[sci.electronics] Audible effects in solid-state electronics

woolsey@nsc.nsc.com (Jeff Woolsey) (01/24/89)

I'm sitting here listening to my CD player being listened to by someone 
else, and find that the noise it makes (sounds like the disc sliding
on something) is not present between tracks, yet the disk still spins.
It occurs to me that this noise is also present in the FPA on Vaxen.
It's a sort of hissing sound.

So, is there any validity to the theory that certain logic families
exhibit piezoelectric effects when operating?

Am I rediscovering some fundamental physics here?
-- 
-- 
When it comes to humility, I'm the greatest.  -- Bullwinkle J. Moose

Jeff Woolsey  woolsey@nsc.NSC.COM  -or-  woolsey@umn-cs.cs.umn.EDU

david@daisy.UUCP (David Schachter) (01/25/89)

A hissing noise on a graphics product I worked on occurred only during
graphics operations; when the unit was idle, it was silent.

This was traced down to the switching power supply: when the unit was busy,
it demanded more current and the pattern of current demand fell into the
audio range.  The noise was the switching power supply responding to the
sudden demands.

Perhaps the noise you are hearing is similar?  Or perhaps DAC switching
transients improperly filtered?

				-- David "Worthless Opinion" Schachter

henry@utzoo.uucp (Henry Spencer) (01/25/89)

In article <9359@nsc.nsc.com> woolsey@nsc.nsc.com (Jeff Woolsey) writes:
>...So, is there any validity to the theory that certain logic families
>exhibit piezoelectric effects when operating?

Much more probably you are rediscovering the fact that switching power
supplies can "sing" and the music :-) is load-dependent.  It's quite
noticeable on an Iris 4D when the graphics hardware fires up.
-- 
Allegedly heard aboard Mir: "A |     Henry Spencer at U of Toronto Zoology
toast to comrade Van Allen!!"  | uunet!attcan!utzoo!henry henry@zoo.toronto.edu

knudsen@ihlpl.ATT.COM (Knudsen) (01/26/89)

A year ago I was using a Sun 3/xx workstation, and it seemed like
I could hear the pixels hitting the raster!  Turned out I heard
the same sizzling squeaky sounds during non-graphic but CPU-intensive
tasks like C compiles and nroffs.  So the sound must have been coming
from the CPU, RAM, or associated glue logic (all TTL I'd think,
for the glue).  Some folks thought I was nuts, others could hear it.

I now have a Sun 3/50 with color monitor, and can't hear it.

BTW, I don't think we're talking about switching power supply noises;
I doubt any CD player would use these.
-- 
Mike Knudsen  Bell Labs(AT&T)   att!ihlpl!knudsen
"Five hundred twelve K bytes of RAM ... out of control ... "

ch@maths.tcd.ie (Charles Bryant) (01/29/89)

In article <1989Jan25.054823.2166@utzoo.uucp> henry@utzoo.uucp (Henry Spencer) writes:
>In article <9359@nsc.nsc.com> woolsey@nsc.nsc.com (Jeff Woolsey) writes:
>>...So, is there any validity to the theory that certain logic families
>>exhibit piezoelectric effects when operating?
>
>Much more probably you are rediscovering the fact that switching power
>supplies can "sing" and the music :-) is load-dependent.  It's quite
>noticeable on an Iris 4D when the graphics hardware fires up.
>-- 

I have heard the same effect from my Sinclair Spectrum which has a linear power
supply. The sound changes depending on what the CPU was doing. In fact it was a
good way to tell if your program had crashed as the tone changed! It may have
come partly from the video circuits as it also changed depending on what was
on the screen, but there were different sounds produced for blank screens.

inews
gives
no
prizes
for
conciseness
-- 

		Charles Bryant.
Working at Datacode Electronics Ltd.

brian@ucsd.EDU (Brian Kantor) (01/29/89)

I've heard memory boards sing; it wasn't the power supply because I was
running the system off of batteries at the time, using just simple monolythic
pass regulators.  I believe that it was the high-current output driver
structures that were the cause, since in write-only tests, and during
refresh (neither of which turn on the output drivers in the chips) did I
hear the crickets chirping.  I don't know what the physical effect is
(thermal, magnetic, poltergeisten), but it's real.

Of course, in larger machines, it's most likely the switching supply
loading down into the audible range, but there could be chirping chips too.

	Brian Kantor	UC San Diego

dave@onfcanim.UUCP (Dave Martindale) (02/03/89)

In article <9359@nsc.nsc.com> woolsey@nsc.nsc.com (Jeff Woolsey) writes:
>I'm sitting here listening to my CD player being listened to by someone 
>else, and find that the noise it makes (sounds like the disc sliding
>on something) is not present between tracks, yet the disk still spins.

What you're hearing is white noise, more or less.

The pattern of pits on the disk modulates the light that reaches an
array of 4 or 6 photodiodes in the laser pickup.  The output of some
of these is summed to produce a data signal, from which the digital
data is extracted.  The difference between pairs of these photodiodes
is used to provide focus and tracking error signals.  The data bits
is still present in the "difference" signals to some extent, and
for most intents and purposes is similar to random bits, which is
white noise.

This noise is then fed through the servo amplifiers
and to the mechanical actuators that actually move the little lens.
The frequency response of these actuators extends into the audio range,
and so they reproduce the digital information as a hissing noise that
you can hear.

Between tracks, the digital data is mostly zeros in a very regular pattern,
so the "white noise" sound goes away.  As soon as the music starts, it
comes back.

If you don't believe me, follow the tracking or focus servo circuits
with an oscilloscope.

>It occurs to me that this noise is also present in the FPA on Vaxen.
>It's a sort of hissing sound.

You're probably hearing a switching voltage regulator's noise change
as the load on it changes.  (But on our VAX, all I can hear is the fans).