Re: Clicking IIgs'

[dempson@actrix.gen.nz (David Empson)]

oz390gta <oz390gta@gmail.com> wrote:

> On Aug 11, 10:03 am, demp...@actrix.gen.nz (David Empson) wrote:
> >
> > oz390gta <oz390...@gmail.com> wrote:
> > > I have a ROM01 and a ROM03 gs which both make a continous clicking
> > > sound through the speaker when running GS/OS. This clicking does not
> > > happen when in Prodos or when the system is interupted like holding
> > > down the mouse button or a key. I once read it had something to do
> > > with the 5.25 drive, they even do it with no drives connected.
> >
> > It isn't the 5.25" drive: it is the 3.5" drive that is being polled.
> >
> > As long as Finder is running or you have a file selection dialog open,
> > and sitting idle in the main event loop, the system is polling every
> > connected 3.5" drive once per second to check for a changed disk.
> > Holding down the mouse button gets out of the main event loop until you
> > release the button, which temporarily stops the polling.
> >
> > The clicking sound is due to induced noise travelling into the sound
> > circuitry via the power rails, as polling the 3.5" drive causes a brief
> > power surge on the +12V rail.
> >
> > I was under the impression that the improvements in the ROM 3 design
> > reduced or eliminated much of the noise being picked up the sound
> > circuitry, but clearly your ROM 3 still has the problem (mine didn't).
> >
> > > I have another ROM01 that does not click so I put that power supply into
> > > the other two, they still click.
> >
> > It isn't the power supply - it is the circuitry on the motherboard not
> > dampening the noise sufficiently, possibly influenced by the specific
> > characteristics of the 3.5" drives involved, or other devices installed
> > in or connected to the computer and the load/dampening effect they
> > impose on the +12V rail.
> 
> In regard to issue, thanks David for the explanation, I will go home
> tonight and disconnect the 3.5 drive and see what happens. Any idea
> how do shield/surpress the sound circuit from the this?

I haven't looked at this in detail, just observed the connection between
the audio noise and polling of the 3.5" drive.

Someone may have done a detailed analysis (with a fast storage
oscilloscope) but I haven't gone looking for one. I'm only speculating
that it is the +12V rail, as that seems the most likely source.

Assuming I'm right, the noise is either a dip in the +12V rail due to
sudden current drain, or some form of ringing or other digital noise in
that rail. As is only produces a brief click it is high frequency, which
suggests a low-pass filter may fix the problem.

At the simplest, a reasonable sized capacitor on the +12V rail near the
drive connector may be sufficient, as it will effectively decouple the
+12V rail by supplying inrush current when the drive motor is activated
briefly.

If further filtering is needed, it is likely to require a more serious
motherboard modification, such as also inserting a low value series
resistor in series with the +12V rail going to the audio circuitry, or
an inductive filter may be required. This will depend to a major degree
on the motherboard layout, which I've never investigated in detail.

I don't remember enough about the electronics involved in filtering, so
perhaps someone else can comment further.

Another possible explanation is the analog audio signal lines are
picking up digital noise from nearby tracks on the motherboard which are
involved in accessing the disk drive. If so, there is little that can be
done to improve it short of a better motherboard layout.

-- 
David Empson
dempson@actrix.gen.nz