Noctua fan speed not showing in Speedfan

Control: management of fans, temp/rpm monitoring via soft/hardware

Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Lawrence Lee

Post Reply
allenlux
Posts: 39
Joined: Sun Dec 25, 2005 4:56 am
Location: Bofferdange, Luxembourg
Contact:

Noctua fan speed not showing in Speedfan

Post by allenlux » Sat Jun 30, 2007 4:49 am

I've got a Noctua NF-S12 1200 RPM fan hooked up to the mainboard case fan header on an Asrock 4CoreDual-VSTA motherboard.

For some reason the fan speed isn't showing up in Speedfan 4.32.

The CPU fan (Intel stock HSF) on the same mb sensor gives a good speed readout.

I've played around with the "FAN1 divisor" and "PWMOUT1" settings, but still no speed readout.

Any suggestions?

seemingly.random
Posts: 176
Joined: Mon Oct 16, 2006 11:51 am
Location: Southeast, USA

Post by seemingly.random » Sat Jun 30, 2007 9:13 am

CPUfan on an ASUS M2NSLI is reported as 1350000 and a divider max of 20 doesn't really help. This MB has an IT8712F super IO controller - one that SpeedFan doesn't do well with. It does just fine on an abit nfm2 though - it has a Winbond W83627EHF. The temps seem to be reported correctly on all MBs.

ASUS' PCProbe can be used to report missing statistics. Maybe there is an equivalent for the ASRock. It would be nice to have all in one window though.

Since SpeedFan is free, it's hard to complain although it would be nice if there was GPL'd source available.

allenlux
Posts: 39
Joined: Sun Dec 25, 2005 4:56 am
Location: Bofferdange, Luxembourg
Contact:

Post by allenlux » Sat Jun 30, 2007 9:29 am

The Asrock 4CoreDual-VSTA mb has a Winbond W83697HF sensor and Speedfan has no problem using this to measure the CPU fan speed, which is why I wondered if it might be something unusual about the Noctua fan.

seemingly.random
Posts: 176
Joined: Mon Oct 16, 2006 11:51 am
Location: Southeast, USA

Post by seemingly.random » Sat Jun 30, 2007 11:16 am

Don't have experience with the Noctua but there have been some glowing reviews of them and I'm sure it would have come out if there is a problem with the rpm sensor. All I can suggest is the obvious - inspect the wiring for damage, try it on a different fan header or another MB and/or try another fan on the ASRock. I had an ASUS 690G that had a dead fan header.

Das_Saunamies
*Lifetime Patron*
Posts: 2000
Joined: Tue May 15, 2007 1:39 am
Location: Finland

Post by Das_Saunamies » Sun Jul 01, 2007 4:52 am

Can you get the speed to show up at any RPM at all? Are you using the ULNA(the Noctua's adapter)?

My old 7000B won't report its RPM in my setup. It's connected to A8N-E's CPU1 through a Fan Mate 2 that limits the RPM to 2200. QFAN is enabled, and the RPM apparently just drop off the scale for the mobo, and it can't pick them up even at full speed. It either shows 0 or a rock-solid 1380, which isn't true either. Cool'n'Quiet is of course enabled, and before the Fan Mate 2 was added it would report the RPM being between 680 and 14000. :D

Since the stock HSF gives a reading, I bet it's the fan dropping out of scale or having a loose wire in the lead.

allenlux
Posts: 39
Joined: Sun Dec 25, 2005 4:56 am
Location: Bofferdange, Luxembourg
Contact:

Post by allenlux » Sun Jul 01, 2007 5:50 am

Yes, I'm using the ULNA lead which should reduce the RPM to 600.

In Speedfan the reading is always zero RPM, with all divisor settings.

In the interests of science I just tried removing the ULNA lead and now Speedfan does give a reading - 1260 RPM, just as it should!

Obviously the ULNA lead interferes - either by reducing the RPM below a level recognised by Speedfan, or for some electrical reason, although as far as I know it's just dropping the voltage in the +ve line, which should affect the RPM signal.

Unfortunately it doesn't seem possible to control this fan's speed with Speedfan.

allenlux
Posts: 39
Joined: Sun Dec 25, 2005 4:56 am
Location: Bofferdange, Luxembourg
Contact:

Post by allenlux » Sun Jul 01, 2007 7:18 am

Another discovery: I connected the Noctua fan through a Zalman low-voltage lead like this:
http://www.thecoolingemporium.co.uk/pro ... ucts_id=47

Now the fan is turning at 827 RPM and Speedfan is giving an RPM readout.

I guess the Zalman lead - which is just an inline resistor - does not interfere with the RPM sensor, whereas the Noctua ULNA lead does.

I still can't control the fan speed via Speedfan, though.

Felger Carbon
Posts: 2049
Joined: Thu Dec 15, 2005 11:06 am
Location: Klamath Falls, OR

Post by Felger Carbon » Sun Jul 01, 2007 8:18 am

allenlux wrote:I guess the Zalman lead - which is just an inline resistor - does not interfere with the RPM sensor, whereas the Noctua ULNA lead does.
[This contained an explanation of what I thought the problem was, and how to fix it. I tried that fix and it didn't work. My second guess is that the fan controller IC, designed to work at 12V, just doesn't handle the tach logic correctly at lower voltages. Keep in mind my first guess was wrong. Some experiments work; some don't. This one didn't. :(

edit: changed 2.2K to 4.7K - sorry, was thinking 5V logic. Old habits die hard!
edit2: typo, changed "from" to "for"
edit3: announced a "fix" didn't work
Last edited by Felger Carbon on Wed Jul 18, 2007 8:42 am, edited 2 times in total.

allenlux
Posts: 39
Joined: Sun Dec 25, 2005 4:56 am
Location: Bofferdange, Luxembourg
Contact:

Post by allenlux » Sun Jul 01, 2007 8:30 am

Felger Carbon, that's really interesting.

I'm going away for a couple of days (a "management seminar") so can't contribute any more for a day or two - but I'm curious as to the fix you suggest. Soldering holds no terrors - in my early days I fixed my Osborne portable computer by replacing RAM chips which were soldered on the motherboard....

Das_Saunamies
*Lifetime Patron*
Posts: 2000
Joined: Tue May 15, 2007 1:39 am
Location: Finland

Post by Das_Saunamies » Sun Jul 01, 2007 12:24 pm

Thank you, Mr. Carbon, another expert post to refer to when this problem pops up again in the next two weeks. :)

allenlux
Posts: 39
Joined: Sun Dec 25, 2005 4:56 am
Location: Bofferdange, Luxembourg
Contact:

Post by allenlux » Wed Jul 04, 2007 12:22 pm

allenlux wrote:I still can't control the fan speed via Speedfan, though.
Deep embarrassment on my side: I had forgotten that PWM speed control normally requires a 4-wire fan connection, whereas the case fan header on my mb and the Noctua fan both use 3 wires. The CPU fan header on the mb is a 4-wire header so of course it does provide speed control.

Das_Saunamies
*Lifetime Patron*
Posts: 2000
Joined: Tue May 15, 2007 1:39 am
Location: Finland

Post by Das_Saunamies » Wed Jul 04, 2007 1:03 pm

Oh well, at least you realized it on your own! 8)

jedi master
Posts: 18
Joined: Sat Oct 07, 2006 2:13 pm

Post by jedi master » Wed Jul 04, 2007 1:54 pm

I have a Noctua 1200RPM fan that does'nt show up the rpm in mbm5, speedfan, evertest home edition and the bios. Tried on a NF7 and a A7N8X-Deluxe, i guessing this is just something to do with Noctua.

Das_Saunamies
*Lifetime Patron*
Posts: 2000
Joined: Tue May 15, 2007 1:39 am
Location: Finland

Post by Das_Saunamies » Thu Jul 05, 2007 2:43 am

jedi master wrote:--i guessing this is just something to do with Noctua.
Have you read the above, especially Felger Carbon's post? How does the above apply to your situation? Have you tried at full speed, direct contact?

My 1200 and 800 worked just fine, clear RPM readings all the way without the ULNA.

jaganath
Posts: 5085
Joined: Tue Sep 20, 2005 6:55 am
Location: UK

Post by jaganath » Thu Jul 05, 2007 3:22 am

allenlux wrote:
allenlux wrote:I still can't control the fan speed via Speedfan, though.
Deep embarrassment on my side: I had forgotten that PWM speed control normally requires a 4-wire fan connection, whereas the case fan header on my mb and the Noctua fan both use 3 wires. The CPU fan header on the mb is a 4-wire header so of course it does provide speed control.
Speedfan should be able to control 3-pin fans.

Das_Saunamies
*Lifetime Patron*
Posts: 2000
Joined: Tue May 15, 2007 1:39 am
Location: Finland

Post by Das_Saunamies » Thu Jul 05, 2007 3:24 am

Don't know how it is with the PWM connectors. If you only connect a regular 3-pin fan lead to it, does it allow the same basic controls as a regular 3-pin connector, or does it all somehow fall back on the 4th pin?

Felger Carbon
Posts: 2049
Joined: Thu Dec 15, 2005 11:06 am
Location: Klamath Falls, OR

Post by Felger Carbon » Thu Jul 05, 2007 9:44 am

jaganath wrote:Speedfan should be able to control 3-pin fans.
Only if the mobo replaces the +12V center connection with a PWM signal. Thus, your statement is mobo-dependent.

Post Reply