All of lore.kernel.org
 help / color / mirror / Atom feed
From: Matt Roberds <mattroberds@cox.net>
To: lm-sensors@vger.kernel.org
Subject: Re: [lm-sensors] [Bug?] W83697: Broken readings for fan speed 10%
Date: Mon, 09 Jun 2008 00:35:36 +0000	[thread overview]
Message-ID: <Pine.LNX.4.64.0806081915360.4011@birdbird.example.com> (raw)
In-Reply-To: <20080531220339.05dc83fd@neptune.home>

[-- Attachment #1: Type: TEXT/PLAIN, Size: 2520 bytes --]

On Sun, 8 Jun 2008, Bruno Prémont wrote:
> I will look if I can find a way do determine the real fan speed in
> order to compare with the values both chips report.

The exactly right instrument to use for this is probably an optical
tachometer.  It looks for reflections from a piece of reflective tape or
a white stripe painted on a rotating machine, and tells you the speed in
rpm.  If you're in university (or know someone who is), the physics or
mechanical engineering labs will often have such a device.
http://www.extech.com/instrument/products/451_499/461700.html is one
example but they are made by several companies.

There is another kind of tachometer that has a small wheel or probe that
you hold against the rotating machine; the machine spins the wheel and
the tachometer displays the rpm.  A computer fan motor is not very
strong, though, and making the measurement may tend to slow down the
fan.  http://www.extech.com/instrument/products/451_499/461750.html
is an example, but again, several companies make them.

The previous way to do this was with a stroboscope - a flashing light
with adjustable flash rate.  You point the light at the rotating machine
and turn a knob until the flashes of light "stop" the motion of the
machine.  You then read the flash rate from the knob and compute the
rpm.  http://www.ietlabs.com/Genrad/1531.html is a classic example.

If you have a function generator, you might be able to connect a LED to
its output and use it as a stroboscope.  (Make sure the output of the
function generator is strong enough to drive a LED; some of them are not
designed for this.)

Another way to do it is to use a plain old fluorescent lamp (with
magnetic ballast) or neon glow lamp.  Either one will flash at 100 Hz
(Europe) or 120 Hz (North America).  It may help to be in a dark room
with the fluorescent or neon lamp as the only source of illumination.
You can't adjust the flash rate, but you can use it to get an idea of
the fan speed.  For instance, if you mark one fan blade, and it appears
to "stop" in the same spot every time, the fan is probably turning at
6000 rpm (Europe): 100 Hz * 60 sec.  (It could also be turning at 12000
or 18000 or 24000 rpm, but most computer fans don't go that fast.)  If
the mark appears to "stop" in two places, 180 degrees apart, the fan is
probably turning at 3000 rpm, and so on.

Standard disclaimers apply; I don't get money or other consideration
from any companies mentioned.

Matt Roberds

[-- Attachment #2: Type: text/plain, Size: 153 bytes --]

_______________________________________________
lm-sensors mailing list
lm-sensors@lm-sensors.org
http://lists.lm-sensors.org/mailman/listinfo/lm-sensors

  parent reply	other threads:[~2008-06-09  0:35 UTC|newest]

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2008-05-31 20:03 [lm-sensors] [Bug?] W83697: Broken readings for fan speed 10% of Bruno Prémont
2008-05-31 20:17 ` [lm-sensors] [Bug?] W83697: Broken readings for fan speed 10% Grant Coady
2008-05-31 20:30 ` Bruno Prémont
2008-06-02 10:32 ` Jean Delvare
2008-06-02 19:09 ` Bruno Prémont
2008-06-02 19:35 ` Jean Delvare
2008-06-02 20:04 ` Bruno Prémont
2008-06-08 20:30 ` Bruno Prémont
2008-06-09  0:35 ` Matt Roberds [this message]
2008-06-11 15:59 ` Jean Delvare
2008-06-11 21:09 ` Bruno Prémont
2008-06-12  8:22 ` Jean Delvare
2008-06-12 18:37 ` Bruno Prémont
2008-06-13 12:04 ` Jean Delvare
2008-06-13 21:31 ` Bruno Prémont
2008-06-14 20:57 ` Jean Delvare
2008-06-14 21:24 ` Bruno Prémont
2008-06-15  7:35 ` Jean Delvare
2008-06-15 10:02 ` Bruno Prémont

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=Pine.LNX.4.64.0806081915360.4011@birdbird.example.com \
    --to=mattroberds@cox.net \
    --cc=lm-sensors@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.