1008. I've
haven't read much good about 1009 in this forum. My board has been
delivered with 1009. If this doesn't help, I'm gonna by a new power
supply.

Hans
 
Roy Coorne said:
Does your PSU fulfill the AMD recommendation, i.e.

+3.3V: 28A, +5V: 30A, +12V: 15/18A?
Click to expand...

These are the specs of my PSU

+5V: 35A
+12V: 14A
-12V: 1A
-5V: 0.5A
3.3V: 24A

Should be OK, nope? My 12V is actually 12.9V which makes me worry a
bit. And it goes up to 13.3 when I run stress on the CPU. And then it
turns off.

Hans
 
Hans J. Ude said:
These are the specs of my PSU

+5V: 35A
+12V: 14A
-12V: 1A
-5V: 0.5A
3.3V: 24A

Should be OK, nope? My 12V is actually 12.9V which makes me worry a
bit. And it goes up to 13.3 when I run stress on the CPU. And then it
turns off.

Hans
Click to expand...

OK, the outputs of an ATX supply are tied together. When you see
one output rise under load, it means another output is falling
at the same time. So, one of the other rails is being loaded.
Now, the question would be, how is your +3.3V right now. Is
it much lower than +3.3 or not ? Maybe the PSU is just shutting
down on overheat or something. Is the PSU physically hot to the
touch ? Any signs it is feeling the stress ? You should quote
all voltages when you post them, as sometimes the values tell
you what is going on.

The way ATX works, for most older supplies, is the +3.3V has a
feedback wire that goes to the PSU. If you examine the ATX 20
pin connector, one pin (a +3.3V pin) will have two wires going
to it. The thinner wire is used by the PSU, to monitor how the
+3.3V is doing. The switcher inside the PSU is designed to
respond to a drop in output, of any of the outputs, but the
strongest response (tightest regulation) would be on +3.3V.
The switcher drives a transformer, and all the outputs of the
supply are wound on the same transformer. When the primary
voltage is increased, all secondary voltages increase. If the
switcher sees one of the outputs drop below normal, it cranks
up all outputs. So, when you tell me the +12V is at +13.3, it
means one of the other rails is being loaded, and when viewed
in the monitor, it will either have a normal value (like +4.9
say), or if the value is really low (say +4.7), the PSU might
be on the verge of shutting down.

Paul
 
Back
Top