[lm-sensors] ASUS P5B DELUXE WiFi with Winbond W83627DHG

[lm-sensors] ASUS P5B DELUXE WiFi with Winbond W83627DHG

[Hamlet]

Winbond W83627DHG seems to be compatible with lm78

# sensors
lm78-isa-0290
Adapter: ISA adapter
VCore 1:   +2.43 V  (min =  +2.85 V, max =  +3.15 V)
VCore 2:   +3.66 V  (min =  +2.85 V, max =  +3.15 V)
+3.3V:     +3.31 V  (min =  +3.14 V, max =  +3.47 V)
+5V:       +5.56 V  (min =  +4.76 V, max =  +5.24 V)
+12V:     +12.77 V  (min = +11.37 V, max = +12.59 V)
-12V:     -11.24 V  (min = -12.63 V, max = -11.40 V)
-5V:       -4.98 V  (min =  -5.25 V, max =  -4.74 V)
fan1:        0 RPM  (min = 337500 RPM, div = 2)
fan2:     20149 RPM  (min = 10074 RPM, div = 1)
fan3:        0 RPM  (min = 112500 RPM, div = 2)
temp:      +41.0?C  (high =   +81?C, hyst =  -124?C

# isadump 0x295 0x296
WARNING! Running this program can cause system crashes, data loss and worse!
I will probe address register 0x295 and data register 0x296.
Continue? [Y/n]
     0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f
00: 04 ff 04 4b 31 00 4c 10 01 4b 01 01 3c 3c 0a 05
10: 04 ff 50 00 00 01 01 3c 43 17 00 00 ff ff ff c7
20: 98 e5 cf cf d2 ca cf 29 ff 43 ff c5 b2 c5 b2 d9
30: c4 c3 b1 cf bb e3 cd da c5 51 84 02 86 06 85 ff
40: 03 00 00 de ff ff 00 15 2d 00 20 44 18 95 06 a3
50: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
60: 01 9b 31 56 9b 01 3c ff 12 ff 0a ff ff ff ff ff
70: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
80: 04 ff 04 4b 31 00 4c 10 01 4b 01 01 3c 3c 0a 05
90: 04 ff 50 00 00 01 01 3c 43 17 00 00 ff ff ff c7
a0: 98 e5 cf cf d2 ca cf 29 ff 43 ff c5 b2 c5 b2 d9
b0: c4 c3 b1 cf bb e3 cd da c5 51 84 02 86 06 85 ff
c0: 03 00 00 de ff ff 00 15 2d 00 20 44 18 95 06 a3
d0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
e0: 01 9b 31 56 9b 01 3c ff 12 ff 0a ff ff ff ff ff
f0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff

I will stay with lm78 because w83627ehf doesn't work, at least for now.

http://forums.gentoo.org//viewtopic-t-508476.html

[lm-sensors] ASUS P5B DELUXE WiFi with Winbond W83627DHG

[Linus Wenell]

On Wed, 18 Oct 2006 15:55:10 -0300
Hamlet <hamletmun at fibertel.com.ar> wrote:

> Winbond W83627DHG seems to be compatible with lm78

well, no not really.. the temp you get is MB temp, the cpu temp can't be read.. and the voltages are all messed up too.. 

[lm-sensors] ASUS P5B DELUXE WiFi with Winbond W83627DHG

[David Hubbard]

Hello,

I am working on the DHG driver, and a prototype seems to work OK but
we're still working on the version that will go into the mainline
kernel. I apologize for not having it ready yet. Would you be willing
to test it when we are ready?

David

On 10/18/06, Linus Wenell <zidern at gmail.com> wrote:
> On Wed, 18 Oct 2006 15:55:10 -0300
> Hamlet <hamletmun at fibertel.com.ar> wrote:
>
> > Winbond W83627DHG seems to be compatible with lm78
>
> well, no not really.. the temp you get is MB temp, the cpu temp can't be read.. and the voltages are all messed up too..

[lm-sensors] ASUS P5B DELUXE WiFi with Winbond W83627DHG

[David Hubbard]

The patch will be for 2.6.19-rc2 or whichever kernel is the latest
version. You should probably build a new kernel, as one of the things
we would like to test is whether the patch compiles correctly or not.

I'll definitely keep you posted.

David

On 10/18/06, Linus Wenell <zidern at gmail.com> wrote:
> I'd be happy to! which kernel version are you working at? or will there be a patch for any kernel? I'm using 2.6.18 at the moment, with a bunch of patches..
>
> /Zider

[lm-sensors] ASUS P5B DELUXE WiFi with Winbond W83627DHG

[David Hubbard]

Hi Christian,

On 10/28/06, Christian Mahr <christian.mahr.ulm at arcor.de> wrote:
> Hi David
>
> I have the same board ASUS P5B Deluxe wifi for 1 week now. The one with the
> W83627DHG chip
>
> I?d like to give the modified driver for w83627 a try - if you don?t mind.
> Currently I am running a 2.6.18.1 kernel (already withe coretemp patches) but
> I can also switch to a more recent one if required.

There is this file which Michael Nelson used on his 2.6.18 kernel successfully:
http://lists.lm-sensors.org/pipermail/lm-sensors/attachments/20060906/b139bc4d/attachment-0001.obj

It is not a part of the kernel yet. In the thread where Michael tested
it, we discussed the changes that need to be made. When a version is
ready for the kernel, would you be willing to test it also?

David

[lm-sensors] ASUS P5B DELUXE WiFi with Winbond W83627DHG

[David Hubbard]

Hi Christian,

> Of yourse I can also test any newer version.
> I understand you want to make a separate driver for the W83627DHG for some
> 2.6.19-RC-xx?

To be exact, the driver I sent you is for 2.6.18 because it is that
old. *sigh* Yes, the new driver will show up in 2.6.19-RC-xx. It will
be an integrated driver for the EHF/EHG and the DHG.

David

[lm-sensors] ASUS P5B DELUXE WiFi with Winbond W83627DHG

[David Hubbard]

Hi Christian,

Please include lm-sensors at lm-sensors.org in your reply so this will
benefit others and get archived on the LM-sensors list.

On 10/29/06, Christian Mahr <christian.mahr.ulm at arcor.de> wrote:
> Hi David,
>
> Thanks again for the driver. So far I can judge - it works!

That's good.

> I am still studying the result.
>
> Still guessing how the chip is used on the Motherboard - do you have any
> documentation on the chip itself to support my guessing?
>
> Some of my ovservations so far:
>
> 1. Fans
>
>  I can observe the 3 fans connect to my board (did not chek the rest).
> Anyhow the sequence is quite different to the numbering on the mother board.

That is normal.

> I adapted /etc/sensors.conf so I can see the 3 fans properly:

Good. Every motherboard may be different. That's the purpose of
/etc/sensors.conf.

> fan1: 948 RPM (min = 803 RPM, div = 16) also named CHA_Fan1
>
> fan4: 1004 RPM (min = 811 RPM, div = 16) named CHA_Fan2
>
> fan5: 819 RPM (min = 602 RPM, div = 16) named CHA_Fan3
>
> (I do not use the cpu fan, chasis only)

So, are you saying the CPU fans are not connected?

> The fans (mine are quite slow) need half a minute to settle. During this
> time the "div" (which I did not set) seems to vary.

Yes. The div is automatically set. While it is settling, the div may
have incorrect values and your fans will read incorrect values (0, if
they are running slowly).

> 2. Voltages
>
> I have no clear idea what I do measure on which input. Here some
> documentation could help. f.e. on the voltage ratio between measured and
> reasl +5V seems to be a factor of 3.16
>
>
>
> Current output is:
>
>
>
> w83627ehf-isa-0290
>
> Adapter: ISA adapter
>
> VCore: +1.17 V (min = +0.00 V, max = +1.74 V)
>
> +12V: +12.04 V (min = +10.82 V, max = +13.20 V)
>
> AVCC: +3.23 V (min = +3.14 V, max = +3.47 V)
>
> 3VCC: +3.23 V (min = +3.14 V, max = +3.47 V)
>
> in4: +1.46 V (min = +1.29 V, max = +1.58 V)
>
> in5: +1.58 V (min = +1.42 V, max = +1.74 V)
>
> +5V: +5.07 V (min = +4.75 V, max = +5.24 V)
>
> VSB: +3.23 V (min = +3.14 V, max = +3.47 V)
>
> VBAT: +3.18 V (min = +3.14 V, max = +3.47 V)
>
>
>
> So on some voltages I am just guessing....

The voltage scaling is done externally and although the datasheet
specifies the resistor ratio, it is possible that the motherboard
manufacturer has wired it differently. The way to check the voltages
is to look in the BIOS. Can you see all the voltages listed there?

> 3. Temperatures
>
>
>
> I can see 2 out of 3 valid temperatures -
>
> Sys Temp: +36?C (high = +45?C, hyst = +40?C)
>
> CPU Temp: +35.0?C (high = +45.0?C, hyst = +40.0?C)
>
> AUX Temp: +119.0?C (high = +80.0?C, hyst = +75.0?C) ALARM
>
>
>
> The 3rd temperature sensoris obviously not used (says 119 degrees)
>
> But I have no real Idea what the 2 sensors are doing.

The AUX temp may need to be configured correctly. Can you see any
variation? It may also be that the pin on the chip is not connected to
anything, and just floating.

> 4. in /sys
>
> ther is nothing in /proc/ACPI/fan, but I have observed the new device in
> /sys/devices/platform/i2c-9191/9191-0290:
>
>
>
> . fan3_min in1_input in5_alarm in8_min pwm2_target pwm4_tolerance
>
> .. fan4_alarm in1_max in5_input name pwm2_tolerance subsystem
>
> bus fan4_div in1_min in5_max power pwm3 temp1_alarm
>
> driver fan4_input in2_alarm in5_min pwm1 pwm3_enable temp1_input
>
> fan1_alarm fan4_min in2_input in6_alarm pwm1_enable pwm3_min_output
> temp1_max
>
> fan1_div fan5_alarm in2_max in6_input pwm1_min_output pwm3_mode
> temp1_max_hyst
>
> fan1_input fan5_div in2_min in6_max pwm1_mode pwm3_stop_time temp2_alarm
>
> fan1_min fan5_input in3_alarm in6_min pwm1_stop_time pwm3_target temp2_input
>
> fan2_alarm fan5_min in3_input in7_alarm pwm1_target pwm3_tolerance temp2_max
>
> fan2_div hwmon:hwmon2 in3_max in7_input pwm1_tolerance pwm4 temp2_max_hyst
>
> fan2_input in0_alarm in3_min in7_max pwm2 pwm4_enable temp3_alarm
>
> fan2_min in0_input in4_alarm in7_min pwm2_enable pwm4_min_output temp3_input
>
> fan3_alarm in0_max in4_input in8_alarm pwm2_min_output pwm4_mode temp3_max
>
> fan3_div in0_min in4_max in8_input pwm2_mode pwm4_stop_time temp3_max_hyst
>
> fan3_input in1_alarm in4_min in8_max pwm2_stop_time pwm4_target uevent
>
>
>
>
>
> and I can get the raw values out of fan* amd in*, which seem to be matching
> the "sensors" output.

Yes, sensors is just accessing the files in the sysfs.

> My question: how do I operate the "pwm*" ports? Am I right that I can
> control the fan speed here?
>
> How to do? Which values to echo in?

There is documentation in the kernel source under
linux/documentation/hwmon. Basically, set pwmN_enable to 1 for manual
control, 2 for automatic "thermal cruise" or "SmartFan 1". In manual
mode, echo "0" > pwmN stops the fan, and echo "255" > pwmN to run the
fan at full speed.

> If you have a data sheet in pdf of this chip with a "recommended
> application" or the like - this might help me.

I do have a copy of the datasheet from Winbond, but it is not
currently available to the public. A "recommended application"? I am
not sure how to help you there. If the values reported look correct,
then it would seem the driver is working.

Hope that helps,
David

[lm-sensors] ASUS P5B DELUXE WiFi with Winbond W83627DHG

[Christian Mahr]

Hi David,
as you wished - reply goes also to <lm-sensors at lm-sensors.org>.
Please see my answers/notes below.
Maybe other owners of P5B Deluxe WiFi are interested what sensor readings really do mean for this board - or maybe somebody can give me a hint on what they found out already?
Thanks
	Christian



Am Sonntag, 29. Oktober 2006 21:33 schrieb David Hubbard:
> Hi Christian,
>
> Please include lm-sensors at lm-sensors.org in your reply so this will
> benefit others and get archived on the LM-sensors list.
>
> On 10/29/06, Christian Mahr <christian.mahr.ulm at arcor.de> wrote:
> > Hi David,
> >
> > Thanks again for the driver. So far I can judge - it works!
>
> That's good.
>
> > I am still studying the result.
> >
> > Still guessing how the chip is used on the Motherboard - do you have any
> > documentation on the chip itself to support my guessing?
> >
> > Some of my ovservations so far:
> >
> > 1. Fans
> >
> >  I can observe the 3 fans connect to my board (did not chek the rest).
> > Anyhow the sequence is quite different to the numbering on the mother
> > board.
>
> That is normal.
>
> > I adapted /etc/sensors.conf so I can see the 3 fans properly:
>
> Good. Every motherboard may be different. That's the purpose of
> /etc/sensors.conf.
>
> > fan1: 948 RPM (min = 803 RPM, div = 16) also named CHA_Fan1
> >
> > fan4: 1004 RPM (min = 811 RPM, div = 16) named CHA_Fan2
> >
> > fan5: 819 RPM (min = 602 RPM, div = 16) named CHA_Fan3
> >
> > (I do not use the cpu fan, chasis only)
>
> So, are you saying the CPU fans are not connected?
>
Sorry, I should have explained. The CPU-Fan connector on the P5BDeluxe Wifi is a 4-pin, but my CPU-Fan itself has only 3 pins. As I tried to use the marked 3 pins(the 3-pin connector still fits into the 4 pin) I could not control the speed via the Bios (this is probably the purpose of the 4th pin on a 4 pin connector?) --> it stayed on a somewhat noisy speed.
So I decided to use one of the 3-pin-CHA1_FAN-connector (all other Fan connectors do have 3 pins) for the CPU-fan and tried this one instead --> it works. All calm cool and quiet now.

> > The fans (mine are quite slow) need half a minute to settle. During this
> > time the "div" (which I did not set) seems to vary.
>
> Yes. The div is automatically set. While it is settling, the div may
> have incorrect values and your fans will read incorrect values (0, if
> they are running slowly).

This is pretty neat as the Bios apparently cannot handle this correctly. If the speed is below, say, 700 rpm or so, the Bios (when the "FAN page" is opened) becomes very slow and behaves strange sometimes.
Is this a "auto-div" function of the chip or the driver?
Does this mean you do not need to set it in the sensors.conf ? This should be mentioned later specifically in the comments in sensors.conf, isn?t it?

>
> > 2. Voltages
> >
> > I have no clear idea what I do measure on which input. Here some
> > documentation could help. f.e. on the voltage ratio between measured and
> > reasl +5V seems to be a factor of 3.16
> >
> >
> >
> > Current output is:
> >
> >
> >
> > w83627ehf-isa-0290
> >
> > Adapter: ISA adapter
> >
> > VCore: +1.17 V (min = +0.00 V, max = +1.74 V)
> >
> > +12V: +12.04 V (min = +10.82 V, max = +13.20 V)
> >
> > AVCC: +3.23 V (min = +3.14 V, max = +3.47 V)
> >
> > 3VCC: +3.23 V (min = +3.14 V, max = +3.47 V)
> >
> > in4: +1.46 V (min = +1.29 V, max = +1.58 V)
> >
> > in5: +1.58 V (min = +1.42 V, max = +1.74 V)
> >
> > +5V: +5.07 V (min = +4.75 V, max = +5.24 V)
> >
> > VSB: +3.23 V (min = +3.14 V, max = +3.47 V)
> >
> > VBAT: +3.18 V (min = +3.14 V, max = +3.47 V)
> >

reference reading in /sys/..../in*input	just right now:
1168
1824
3232
3232
1456
1584
1400
3232
3184

versus "sensors interpretation" (some seconds later)
VCore:     +1.16 V  		this seems smewhat low as BIOS says 1.296
+12V:     +12.04 V		this seems smewhat high as BIOS says 11,985
AVCC:      +3.23 V 		matches the bios readin of 3.216 Volts
3VCC:      +3.22 V  
in4:       +1.45 V 		
in5:       +1.58 V  
+5V:       +5.07 V 		Bios says 5.068, but here I adjusted the ratio for a fit.
VSB:       +3.23 V  
VBAT:      +3.18 V  
So Bios shows only 4 voltages out of the 8 sensors that are available!

Interestingly enough one can set some more Voltages when performing overclocking (I don?t dare to). 
The manual says, one can set 6 voltages on this board:
	memory voltage[1.80..2.45]
	CPU-Vcore         [1.275..1,7V], 
	FSB-voltage       [1.2 .. 1.45V]
	NB-Vcore	          [1.25..1,55]
	SB-vcore	          [1.5..1,8]
	ICH-voltage      [1.057 and 1,215V]

Since the Bios does not show these (except Vcore), these seem to be "write-only".
Nevertheless it could be that in4 and in5 relate to one of those values?
Maybe somebody on the list knows?


> >
> >
> > So on some voltages I am just guessing....
>
> The voltage scaling is done externally and although the datasheet
> specifies the resistor ratio, it is possible that the motherboard
> manufacturer has wired it differently. The way to check the voltages
> is to look in the BIOS. Can you see all the voltages listed there?
>
see above....
I guess the load situation is Bios is different to having Linux started?

> > 3. Temperatures
> >
> >
> >
> > I can see 2 out of 3 valid temperatures -
> >
> > Sys Temp: +36?C (high = +45?C, hyst = +40?C)
> >
> > CPU Temp: +35.0?C (high = +45.0?C, hyst = +40.0?C)
> >
> > AUX Temp: +119.0?C (high = +80.0?C, hyst = +75.0?C) ALARM
> >
> >
> >
> > The 3rd temperature sensoris obviously not used (says 119 degrees)
> >
> > But I have no real Idea what the 2 sensors are doing.
>
Interestingly enough the "sys" is very close to "CPU" in temperature wheras Bios shows 7 degrees more for the CPU (???).
Question: 
Which CPU sensor is meant here? The Core2Duo has got 2 of them internally (see coretemp driver) - is this a 3rd one?


> The AUX temp may need to be configured correctly. Can you see any
> variation? It may also be that the pin on the chip is not connected to
> anything, and just floating.
>
> > 4. in /sys
> >
> > ther is nothing in /proc/ACPI/fan, but I have observed the new device in
> > /sys/devices/platform/i2c-9191/9191-0290:
> >
> >
> >
> > . fan3_min in1_input in5_alarm in8_min pwm2_target pwm4_tolerance
> >
> > .. fan4_alarm in1_max in5_input name pwm2_tolerance subsystem
> >
> > bus fan4_div in1_min in5_max power pwm3 temp1_alarm
> >
> > driver fan4_input in2_alarm in5_min pwm1 pwm3_enable temp1_input
> >
> > fan1_alarm fan4_min in2_input in6_alarm pwm1_enable pwm3_min_output
> > temp1_max
> >
> > fan1_div fan5_alarm in2_max in6_input pwm1_min_output pwm3_mode
> > temp1_max_hyst
> >
> > fan1_input fan5_div in2_min in6_max pwm1_mode pwm3_stop_time temp2_alarm
> >
> > fan1_min fan5_input in3_alarm in6_min pwm1_stop_time pwm3_target
> > temp2_input
> >
> > fan2_alarm fan5_min in3_input in7_alarm pwm1_target pwm3_tolerance
> > temp2_max
> >
> > fan2_div hwmon:hwmon2 in3_max in7_input pwm1_tolerance pwm4
> > temp2_max_hyst
> >
> > fan2_input in0_alarm in3_min in7_max pwm2 pwm4_enable temp3_alarm
> >
> > fan2_min in0_input in4_alarm in7_min pwm2_enable pwm4_min_output
> > temp3_input
> >
> > fan3_alarm in0_max in4_input in8_alarm pwm2_min_output pwm4_mode
> > temp3_max
> >
> > fan3_div in0_min in4_max in8_input pwm2_mode pwm4_stop_time
> > temp3_max_hyst
> >
> > fan3_input in1_alarm in4_min in8_max pwm2_stop_time pwm4_target uevent
> >
> >
> >
> >
> >
> > and I can get the raw values out of fan* amd in*, which seem to be
> > matching the "sensors" output.
>
> Yes, sensors is just accessing the files in the sysfs.
>
> > My question: how do I operate the "pwm*" ports? Am I right that I can
> > control the fan speed here?
> >
> > How to do? Which values to echo in?
>
> There is documentation in the kernel source under
> linux/documentation/hwmon. Basically, set pwmN_enable to 1 for manual
> control, 2 for automatic "thermal cruise" or "SmartFan 1". In manual
> mode, echo "0" > pwmN stops the fan, and echo "255" > pwmN to run the
> fan at full speed.

Thanks , I should have known to look there
I?ll try.


>
> > If you have a data sheet in pdf of this chip with a "recommended
> > application" or the like - this might help me.
>
> I do have a copy of the datasheet from Winbond, but it is not
> currently available to the public. A "recommended application"? I am
> not sure how to help you there. If the values reported look correct,
> then it would seem the driver is working.
>
> Hope that helps,
> David
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.lm-sensors.org/pipermail/lm-sensors/attachments/20061029/4fb4a583/attachment-0001.html 

[lm-sensors] ASUS P5B DELUXE WiFi with Winbond W83627DHG

[Christian Mahr]

Hi David

a small note on fan control below.

Best regards
	Christian

Am Sonntag, 29. Oktober 2006 21:33 schrieben Sie:
> >
> > How to do? Which values to echo in?
>
> There is documentation in the kernel source under
> linux/documentation/hwmon. Basically, set pwmN_enable to 1 for manual
> control, 2 for automatic "thermal cruise" or "SmartFan 1". In manual
> mode, echo "0" > pwmN stops the fan, and echo "255" > pwmN to run the
> fan at full speed.
>


I did some study of the documentation, assuming that the notes for either 
w83792d or w83627hf do apply whatever sounds more likely - no precise info on w83627HDG ...

fan connections
For me, the relatationship between the 5 fan tacho reads and the 4 pwm-control ports are somewhat unclear to me: All "chassis" type fans I have installed (1,4,5) seem to be connected to pwm4 ? I cannot control them individually, they all move in sync.

pwm*enable
- I can read value "4" from pwm*_enable after having set "Silent" in Bios. Is this out of range? which values are allowe 0..3?

- When I try to set values, I can set "1" and "2" but neither "0" for "off" or "4" for "Thermal cruise" (whatever this is).

The driver prevents this. Please have a look to line 920 ff:

static ssize_t
store_pwm_enable(struct device *dev, struct device_attribute *attr,
			const char *buf, size_t count)
{
	struct i2c_client *client = to_i2c_client(dev);
	struct w83627ehf_data *data = i2c_get_clientdata(client);
	struct sensor_device_attribute *sensor_attr = to_sensor_dev_attr(attr);
	int nr = sensor_attr->index;
	u32 val = simple_strtoul(buf, NULL, 10);
	u16 reg;

	if (!val || (val > 2))	/* only modes 1 and 2 are supported */
		return -EINVAL;
	mutex_lock(&data->update_lock);
	reg = w83627ehf_read_value(client, W83627EHF_REG_PWM_ENABLE[nr]);
	data->pwm_enable[nr] = val;
	reg &= ~(0x03 << W83627EHF_PWM_ENABLE_SHIFT[nr]);
	reg |= (val - 1) << W83627EHF_PWM_ENABLE_SHIFT[nr];
	w83627ehf_write_value(client, W83627EHF_REG_PWM_ENABLE[nr], reg);
	mutex_unlock(&data->update_lock);
	return count;
}

Can you check in the data sheet for W83627DHG whether its correct to expand the range to 0..3 or whatever suitable? will the shift masks be still correct also for the  W83627DHG ?

1=Manual mode
when I set  pwm4_enable to 1 I can control all 3 fan speeds simultanously by  "echo xx >pwm4" 
2= Smartfan 1 mode
this brings the fans to such a low speed that the fan reading becomes jumping back and forth -  fan registers showing erraneously either zero or 5000...15000 rpm
I am still experimenting here, to lower the CPU temp (pwm-target?) to check 
 pwm*-registers
pwm*min_output:  purpose seems unclear to me. seems to work as lower fan limit value in case of smartfan mode1. no function in manual mode.
pwm*_stop_time : no idea
pwm*_target : seems to compare to temp2_input when in smartfan 1 mode
pwm*_tolerance  : hysteresis for pwm_target?
... my studies so far....
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.lm-sensors.org/pipermail/lm-sensors/attachments/20061031/a5b50538/attachment.html 

[lm-sensors] ASUS P5B DELUXE WiFi with Winbond W83627DHG

[Christian Mahr]

..2nd try

Hi David

I am still researching on the fans on the asus board. 

One problem I came across:
for fan5 the reading drops sharply from about 1325 rpm to zero when I lower 
the voltage. Note the fan is definitely moving still, the fan physicall can 
do much slower.

fan5_div shows 4 and does not step up as with fan 1 and 4 where I can see _div 
on 128.

fans 1,4 and 5 seem all to be controlled by pwm4 on this board.

to my understanding fan5 should behave as the others and step up the divider 
when lowering the voltage, isn?t it?

Best regards
????????Christian
Am Montag, 30. Oktober 2006 00:12 schrieb David Hubbard:
> Hi Christian,
>
> > Sorry, I should have explained. The CPU-Fan connector on the P5BDeluxe
> > Wifi is a 4-pin, but my CPU-Fan itself has only 3 pins. As I tried to use
> > the marked 3 pins(the 3-pin connector still fits into the 4 pin) I could
> > not control the speed via the Bios (this is probably the purpose of the
> > 4th pin on a 4 pin connector?) --> it stayed on a somewhat noisy speed.
> >
> > So I decided to use one of the 3-pin-CHA1_FAN-connector (all other Fan
> > connectors do have 3 pins) for the CPU-fan and tried this one instead -->
> > it works. All calm cool and quiet now.
>
> Okay, your sensors.conf will be different from the "standard" one for
> this motherboard, because you map the CPU fan to a different fan
> number.
>
> > This is pretty neat as the Bios apparently cannot handle this correctly.
> > If the speed is below, say, 700 rpm or so, the Bios (when the "FAN page"
> > is opened) becomes very slow and behaves strange sometimes.
> >
> > Is this a "auto-div" function of the chip or the driver?
> >
> > Does this mean you do not need to set it in the sensors.conf ? This
> > should be mentioned later specifically in the comments in sensors.conf,
> > isn?t it?
>
> You don't need to set the divs. In fact, I don't think you can. The
> driver does the div scaling.
>
> > Interestingly enough the "sys" is very close to "CPU" in temperature
> > wheras Bios shows 7 degrees more for the CPU (???).
>
> The "sys" temperature is measured inside the W83627DHG. You can test
> this, if you like, by chilling a large metal object and then holding
> it to the chip. But your chip may be integrated with the rest of the
> chipset. Your mileage may vary.
>
> > Question:
> >
> > Which CPU sensor is meant here? The Core2Duo has got 2 of them internally
> > (see coretemp driver) - is this a 3rd one?
> >
> > > The AUX temp may need to be configured correctly. Can you see any
> > >
> > > variation? It may also be that the pin on the chip is not connected to
> > >
> > > anything, and just floating.
>
> The AUX temperature input is a pin on the W83627DHG which measures the
> voltage from a thermal diode or thermistor. It is up to the
> motherboard manufacturer to decide whether to connect it to a sensor,
> or leave it floating. I don't know what it is measuring. The coretemp
> driver measures the temperature of your CPU using a completely
> different interface, and is not related to temperatures reported by
> the W83627DHG.
>
> David



	
		
___________________________________________________________ 
Der fr?he Vogel f?ngt den Wurm. Hier gelangen Sie zum neuen Yahoo! Mail: http://mail.yahoo.de

[lm-sensors] ASUS P5B DELUXE WiFi with Winbond W83627DHG

[Christian Mahr]

Hi David,

I found out: if the lower fan limit is too high, i do have this effect.
when I lowered the _min value. it works correctly.

.. still working on how to control fan 2 and 3 .... does not work yet

christian


Am Donnerstag, 2. November 2006 11:00 schrieb Christian Mahr:
> ..2nd try
>
> Hi David
>
> I am still researching on the fans on the asus board.
>
> One problem I came across:
> for fan5 the reading drops sharply from about 1325 rpm to zero when I lower
> the voltage. Note the fan is definitely moving still, the fan physicall can
> do much slower.
>
> fan5_div shows 4 and does not step up as with fan 1 and 4 where I can see
> _div on 128.
>
> fans 1,4 and 5 seem all to be controlled by pwm4 on this board.
>
> to my understanding fan5 should behave as the others and step up the
> divider when lowering the voltage, isn?t it?
>
> Best regards
> ????????Christian
>
> Am Montag, 30. Oktober 2006 00:12 schrieb David Hubbard:
> > Hi Christian,
> >
> > > Sorry, I should have explained. The CPU-Fan connector on the P5BDeluxe
> > > Wifi is a 4-pin, but my CPU-Fan itself has only 3 pins. As I tried to
> > > use the marked 3 pins(the 3-pin connector still fits into the 4 pin) I
> > > could not control the speed via the Bios (this is probably the purpose
> > > of the 4th pin on a 4 pin connector?) --> it stayed on a somewhat noisy
> > > speed.
> > >
> > > So I decided to use one of the 3-pin-CHA1_FAN-connector (all other Fan
> > > connectors do have 3 pins) for the CPU-fan and tried this one instead
> > > --> it works. All calm cool and quiet now.
> >
> > Okay, your sensors.conf will be different from the "standard" one for
> > this motherboard, because you map the CPU fan to a different fan
> > number.
> >
> > > This is pretty neat as the Bios apparently cannot handle this
> > > correctly. If the speed is below, say, 700 rpm or so, the Bios (when
> > > the "FAN page" is opened) becomes very slow and behaves strange
> > > sometimes.
> > >
> > > Is this a "auto-div" function of the chip or the driver?
> > >
> > > Does this mean you do not need to set it in the sensors.conf ? This
> > > should be mentioned later specifically in the comments in sensors.conf,
> > > isn?t it?
> >
> > You don't need to set the divs. In fact, I don't think you can. The
> > driver does the div scaling.
> >
> > > Interestingly enough the "sys" is very close to "CPU" in temperature
> > > wheras Bios shows 7 degrees more for the CPU (???).
> >
> > The "sys" temperature is measured inside the W83627DHG. You can test
> > this, if you like, by chilling a large metal object and then holding
> > it to the chip. But your chip may be integrated with the rest of the
> > chipset. Your mileage may vary.
> >
> > > Question:
> > >
> > > Which CPU sensor is meant here? The Core2Duo has got 2 of them
> > > internally (see coretemp driver) - is this a 3rd one?
> > >
> > > > The AUX temp may need to be configured correctly. Can you see any
> > > >
> > > > variation? It may also be that the pin on the chip is not connected
> > > > to
> > > >
> > > > anything, and just floating.
> >
> > The AUX temperature input is a pin on the W83627DHG which measures the
> > voltage from a thermal diode or thermistor. It is up to the
> > motherboard manufacturer to decide whether to connect it to a sensor,
> > or leave it floating. I don't know what it is measuring. The coretemp
> > driver measures the temperature of your CPU using a completely
> > different interface, and is not related to temperatures reported by
> > the W83627DHG.
> >
> > David
>
> ___________________________________________________________
> Der fr?he Vogel f?ngt den Wurm. Hier gelangen Sie zum neuen Yahoo! Mail:
> http://mail.yahoo.de
>
>
> _______________________________________________
> lm-sensors mailing list
> lm-sensors at lm-sensors.org
> http://lists.lm-sensors.org/mailman/listinfo/lm-sensors



	
		
___________________________________________________________ 
Der fr?he Vogel f?ngt den Wurm. Hier gelangen Sie zum neuen Yahoo! Mail: http://mail.yahoo.de

[lm-sensors] ASUS P5B DELUXE WiFi with Winbond W83627DHG

[David Hubbard]

Hi Christian,

That's interesting that lowering the _min value causes the _div to
work correctly. I'll study the driver and see if I can find out why.

What is wrong with fan 2 and 3? I didn't see that in the email.

David

[lm-sensors] ASUS P5B DELUXE WiFi with Winbond W83627DHG

[Christian Mahr]

Hi David

more investigations:

On this board, fan1, fan4 and fan5 (named CHA_FAN1, CHA_FAN3, CHA_FAN2 externally) are synchroneously controlled by pwm4.  (I have not checked CHA_FAN4 and 5) since I could not finde a corresponding fan read out. Maybe they are not monitored at all)

I succeded to establish a control loop by setting _enable=2, _target8000,_mode=1, _min_output•. Now I can observe when loading the cpu with "while true ; do  a=1; done" ,after some while the Fans do really speed up.

Interesting enough, the circuit seems to react on the CPU temperature, not on the SYS-temperature.

I am still working on fan2 and 3. Monitoring fan speed works.
I expected them to be controlled by one of the other 3 pwm?s but I am still not able to control them.

Fan3 ist connected to "PWR-FAN" which seems to be non-controllable at all. The BIOS does not offer any choice either. Only fan speed is supervised.

Fan2 seems to be connected to "CPU-FAN", the 4-pin connector. Here I am still experimenting. The main problem is that the CPU cooler I do use now has only 3 pins. If I connect any 3 pin fan to these 4 pins, the fan just turns at high speed - no control. Is there any adapter for such purposes?

On the fans measurements: I do have fans that are able to operate lower than 600 rpm. But when I turn down power below "95" and the fan reading is below 620 rpm, the measurement becomes erroneous - it jumps up to values like 10.546 rpm even at a _div\x128. This might be area of improvement in the driver? 

Best regards
	Christian


Am Donnerstag, 2. November 2006 18:53 schrieb David Hubbard:
> Hi Christian,
>
> That's interesting that lowering the _min value causes the _div to
> work correctly. I'll study the driver and see if I can find out why.
>
> What is wrong with fan 2 and 3? I didn't see that in the email.
>
> David
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.lm-sensors.org/pipermail/lm-sensors/attachments/20061102/5ab22d66/attachment.html 

[lm-sensors] ASUS P5B DELUXE WiFi with Winbond W83627DHG

[David Hubbard]

Hi Christian,

> On this board, fan1, fan4 and fan5 (named CHA_FAN1, CHA_FAN3, CHA_FAN2
> externally) are synchroneously controlled by pwm4. (I have not checked
> CHA_FAN4 and 5) since I could not finde a corresponding fan read out. Maybe
> they are not monitored at all)

Okay, that is good to know. It's a popular board, and I expect more
people will ask similar questions about their fan behavior.

> I succeded to establish a control loop by setting _enable=2,
> _target8000,_mode=1, _min_output•. Now I can observe when loading the
> cpu with "while true ; do a=1; done" ,after some while the Fans do really
> speed up.

That sounds like correct behavior to me also.

> Interesting enough, the circuit seems to react on the CPU temperature, not
> on the SYS-temperature.

In the future, we will provide sysfs entries which will document how
the chip connects a temperature to a fan output when doing automatic
fan control (like _enable=2). You are correct in noting that CPU
temperature is the current configuration.

> I am still working on fan2 and 3. Monitoring fan speed works.
>
> I expected them to be controlled by one of the other 3 pwm?s but I am still
> not able to control them.
>
> Fan3 ist connected to "PWR-FAN" which seems to be non-controllable at all.
> The BIOS does not offer any choice either. Only fan speed is supervised.

That's probably the way the motherboard is designed. Even though the
chip outputs a control voltage / PWM signal for the fan, the fan is
not designed to be controlled like this. It is still a good thing to
be able to monitor the fan so that an alarm can trigger if it fails.

> Fan2 seems to be connected to "CPU-FAN", the 4-pin connector. Here I am
> still experimenting. The main problem is that the CPU cooler I do use now
> has only 3 pins. If I connect any 3 pin fan to these 4 pins, the fan just
> turns at high speed - no control. Is there any adapter for such purposes?
The pins on a 3-pin fan are (not in order): ground, PWM signal
(typically +12V, but the signal switches on and off rapidly) and RPM
sensor. The problem is that the RPM sensor is powered by the PWM
signal, resulting in an inaccurate RPM reading. The pins on a 4-pin
fan are (not in order): ground, PWM signal drives the fan, power for
the RPM sensor (I think it's +12V, or it might be +5V) and the RPM
reading. This gives a more accurate reading. If you found the ground,
PWM, and RPM pins, you could connect a 3-pin fan. You should use a
voltmeter, and you might still damage your board. (If, for instance,
the RPM sensor on the 3-pin fan outputs a +12V sense signal and the
motherboard is designed for a +5V sense signal, you could damage it.)
That's all I know on the subject, but you could google for a while and
look for more information.

> On the fans measurements: I do have fans that are able to operate lower than
> 600 rpm. But when I turn down power below "95" and the fan reading is below
> 620 rpm, the measurement becomes erroneous - it jumps up to values like
> 10.546 rpm even at a _div\x128. This might be area of improvement in the
> driver?
No, the driver can only report readings from the chip. If the _div is
128, the readings are as low as they can get. Erroneous readings are
typical at low speeds. One way to get a more accurate reading would be
to average multiple values (two options: a moving window average, or
an alpha filter) -- this is not something that a kernel driver should
do, in keeping with the lm-sensors design philosophy. The sensors
package might be the right place to add this feature, but I am not
certain.

Hope that helps,
David

[lm-sensors] ASUS P5B DELUXE WiFi with Winbond W83627DHG

[Christian Mahr]

Hi David,

I did some more research on the 4 pin fan.

> > Fan2 seems to be connected to "CPU-FAN", the 4-pin connector. Here I am
> > still experimenting. The main problem is that the CPU cooler I do use now
> > has only 3 pins. If I connect any 3 pin fan to these 4 pins, the fan just
> > turns at high speed - no control. Is there any adapter for such purposes?
>
> The pins on a 3-pin fan are (not in order): ground, PWM signal
> (typically +12V, but the signal switches on and off rapidly) and RPM
> sensor. The problem is that the RPM sensor is powered by the PWM
> signal, resulting in an inaccurate RPM reading. The pins on a 4-pin
> fan are (not in order): ground, PWM signal drives the fan, power for
> the RPM sensor (I think it's +12V, or it might be +5V) and the RPM
> reading. This gives a more accurate reading. If you found the ground,
> PWM, and RPM pins, you could connect a 3-pin fan. You should use a
> voltmeter, and you might still damage your board. (If, for instance,
> the RPM sensor on the 3-pin fan outputs a +12V sense signal and the
> motherboard is designed for a +5V sense signal, you could damage it.)
> That's all I know on the subject, but you could google for a while and
> look for more information.


- pwm2 seems to work only in pwm mode.
- pin 1= GND, Pin2 = +12v, pin3 = sensor, pin4 = speed control (as you 
indicated)

When measuring pin4, it turns out that this delivers zero to almost 5 volts.
This means that I cannot drive a fan directly but will need a transistor or so 
to translate to 0..12 V.


>
> No, the driver can only report readings from the chip. If the _div is
> 128, the readings are as low as they can get. Erroneous readings are
> typical at low speeds. One way to get a more accurate reading would be
> to average multiple values (two options: a moving window average, or
> an alpha filter) -- this is not something that a kernel driver should
> do, in keeping with the lm-sensors design philosophy. The sensors
> package might be the right place to add this feature, but I am not
> certain.
>
can one extend the mesuremnt period in the driver?

Best regards
		Chris



	
		
___________________________________________________________ 
Der fr?he Vogel f?ngt den Wurm. Hier gelangen Sie zum neuen Yahoo! Mail: http://mail.yahoo.de

[lm-sensors] ASUS P5B DELUXE WiFi with Winbond W83627DHG

[David Hubbard]

Hi Christian,

> You ar right, but the meter should integrate, shouln?t it (the fan does
> anyway)
> What I can see is, when setting 0...255 in pwm2, this translates to 0..5 V
> roughly linear on the meter.
> 255 should mean "almost always on" and should result in something close to
> 12V, isn?t it? But I can measure only 5V....
The meter is digital. The ADC has a high sample rate and integrates
piecemeal. Or I could be wrong about your meter. However, you are
right about needing to do more research. A cheap way to measure peak
voltage is to charge a capacitor in series with a diode. It will reach
peak voltage but not discharge. Then you (might) be able to measure it
with your meter.

> No I think there might be more secrets behind 4-pin fans than I know...
> maybe these fans use TTL logic type voltages to influence electronic
> communtation internally.....
> We might need to reseearch in the internet her

David

[lm-sensors] ASUS P5B DELUXE WiFi with Winbond W83627DHG - here

[Christian Mahr]

Hi David and others on the list,


some more internet research revealed more details on the differences between 3-pin and 4-pin fans (at least to me):

First: Intels intention with this 4-pin definition (black, yellow, green, blue wires) and the related pinning from "PC-user" perspective are shown here:
 
http://www.intel.com/support/motherboards/desktop/sb/cs-012074.htm 
      and
http://www.intel.com/cd/channel/reseller/asmo-na/eng/products/box_processors/desktop/proc_dsk_p4/technical_reference/188115.htm

So Intel made sure that 4-pin fans ("the newer ones") will work on 3 pin MBs, but 3-pin fans on a 4-pin MB connectors do work only with speed supervision, but without speed control.

Some earlier motherboards seem to have a jumper to be able to select between power control on pin 2 and pwm control on pin 4, but the ASUS P5B Deluxe/Wifi seems not to offer this choice.

Some more technical explanation I found here:

http://www.analog.com/library/analogDialogue/archives/38-02/fan_speed.html
 
including the explanation why 4-pin Fans are the technical better options on the long term.

The very technical details of the 4-pin fan interface definition are given here by Intel:

http://www.formfactors.org/developer/specs/REV1_2_Public.pdf

Intel issued the spec in July 2004 and seems to promote this solution somehow since then. Nevertheless such fans are far less widespread than the 3-pin ones. Most cooler/fan manufacturer still offer mainly 3-pin fans as part of their CPU-cooler solution.
In this specification we are told the the pwm control signal is actually not modulating the 12V power as it does with 3-pin fans. 
The pwm fan control pin (pin4) contains a 0..5 V logic ("TTL-like") pwm signal (only 5mA loadable!) at 25kHz, which is intented to control the internal eletronics of the respective fan directly. 

So this type of pwm control is completely useless for a 3 pin fan which expects modulated 12-V-power, unless....
 
What might be possible is to use some external circuitry to translate the 0..5V logic signal into a pwm-modulation of the +12V supply, i.e. a power switch.


Before I start experimenting in hardware, just my question: 
 - did anybody find already a ready-made-solution for this kind of adapation circuitry?
 - if not: Is there any suitable IC known which does all in one (otherwise I might need at least 4 resistors and 2 transistors...)

Sorry for the lengthy explanation but the whole might be of interest for other MB owners as well who try to control their CPU fan and don?t succeed....

Best regards
	Christian

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.lm-sensors.org/pipermail/lm-sensors/attachments/20061103/74985e78/attachment.html 

[lm-sensors] ASUS P5B DELUXE WiFi with Winbond W83627DHG - here

[David Hubbard]

Hi Christian,

>  - if not: Is there any suitable IC known which does all in one (otherwise I
> might need at least 4 resistors and 2 transistors...)

Just off the top of my head, almost any transistor, MOSFET, or Op-Amp
would be able to switch 12V at 25 kHz and source > 100 mA. Do you have
a favorite online parts distributor?

David

[lm-sensors] ASUS P5B DELUXE WiFi with Winbond W83627DHG - here

[Pavel Ruzicka]

Hello,

I drawed few schematics of 2 and 3 pin fans.
http://www.pavouk.org/hw/fan/fan2.html

They can be useful for you. Sorry for descriptions, that are
not in English now.

> Sorry for the lengthy explanation but the whole might be of interest for
> other MB owners as well who try to control their CPU fan and don?t
> succeed....

Best regards,

Pavel Ruzicka

[lm-sensors] ASUS P5B DELUXE WiFi with Winbond W83627DHG - here

[Jean Delvare]

On Fri, 3 Nov 2006 22:39:55 +0100, Christian Mahr wrote:
> Some more technical explanation I found here:
> 
> http://www.analog.com/library/analogDialogue/archives/38-02/fan_speed.html

Nice article, I've added a link to it to our "userful addresses" page.

> So this type of pwm control is completely useless for a 3 pin fan
> which expects modulated 12-V-power, unless....
>  
> What might be possible is to use some external circuitry to
> translate the 0..5V logic signal into a pwm-modulation of the
> +12V supply, i.e. a power switch.
> 
> 
> Before I start experimenting in hardware, just my question: 
>  - did anybody find already a ready-made-solution for this
>    kind of adapation circuitry?
>  - if not: Is there any suitable IC known which does all in one
>    (otherwise I might need at least 4 resistors and 2 transistors...)

Why don't you simply use 4-wire fans if that's what the board was
designed for?

-- 
Jean Delvare

[lm-sensors] ASUS P5B DELUXE WiFi with Winbond W83627DHG - here

[Christian Mahr]

Hi Jean,

>
> Why don't you simply use 4-wire fans if that's what the board was
> designed for?

yes, that is a solution - of course. But if you look to the market, there is a very big choice of 3-pin fans (several 100-dreds) and very few 4-pin ones (maybe 10 or so).
 
If you want to use a specific CPU-fan, where you are f.e. convinced of the low-noise properties, its almost impossible to find a 4-pin version. 

Christian
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.lm-sensors.org/pipermail/lm-sensors/attachments/20061104/3dc22d1d/attachment.html 

[lm-sensors] ASUS P5B DELUXE WiFi with Winbond W83627DHG - here

[Jean Delvare]

On Sat, 4 Nov 2006 15:03:59 +0100, Christian Mahr wrote:
> > Why don't you simply use 4-wire fans if that's what the board was
> > designed for?
> 
> yes, that is a solution - of course. But if you look to the
> market, there is a very big choice of 3-pin fans (several
> 100-dreds) and very few 4-pin ones (maybe 10 or so).

I'm not surprised, 4-pin fans is a relatively new technology.

> If you want to use a specific CPU-fan, where you are f.e.
> convinced of the low-noise properties, its almost impossible to
> find a 4-pin version. 

OTOH 4-pin fans promise much better fan speed control possibilities, so
maybe a "standard" 4-pin fan will do a better job in that respect than
an "advanced" 3-pin fan. But well that's your system and your money ;)

-- 
Jean Delvare

[lm-sensors] ASUS P5B DELUXE WiFi with Winbond W83627DHG - here

[Christian Mahr]

>
> > If you want to use a specific CPU-fan, where you are f.e.
> > convinced of the low-noise properties, its almost impossible to
> > find a 4-pin version.
>
> OTOH 4-pin fans promise much better fan speed control possibilities, so
> maybe a "standard" 4-pin fan will do a better job in that respect than
> an "advanced" 3-pin fan. But well that's your system and your money ;)

Hi Jean

I?d tend to agree on the long term - but I want to see a solid market choice here. 
I know at least one negative example: the cooler that comes with a boxed Core2Duo cannot be controlled below a certain minimum speed even if you manually set the pwm2 value to zero - the fan itself simply insists on a minimum speed. 

In my case I prefer a bigger/better cooler where I have full control on the fan speed even down to zero if i like. 
Just in this moment (in almost zero load condition) I run a good  3 pin cooler at 800 rpm (very quiet) resulting in a CPU temperature of only 36 degrees. 
If I load the system I can keep fan speed below 1200 (still quiet) and CPU temperature below 45 degrees. 

Best regards
	Christian
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.lm-sensors.org/pipermail/lm-sensors/attachments/20061111/81254404/attachment.html 

[lm-sensors] ASUS P5B DELUXE WiFi with Winbond W83627DHG - here

[David Hubbard]

Hi Christian,

Very interesting result. It sounds like you did it right. I'm
curious--would an OPamp with enough driving power give you the same
result?

The w83627ehf and w83627dhg allow the fan output to be PWM or voltage
controlled. This is the fan*_mode file in sysfs. If you can hear the
PWM switching, you would know if you successfully set it in DC mode.
When I did diagnostic testing on the w83627ehf, I concluded that DC
mode runs the fan at lower RPMs for the same setting. I assume this is
because of the inductance in the fan, so PWM values in the top half of
the range all produce relatively high RPMs.

Anyway, thanks for the info.

David

On 11/16/06, Christian Mahr <christian.mahr.ulm at arcor.de> wrote:
>
>
> Hi David,
>
>
>
>
>
> in the mean time I tried the conversion form 4 to 3 pin fans but I am not
> content with the result.
>
>
>
> Circuitry verbally - brief
>  a voltage divider 4k7:1k8 connected to the pwm signal/pin4 drives a BC107
> npn transistor versus GND. draws a little less than 1 mA from the driving
> chip.
>  The collector of this drives a BD140 E versus +12V via a voltage divider
> 390 Ohm: 120 Ohm. The collector directly connected to the +12V of the fan.
>
>
> With this (maybe old-fashioned) contruction I had a lot of "pwm-noise" in
> the fan, but works somehow.
>
> The I tried to filter this a bit by inserting 3,3 Ohms and 100 micro-F into
> the fan-power +12V -line.
>
> This does make it better, but when driving low voltages, its still quite
> "pwm-noisy". Of course a RC-elemt is not a good contruction here because it
> is much more non-linar and depends on the fan impedance.
>
>
>
> Conclusion: The 3 pin fans need really good analogue voltage control.
>
>  I probably need to do something better by integrating the pwm signal maybe
> by an OPamp or so.... and driving the fan by BD139 as emitter follower.
>
>
>
> Incidentaly, the fan control on the board for the other fans seem much
> better - bu I dont have any circuirtry from the motherboard.
>
>
>
> Just to keep you informed.
>
>
>
> Christian
>
>
>
>
>
> Am Freitag, 3. November 2006 22:56 schrieben Sie:
>
>
> > Hi Christian,
>
> >
>
> > > - if not: Is there any suitable IC known which does all in one
>
> > > (otherwise I might need at least 4 resistors and 2 transistors...)
>
> >
>
> > Just off the top of my head, almost any transistor, MOSFET, or Op-Amp
>
> > would be able to switch 12V at 25 kHz and source > 100 mA. Do you have
>
> > a favorite online parts distributor?
>
> >
>
> > David
>
>

[lm-sensors] ASUS P5B DELUXE WiFi with Winbond W83627DHG

[endemic eden]

I wish for my fan on my cpu to run 2800 rpm at least I think I have seen it
that High and I am not sure how hi it can go.  I Have tried using Qfan and
turning it on to performance but it always resets and it is just a pain. Do
you know of any useful program I can use to keep the fan speed hi so I can
OC and Game on the board?




Thank you for any advice.

Michele
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.lm-sensors.org/pipermail/lm-sensors/attachments/20061126/0190cb9b/attachment.html 

[lm-sensors] ASUS P5B DELUXE WiFi with Winbond W83627DHG

[David Hubbard]

Hi Michele,

I am currently working on a driver for the W83627DHG on your
motherboard. I will be sure and let you know when we have a testing
version ready.

David

On 11/26/06, endemic eden <endemic.eden at gmail.com> wrote:
> I wish for my fan on my cpu to run 2800 rpm at least I think I have seen it
> that High and I am not sure how hi it can go.  I Have tried using Qfan and
> turning it on to performance but it always resets and it is just a pain. Do
> you know of any useful program I can use to keep the fan speed hi so I can
> OC and Game on the board?
>
> Thank you for any advice.
>
> Michele
>
> _______________________________________________
> lm-sensors mailing list
> lm-sensors at lm-sensors.org
> http://lists.lm-sensors.org/mailman/listinfo/lm-sensors
>
>