If this is your first visit, be sure to check out the FAQ by clicking the link above. You may have to register before you can post: click the register link above to proceed. To start viewing messages, select the forum that you want to visit from the selection below. |
|
|
Thread Tools | Display Modes |
#1
|
|||
|
|||
CPUID HWMonitor
Hi,
CPUID HWMonitor warned me that the MOBO value of -12 V was -0256 V. Is this anything to sweat? |
#2
|
|||
|
|||
CPUID HWMonitor
Norm X wrote:
Hi, CPUID HWMonitor warned me that the MOBO value of -12 V was -0256 V. Is this anything to sweat? I would want to check that under a second or third tool. If you have a Linux DVD, you could try from that side of the fence. To measure -12V, the 0 to 4.095V input needs a scaling resistor. Say for example, I fitted a two resistor attenuator network that scaled the input by 1/4. The ADC would measure "3 volts". It's up to the HWMonitor program to multiply the value by 4/1 to get the correct voltage. (I presume the ADC has a sign bit but sometimes even that is fudged with an external bias network.) At one time, we only knew by trial and error, what the scaling resistors on each input were. But modern programs seem to be getting scaling info from the BIOS some how. I've not seen a description of the ACPI table that might be providing that info. I no longer see "lots of questions" about how a new motherboard behaves, so the info is coming from places other than end-user testing. The -12V is not critical to hardware operation. Typically, +12V and -12V are used by TI 75232 RS232 driver chips. You can get RS232 chips with charge pumps, which means a number of larger value ceramic capacitors to hold the charge that the pumps generate. But some motherboard makers still use +12V and -12V and conventional RS232 chips. That's about the only consumer of -12V I know of right off hand. Several generations of motherboard ago, there were a few OP Amps used for doing voltage regulation, and those could benefit from a negative swing supply. So there may be other fairly important things like that, needing a negative voltage, in which case you would not want the -12V going too nuts. The voltages on the ATX supply are established by "turns ratio". If your 3.3V, 5V, 12V are correct, it's pretty hard for the voltage to go "purely nuts". Some supplies shut down (PWR_GOOD de-asserted) if rails are out of spec, but I don't know if -12V is actually a term in PWR_GOOD and maybe -12V is being ignored. The main ATX connector is always available for probing with a $20 multimeter, which will give a means of verifying the values. The connector pinout can be seen on the Playtool site, about half way down the page. http://www.playtool.com/pages/psucon...onnectors.html You can ground the multimeter on an I/O plate screw for the black lead of the multimeter. Use the 20V full scale range. Then, use the red lead to check DC voltage. All voltages should be within 5% of nominal. The meter probe can be shoved into the plastic shell of the main connector, where the wire goes into the shell. The pin in there, has enough exposed metal, for you to be able to take a reading. By clipping the black meter probe onto the I/O plate as a ground screw, this prevents the meter probes from getting too close together and shorting out the ATX supply. If the supply is actually out of spec, I would change it out. That's if you want everything to work. On relatively modern motherboards, if there's no RS232, most of the onboard regulation tasks are done by dedicated chips (not cheesy OP Amp circuits), and really nothing else should depend on -12V. ******* Try reading it with Speedfan, which also prints the voltages. For best results, don't run two hw monitor programs at the same time. http://almico.com/sfdownload.php http://almico.com/speedfan452.exe Paul Paul |
#3
|
|||
|
|||
CPUID HWMonitor
On Sunday, June 23, 2019 at 7:01:33 PM UTC+8, Paul wrote:
I would want to check that under a second or third tool. Yeah, like the CPU temperature reported by memtest86+ seems to be rather too high. On one rig, it claimed 99 degrees C. I had set BIOS to beep beep at 80, but it wasn't triggered. Also trying lm_sensors with all cores loaded, it only went to high 60s. |
#4
|
|||
|
|||
CPUID HWMonitor
wrote in message
... On Sunday, June 23, 2019 at 7:01:33 PM UTC+8, Paul wrote: I would want to check that under a second or third tool. Yeah, like the CPU temperature reported by memtest86+ seems to be rather too high. On one rig, it claimed 99 degrees C. I had set BIOS to beep beep at 80, but it wasn't triggered. Also trying lm_sensors with all cores loaded, it only went to high 60s. Hey, Start with Real Temp 3.7X. It has sensor calibrations. When calibrated, temperatures seem reasonable and clearly show the benefit of better thermal grease. CPUID HWMonitor temperatures are then credible. BIOS temperatures are incomplete. I set my BIOS Temp alarm to 70 C. In one booboo, CPU use went to 110% and then the BIOS alarm sounded constantly. I needed to reset the BIOS, to return sanity to the PC. What does memtest86+ have to do with temperature? Your memory is worse than mine! I drive a used car out of economic necessity. Likewise my MOBO is 10+ years old but all parts are new. Works good. Recently, after much trial and tribulation, I came to the realization that I need a new CPU cooler. My auto needs a new PCV valve. Go figure. |
Thread Tools | |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
About CPUID, why is it serialization instruction? | Revan76 | Intel | 2 | February 8th 09 11:53 AM |
CPUID results help request. | John IV | Intel | 0 | August 10th 05 10:29 PM |
Does anybody know where to find the CPUID <=> CPU Name mapping ? | forfun | Intel | 2 | February 28th 05 03:36 PM |
P2B - "Update data incorrect.. CPUID =..." | Paul | Asus Motherboards | 7 | December 1st 04 05:02 AM |
Do I believe CPUID or SETUP? | Gene | Overclocking AMD Processors | 11 | January 21st 04 04:47 PM |