You are not logged in.

#1 2017-01-25 07:33:42

matmutant
Member
Registered: 2016-07-19
Posts: 39
Website

Fun fact: ACPI and Fan rpm

This not much of an issue, simply a fun fact,
On my EliteBook 2540p, I couldn't get the fan speed with sensors or whatever, I gave up.

Yesterday, something showed up: one of the acpi temperature sensors (temp3) was oscillating between 0 and 11°C whenever the fan was starting and stopping...

So I launched

sysbench --test=cpu --cpu-max-prime=200000 --num-threads=4  run

As soon as the fan speed increased (that you can definitely hear speeding up), that sensor got its values increasing by steps: 21, 37 and 53"°C" at max. (screenshot here)

psensor and fan rpm shown as T°C

This looks to me like either a speed factor or even the rpm lacking 3 zeros (1100, 2100, 3700, 5300rpm)

This PC might not be the only one reading fan speed as a temperature value, so one of you might be interested by this smile

Have a nice day!
m@


PS: my Toshiba A100 SSD temperature is either shown or not by psensor for no reason xD (one day it is, the following one it isn't)

Last edited by matmutant (2017-01-25 07:34:15)

Offline

#2 2017-01-25 07:42:33

Head_on_a_Stick
Member
From: London
Registered: 2015-09-29
Posts: 8,759
Website

Re: Fun fact: ACPI and Fan rpm

Great tip, thanks matmutant smile

I have always found lm-sensors to be a bit strange with it's output, did you calibrate it to your system?

sudo sensors-detect

FWIW, my ThinkPad X201 shows about 3 items with `sensors` in Linux but in OpenBSD:

SENSOR                                    VALUE  STATUS  DESCRIPTION
cpu0.temp0                           51.00 degC
acpitz0.temp0                        56.00 degC          zone temperature
acpibtn0.indicator0                          On          lid open
acpibat0.volt0                       11.10 V DC          voltage
acpibat0.volt1                       12.76 V DC          current voltage
acpibat0.current0                        0.00 A          rate
acpibat0.amphour0                       2.69 Ah          last full capacity
acpibat0.amphour1                       0.13 Ah          warning capacity
acpibat0.amphour2                       0.02 Ah          low capacity
acpibat0.amphour3                       2.67 Ah    OK    remaining capacity
acpibat0.amphour4                       6.22 Ah          design capacity
acpibat0.raw0                             2 raw    OK    battery charging
acpiac0.indicator0                           On          power supply
acpithinkpad0.fan0                     3283 RPM
acpidock0.indicator0                        Off unknown  not docked
itherm0.temp1                        53.00 degC          Core 1
itherm0.temp4                        57.00 degC          CPU/GPU Max temp
itherm0.temp9                        57.00 degC          GPU/Memory controller abs.
itherm0.temp10                       59.00 degC          PCH abs.
itherm0.power0                           6.00 W          CPU power consumption
aps0.temp0                           38.00 degC
aps0.temp1                           38.00 degC
aps0.indicator0                             Off          Keyboard Active
aps0.indicator1                              On          Mouse Active
aps0.indicator2                              On          Lid Open
aps0.raw0                               503 raw          X_ACCEL
aps0.raw1                               528 raw          Y_ACCEL
aps0.raw2                               503 raw          X_VAR
aps0.raw3                               528 raw          Y_VAR

... even the accelerometer readings show up cool


“Et ignotas animum dimittit in artes.” — Ovid, Metamorphoses, VIII., 18.

Forum Rules   •   How to report a problem   •   Software that rocks

Offline

#3 2017-01-25 10:42:45

matmutant
Member
Registered: 2016-07-19
Posts: 39
Website

Re: Fun fact: ACPI and Fan rpm

Yes I did use sensors-detect (before it wouldn't bother to give me anything btw)

matmutant@BunsenElite:~$ sensors
acpitz-virtual-0
Adapter: Virtual device
temp1:         +0.0°C  (crit = +127.0°C)
temp2:         +0.0°C  (crit = +127.0°C)
temp3:        +11.0°C  (crit = +128.0°C)
temp4:        +37.0°C  (crit = +127.0°C)
temp5:         +0.0°C  (crit = +115.0°C)
temp6:        +29.0°C  (crit = +128.0°C)
temp7:         +0.0°C  (crit = +128.0°C)
temp8:        +39.0°C  (crit = +128.0°C)
temp9:        +47.0°C  (crit = +128.0°C)
temp10:       +59.0°C  (crit = +128.0°C)

coretemp-isa-0000
Adapter: ISA adapter
Core 0:       +39.0°C  (high = +95.0°C, crit = +105.0°C)
Core 2:       +37.0°C  (high = +95.0°C, crit = +105.0°C)

There are still many of those "temperature" sensors that I either don't know what they are for or even if they are really a Thermal probe or something else (I'm pretty sure one of them is the ambient light sensor mislabeled)

This PC is still a big mystery to me...  on many, many  points xD

Offline

#4 2017-01-25 18:14:14

martix
Kim Jong-un Stunt Double
Registered: 2016-02-19
Posts: 1,267

Re: Fun fact: ACPI and Fan rpm

Yeah, that fan is kind of tricky on linux...  https://ubuntuforums.org/showthread.php?t=2121339

Do you maybe get a fan-speed by using this command (thanks to Sector11 for mentioning this useful code)?

sensors && echo =-=-=-=-=-=-= && inxi -s

Offline

#5 2017-01-25 18:33:24

unklar
Member
Registered: 2015-10-31
Posts: 1,255

Re: Fun fact: ACPI and Fan rpm

sensors -u

Is also not bad.   big_smile

Offline

#6 2017-01-26 14:26:49

matmutant
Member
Registered: 2016-07-19
Posts: 39
Website

Re: Fun fact: ACPI and Fan rpm

yeah, that would work if inxi gave me anything but N/A for fan speed xD

inxi -s
Sensors:   System Temperatures: cpu: 39.0C mobo: 0.0C 
           Fan Speeds (in rpm): cpu: N/A 

Offline

#7 2017-01-27 19:01:39

o9000
tint2 developer
From: Network Neighborhood
Registered: 2015-10-24
Posts: 401
Website

Re: Fun fact: ACPI and Fan rpm

I'm happy that HP doesn't make cars. Or airplanes.

Offline

#8 2017-01-27 19:17:43

matmutant
Member
Registered: 2016-07-19
Posts: 39
Website

Re: Fun fact: ACPI and Fan rpm

o9000 wrote:

I'm happy that HP doesn't make cars. Or airplanes.

I'm starting to think that the fan speed is driven directly by hardware, doesn't seem that system is capable of doing anything on it (and there are some bios parameter of the type "fan always ON on AC")

Offline

#9 2017-01-27 19:30:49

o9000
tint2 developer
From: Network Neighborhood
Registered: 2015-10-24
Posts: 401
Website

Re: Fun fact: ACPI and Fan rpm

That sounds smart, I used to have an HP laptop where the fan was controlled by software. Unfortunately the sensor would lock up occasionally, so the fan wouldn't start, or it would stay on low although the CPU was frying the dust inside the heatsink, letting out burning smells... all while the temperature reading was stuck at 30C or so. An X11 restart would unlock it though. Ctrl+Alt+Backspace was my friend, it used to be enabled by default at the time, I remember the uproar when that was changed. Good times.

Offline

#10 2017-01-28 11:41:28

martix
Kim Jong-un Stunt Double
Registered: 2016-02-19
Posts: 1,267

Re: Fun fact: ACPI and Fan rpm

On a gentoo forum I read that HP stands for "Heat Problems".  smile

Anyhow it was also mentioned there that TZ 2 has the following steps: 0, 10, 21, 37, 53, 70.

On windows hwinfo shows the correct fan rpm values.

Offline

#11 2017-01-29 21:09:05

matmutant
Member
Registered: 2016-07-19
Posts: 39
Website

Re: Fun fact: ACPI and Fan rpm

martix wrote:

On a gentoo forum I read that HP stands for "Heat Problems".  smile

Anyhow it was also mentioned there that TZ 2 has the following steps: 0, 10, 21, 37, 53, 70.

On windows hwinfo shows the correct fan rpm values.

Do you know if those steps are arbitrary steps or rpm factor?

btw, I booted the PC by 5°C outside today, and acpi temp 10 was still around 59°C (contrary to other temp sensors that were showing less than 20°C (CPU cores at least)

So I really doubt this temp 10 really is temperature either xD

Offline

#12 2017-03-28 22:26:11

martix
Kim Jong-un Stunt Double
Registered: 2016-02-19
Posts: 1,267

Re: Fun fact: ACPI and Fan rpm

@matmutant Sorry, somehow I missed the question.

I suspect that the core of the problem might be a kernel issue, giving those values instead of the correct rpm results. The steps 0, 10, 21, etc. do not refer to temperature as you mentioned. These are more likely the values instead of the correct rpm numbers. Someone with more kernel knowledge might shed light on the cause for the strange rpm results.

Offline

Board footer

Powered by FluxBB