G8MNY > TECH 18.08.18 08:30l 111 Lines 4564 Bytes #8 (0) @ WW
BID : 34125_GB7CIP
Subj: PA & PSU Fan Noise
Sent: 180818/0627Z @:GB7CIP.#32.GBR.EURO #:34125 [Caterham Surrey GBR]
>From firstname.lastname@example.org Sat Aug 18 08:27:06 2018
Received: from i0ojj.ampr.org by i0ojj.ampr.org (JNOS2.0k.3b) with SMTP
id AA43192 ; Sat, 18 Aug 2018 08:27:06 +0200
X-JNOS-User-Port: Telnet (gb7cip @ 126.96.36.199) -> Sending message
To : TECH@WW
By G8MNY (Updated May 15)
(8 Bit ASCII graphics use code page 437 or 850, Terminal Font)
I have often noticed fan noise from PSUs & powerful RF PAs, this can be quite
distracting on Rx & the fan is normally not necessary in Rx mode.
Quiet fans are made & usually have feathered slots in the blade leading edges.
| | | =<
\ /Leading \ =< Deep
\ _ / Edge \ _=< V Slots
/~ ~\ /~ ~\
There is little loss in efficiency, but nearby stationary support bars now
produce a softer blurred sound pulse considerably reducing the blade pitch
Permanently reducing the fan seed will reduce the noise, & for some over cooled
systems this works well. But for applications where maximum cooling is
occasionally needed a variable speed solution is appropriate.
There are several options.
1/ Underrun the fan in Rx mode.
This needs a series R that is shorted out when on Tx. This can be done with
a relay contact on a mains fan, & to keep the R power down the mains fan
should be Power Factor tuned with a mains C, or use a lower mains voltage.
=== Fan ÚÄÄÁÄÄ¿ _³_ LÄÄÄÄÂÄÄÄÄÄÄ¿230V
CÀ---´ ³Relay³ /_\ ||)____Ú \
ÃÄÄ¿ ÀÄÄÂÄÄÙ ³ OR ||)150V ³
R / ÃÄÄÄÄÄÙ ||) Fan
NÄÄÄÄÄÁÄÄÙ PTT NÄÄÄÄÁÄÄÄÄÄÄÙ
2/ Temperature sense.
By adding a thermal switch a DC or mains fan can be kept off, or running
slow most of the time.
LÄÄÄÄ¿ MODE SPEED UP LÄÄÄÄ¿
Fan MODE Fan
/Thermal R /Thermal
NÄÄÄÄÙ Switch NÄÄÄÄÁÄÄÙ Switch
3/ Variable controlled DC a fan.
This offers the quietest fan control.
³ e\³_Old PNP
\³ /³ Germanium
ÃÄÄÄÙ In heatsink
If an old germanium PNP is not available, it can be a suitable NTC
thermistor & the NPN can have a hold off R base to emitter to set the
operation temperature. The cap is needed on DC fans to handle the pulse
currents & make the motor start better. A permanent quiet run R can be put
across the NPN if wanted.
Feedback from Osvaldo LW1DSE..
In place of a resistor to drop the voltage for the AC motor fan, I used a 1æF
275VAC, keeping in mind that this cap not resonates with the motor inductance.
I used 1æF, but any other cap will do the job, and increasing its value,
reduces the capacitive reactance (Xc = 1/ 2ã * F * C) and then increases motor
speed. If this cap have infinite value, the motor will run at normal direct
speed. And it doesn't discipates heat!
For DC motors, the solution is simpler. There exists a device called LM35 from
National Semiconductor in TO92 case, the LM35 that is a thermal sensor
directly calibrated inside the chip to give an output of 10mV/øC, then, at 30øC
it will give about 300mV. Then, letting this voltage to a comparator like LM311
& putting the wiper of a potentiometer in the other input as voltage
Temperature reference, & giving some hysteresis using a positive feedback, you
can do a precise thermal control, in the form of an on/off circuit. As the
power consumption of a normal fan is in order of tens of mA, the internal LM311
transistor is sufficient to drive this motor directly, the most of them being
brushless, & then the freeweeling diode is unnecesary. Or, you can feed this
voltage to the input of a PWM controller like one of the voltage mode PWM IC's,
& do a PWM analog speed control with low power drain.
See my TECH Buls "Variable Speed Fan for AF PA" & "Variable Speed Thermal Fan"
Why Don't U send an interesting bul?
73 De John, G8MNY @ GB7CIP
Read previous mail | Read next mail