[time-nuts] 20221128: A question re fans and low-noise measurements / techniques

[Andrew Kalman]

All:

I'm wondering if anyone has some wisdom backed up with experience on the
best practices for implementing cooling fans inside of low-noise
instruments. Other than "Don't do it!" :-)

Maybe someone encountered an instrument that had a particularly elegant /
efficient / quiet fan implementation.

I recently repaired the fan circuit on an SRS SR620 (it uses a PTC to
control fan speed through two bipolar transistors via a typical dc
brushless motor fan, PCB had some burnt traces), and in doing so it
occurred to my (naive) mind that there might be advantages to using such
"simple", "analog" fan speed control over temperature compared to using
more modern PWM controllers, from a conducted (and radiated?) noise
perspective.

Also, HP has/had many instruments that just run AC fans at full blast,
irrespective of any temperatures that might be sensed. That's great, except
it's killing my hearing.

So, I guess I'm wondering whether there are any general statements about
best practices (other than "Do lots of filtering") if one needs to
implement a fan in a low-electrical-noise instrument. The main fan schemes
I can think of are:

   - AC fan (full speed, all the time).
   - PTC or similar temperature control over brushless fan voltage.
   - PWM-based fan driver with sensed temperature affecting PWM duty cycle.

Thanks,

--Andrew

--------------------------------
Andrew E. Kalman, Ph.D.