Reducing EMF on cooling fans

F

Fritz Oppliger

Guest
I am tasked with reducing "EMF" in the vicinity of computer type
brushless 12VDC cooling fans. Problem similar to
http://www.sciencedaily.com/releases/2008/04/080430201654.htm
As I am controlling speed with PWM I suspected radiation from the feed
wires. But even going 100% (no switching) the noise is there. Reducing
duty cycle to about 20% reduces the noise but is not useable - it is near
stalling speed and I DO need the fans to fan.
Twisting the supply lines does not help much. What other quick fixes could
I try? I suspect the motors themselves are noisy and all I can do is try
different models / brands?
** Posted from http://www.teranews.com **
 
On Sat, 03 May 2008 23:20:34 -0700, Fritz Oppliger wrote:

I am tasked with reducing "EMF" in the vicinity of computer type
brushless 12VDC cooling fans. Problem similar to
http://www.sciencedaily.com/releases/2008/04/080430201654.htm As I am
controlling speed with PWM I suspected radiation from the feed wires.
But even going 100% (no switching) the noise is there. Reducing duty
cycle to about 20% reduces the noise but is not useable - it is near
stalling speed and I DO need the fans to fan. Twisting the supply lines
does not help much. What other quick fixes could I try? I suspect the
motors themselves are noisy and all I can do is try different models /
brands?
** Posted from http://www.teranews.com **
Click to expand...
Capacitors across the motors? Ferrite beads near the motors?
 
Fritz Oppliger wrote:
I am tasked with reducing "EMF" in the vicinity of computer type
brushless 12VDC cooling fans. Problem similar to
http://www.sciencedaily.com/releases/2008/04/080430201654.htm
As I am controlling speed with PWM I suspected radiation from the feed
wires. But even going 100% (no switching) the noise is there. Reducing
duty cycle to about 20% reduces the noise but is not useable - it is near
stalling speed and I DO need the fans to fan.
Twisting the supply lines does not help much. What other quick fixes could
I try? I suspect the motors themselves are noisy and all I can do is try
different models / brands?
** Posted from http://www.teranews.com **
Click to expand...

Those are 'Brushless DC motor' type fans with an internal three phase
generator. That is the source of the noise, and has to be filtered with
a grounded screen shield that will pass the air, but not the RF.


--
Service to my country? Been there, Done that, and I've got my DD214 to
prove it.
Member of DAV #85.

Michael A. Terrell
Central Florida
 
"Fritz Oppliger" <fritz@fritzop.dot.com> wrote in message
news:eek:p.ual88kqny8zufa@tosh...
I am tasked with reducing "EMF" in the vicinity of computer type
brushless 12VDC cooling fans. Problem similar to
http://www.sciencedaily.com/releases/2008/04/080430201654.htm
As I am controlling speed with PWM I suspected radiation from the feed
wires. But even going 100% (no switching) the noise is there. Reducing
duty cycle to about 20% reduces the noise but is not useable - it is near
stalling speed and I DO need the fans to fan.
Twisting the supply lines does not help much. What other quick fixes could
I try? I suspect the motors themselves are noisy and all I can do is try
different models / brands?
** Posted from http://www.teranews.com **
Click to expand...

Physical proximity... use a hose to the incubator, rather than having the
motor be directly attached. Surround it with a sheet metal cage. It'll be
quieter too, which is, in my view, a more likely culprit than emf.

Regards,
Bob Monsen
 
On Sat, 03 May 2008 23:20:34 -0700, "Fritz Oppliger"
<fritz@fritzop.dot.com> wrote:

I am tasked with reducing "EMF" in the vicinity of computer type
brushless 12VDC cooling fans. Problem similar to
http://www.sciencedaily.com/releases/2008/04/080430201654.htm
Click to expand...
The fans also blow fresh air over the babies, which the tape recorders
don't. I'd suspect that the average baby notices hot/cold more than it
notices weak magnetic fields.


As I am controlling speed with PWM I suspected radiation from the feed
wires. But even going 100% (no switching) the noise is there.
Click to expand...
What noise? Spikey PWM from your driver, or the low-frequency mag
field from the fan?

John
 
On Sun, 04 May 2008 09:21:35 -0700, John Larkin
<jjlarkin@highNOTlandTHIStechnologyPART.com> wrote:


As I am controlling speed with PWM I suspected radiation from the feed
wires. But even going 100% (no switching) the noise is there.

What noise? Spikey PWM from your driver, or the low-frequency mag
field from the fan?

My measurement device does not say. But the PWM is flat DC at 100% and
Click to expand...
since it is bit-banged by a 8051 is not too terribly fast either ;-) So
I'm sure it is the mag field from the fans. easy to modify ? not likely .

** Posted from http://www.teranews.com **
 
On Sun, 04 May 2008 10:49:49 -0700, "Fritz Oppliger"
<fritz@fritzop.dot.com> wrote:

On Sun, 04 May 2008 09:21:35 -0700, John Larkin
jjlarkin@highNOTlandTHIStechnologyPART.com> wrote:


As I am controlling speed with PWM I suspected radiation from the feed
wires. But even going 100% (no switching) the noise is there.

What noise? Spikey PWM from your driver, or the low-frequency mag
field from the fan?

My measurement device does not say.
Click to expand...
---
What are you using for a "measurement device"?
---

But the PWM is flat DC at 100% and
since it is bit-banged by a 8051 is not too terribly fast either ;-) So
I'm sure it is the mag field from the fans. easy to modify ? not likely .
Click to expand...
---
You posted earlier that twisting the leads to the fan helped somewhat,
so you may want to try shielding the leads as well.

JF
 
On Sun, 04 May 2008 10:49:49 -0700, "Fritz Oppliger"
<fritz@fritzop.dot.com> wrote:

On Sun, 04 May 2008 09:21:35 -0700, John Larkin
jjlarkin@highNOTlandTHIStechnologyPART.com> wrote:


As I am controlling speed with PWM I suspected radiation from the feed
wires. But even going 100% (no switching) the noise is there.

What noise? Spikey PWM from your driver, or the low-frequency mag
field from the fan?

My measurement device does not say. But the PWM is flat DC at 100% and
since it is bit-banged by a 8051 is not too terribly fast either ;-) So
I'm sure it is the mag field from the fans. easy to modify ? not likely .

** Posted from http://www.teranews.com **
Click to expand...
Brushless DC fans draw spikey current and locally make AC magnetic
fields. Both are low frequency, numbers roughly like 80 Hz. The
current ripple can get into power supplies and bother low-level analog
circuits, and may cause ground loop noise. The mag fields can induce
low-level (microvolts, typically) voltages into loops on nearby
circuit boards.

What is your "measurement device"?

John
 
On Sun, 04 May 2008 14:23:56 -0700, John Larkin
<jjlarkin@highNOTlandTHIStechnologyPART.com> wrote:

Brushless DC fans draw spikey current and locally make AC magnetic
fields. Both are low frequency, numbers roughly like 80 Hz. The
current ripple can get into power supplies and bother low-level analog
circuits, and may cause ground loop noise. The mag fields can induce
low-level (microvolts, typically) voltages into loops on nearby
circuit boards.

What is your "measurement device"?

OK don't laugh -
Click to expand...
(C) 1995 Connectware PCboard with 10 LEDs on it. It has a 43mm square
antenna loop area (in to 18 mm at the center)
it sports an LM 3914V and a TLE2114C IC. Powered by 9V battery, purchased
from BG Micro about ten years ago. It lights up all the LEDs when I get to
within 3 inches of the fans, tapering off to zero within 10 inches from
the fans.
It WAS sold as EMF detector. It seems to work as such. It is what I have.
It is quite sensitive, a fraction of an inch makes a big difference in the
display.

** Posted from http://www.teranews.com **
 
On Sun, 04 May 2008 16:18:44 -0700, "Fritz Oppliger"
<fritz@fritzop.dot.com> wrote:

On Sun, 04 May 2008 14:23:56 -0700, John Larkin
jjlarkin@highNOTlandTHIStechnologyPART.com> wrote:


Brushless DC fans draw spikey current and locally make AC magnetic
fields. Both are low frequency, numbers roughly like 80 Hz. The
current ripple can get into power supplies and bother low-level analog
circuits, and may cause ground loop noise. The mag fields can induce
low-level (microvolts, typically) voltages into loops on nearby
circuit boards.

What is your "measurement device"?

OK don't laugh -
(C) 1995 Connectware PCboard with 10 LEDs on it. It has a 43mm square
antenna loop area (in to 18 mm at the center)
it sports an LM 3914V and a TLE2114C IC. Powered by 9V battery, purchased
from BG Micro about ten years ago. It lights up all the LEDs when I get to
within 3 inches of the fans, tapering off to zero within 10 inches from
the fans.
It WAS sold as EMF detector. It seems to work as such. It is what I have.
It is quite sensitive, a fraction of an inch makes a big difference in the
display.

** Posted from http://www.teranews.com **
Click to expand...


Why are you "tasked" to satisfy what is probably a crap, uncalibrated
instrument?

John
 
On Tue, 06 May 2008 17:31:05 -0700, Paul E. Schoen <pstech@smart.net>
wrote:

"John Larkin" <jjlarkin@highNOTlandTHIStechnologyPART.com> wrote


Why are you "tasked" to satisfy what is probably a crap, uncalibrated
instrument?

The example of heart rate variability in newborns in incubators may be
caused by 50-60 Hz EM fields, which is a dangerous frequency that can
cause
fibrillation (but usually because of conducted current through heart
muscle). If your instrument shows zero levels at 10" from the fans, then
why not maintain that distance? Also, see if there is a problem if you
run
the fan on a smooth DC signal. You can filter the PWM if that is the
cause
of the noise. And if your project is as sensitive and important as
protecting infants in incubators, you really need to invest in some
accurate and reliable equipment, and you probably should get data on the
field strength that was measured in the experiments you cited in the
article.

A simple magnetic field sensor can be made from an air core choke, and
can
be connected to a scope to see the frequency and intensity. The choke's
axis can be oriented to determine the direction of the field. This will
probably be more useful than your unknown instrument. You can probably
produce a known field strength by passing a measured current through a
toroidal transformer and placing the sensor in the hole.

Paul


Actually our incubator is for exotic bird eggs. The article prompted us
Click to expand...
to investigate. While there may be no evidence suggesting that EMF harms
eggs we thought it wise to check into the EMF angle as far as our hardware
is concerned, and perhaps claim superiority on the basis of "LOW EMF
ENVIRONMENT IS GOOD FOR YOUR $10'000 EGGS" .
What I have found so far: "our" EMF is not from line AC but from the fans,
internally (not PWM). I will find out the frequency as per your suggestion
above - thank you.
Fritz
** Posted from http://www.teranews.com **
 
Fritz Oppliger wrote:

I am tasked with reducing "EMF" in the vicinity of computer type
brushless 12VDC cooling fans.
Click to expand...
I think you mean 'EMI' - electromamgnetic interference.


Problem similar to
http://www.sciencedaily.com/releases/2008/04/080430201654.htm
As I am controlling speed with PWM I suspected radiation from the feed
wires. But even going 100% (no switching) the noise is there. Reducing
duty cycle to about 20% reduces the noise but is not useable - it is near
stalling speed and I DO need the fans to fan.
Twisting the supply lines does not help much. What other quick fixes could
I try? I suspect the motors themselves are noisy and all I can do is try
different models / brands?
Click to expand...
I have found huge variations in the EMI of 80mm axial fans from one
manufacturer to another. I'd suggest you try several different brands.

As you've discovered, not much of the noise is from the supply current but do
put a largish cap (100uF perhaps) across the supply rails where the fan plugs
in (plus a little series DCR or some L too). I expect the EMI's mostly from
the internal magnetics and you can't see those but you can choose by
manufacturer.

Graham
 
On Wed, 07 May 2008 06:33:36 +0100, Eeyore
<rabbitsfriendsandrelations@hotmail.com> wrote:

I expect the EMI's mostly from
the internal magnetics and you can't see those but you can choose by
manufacturer.
Click to expand...
---
Ypu can certainly see the effect of the internal magnetics; just use
one of these:

news:aua3241i15s4ivvvtmd56isav38308u81o@4ax.com
JF
 
John Fields wrote:

Eeyore wrote:

I expect the EMI's mostly from
the internal magnetics and you can't see those but you can choose by
manufacturer.

---
Ypu can certainly see the effect of the internal magnetics; just use
one of these:

news:aua3241i15s4ivvvtmd56isav38308u81o@4ax.com
JF
Click to expand...
I use something very similar ! :~)

Graham
 
You must log in or register to reply here.

Welcome to EDABoard.com

Sponsor

Back
Top