[lm-sensors] [PATCH v4 00/10] hwmon: (w83627ehf) Add support for NCT6775F and NCT6776F
psusi at cfl.rr.com
Wed Feb 23 16:38:05 CET 2011
On 2/23/2011 12:53 AM, Guenter Roeck wrote:
> Actually, sysfanout (pwm1) can be configured for DC output. You should see
> the setting with pwm1_mode. If its value is 0, output is DC, otherwise
> it is pwm. pwm2 and pwm3 only support pwm mode.
> If configured for DC, the upper 6 bit of pwmX translate to output voltage.
> Motherboard pinout (per its datasheet) is a bit odd. Looks like only the CPU fan
> (pwm2) supports pwm mode. The other fan connectors don't have a pwm pin.
> Fan connectors are labeled PWR_FAN1, CPU_FAN, CHA_FAN1 and CHA_FAN2. CHA_FAN1
> has a 4-pin header, but the 4th pin is labeled as "+5V", not pwm.
That is odd. The 4th pin should either be pwm or no connect, not +5V.
>>> Would it be possible to run pwmconfig and let us know what it reports ?
>> It appears to agree with my assessment in that fan1 responds to pwm1.
> And pwm2 and pwm3 don't have an effect ?
> Also - just trying to make sure - if you change pwm1, does it have an effect on
> pwm2 and/or pwm3 ? It should not, but who knows. And what are the values reported
> by pwm2 and pwm3 ?
I was just playing directly with the knobs in /sys rather than running
sensors or pwmconfig. Setting pwm2 and 3 to zero had no effect.
Setting pwm1 to zero caused fan1 and I think it was fan4 to drop to 0.
fan2 and fan3 were still reading as spinning, though I am pretty sure
that only one fan was actually still spinning. I'll open the case
tonight and figure out which fan was connected to which header.
More information about the lm-sensors