What Michael says here makes a great deal of sense. However, as I have a number
of units which, like the Commodore box, have a transformer of considerable age
in their power supply located near the yoke of the CRT, yet have none of the
wobble described here, it seems to me that one could probably allow the
transformer in the Commodore unit to remain at least long enough to check for
some sort of disruption of the power supply to the sweep circuits. Transformers
should last longer than this 20 years or so, while the capacitors have reached
their shelf-life limits, I think.
Partly, I guess, this bias is because I'm quite certain it will be a lot of
trouble replacing the original transformer, though a kit with appropriate spec's
might be solution. In any case, one can check the power supply connections to
the CRT circuitry without leaving the house. This being Saturday, I'd say
Monday will be soon enough to go searching for a replacement transformer.
Let's keep our fingers crossed ...
Dick
----- Original Message -----
From: "Michael Schneider" <ms_at_silke.rt.schwaben.de>
To: <classiccmp_at_classiccmp.org>
Sent: Saturday, October 06, 2001 4:15 PM
Subject: Re: CBM8032 - wobbly screen (The saga continues)
> I have to disagree.
>
> What Adrian describes are the "classical" symptoms for electromagnetic
> inductions,
> resp. distortions caused by this physical effect. Grounding and
> shielding with tin
> foil does not help, since it's the transformers magnetic field (as
> opposed to an
> electrostatic field, that would be shielded by the grounded foil) that
> deflects the
> electrons of the CRT beam and causes the wobble.
>
> Tha most probable cause is indeed the transformer, more precisely: the
> aging core
> of the transformer. This core usually is made of a lot of thin sheets of
> steel.
> The purpose of the core is to a) couple the primary and secondary coils
> very
> closely and, more important b) to "contain" and keep the electromagnetic
> field
> closely within the transformer. Well, ok, that's a bit simplistic, but
> you'll get
> the idea. The core is not made out of a compact piece of steel since
> stray currents
> would be induced within the steel, heating up the core and wasting
> energy. This stray
> currents ("eddies" they're called in english, i think...) would be
> induced perpendicular
> to the orientation of the sheets the transformer core is made of.
> Thus, these sheets must be insulated by very thin layers of plastic or
> resin or something
> similar.
> Now, this is the point: With old transformers, this insulation starts to
> literally crumble
> away, which leads to a significantly larger stray magnetic field around
> the transformer.
> Later, you can usually **HEAR** this effect: the steel sheets start to
> humm with the
> line frequency. The transformer is close to dead now...
>
> To make this log story short: Yes, a new transformer should help. New,
> like in
> "factory fresh"...
> Even better would be one of these fancy and expensive torroid
> trasformers, since
> they have literally no measurable magnetic field outside the core.
>
> Shielding of the old transformer is no real option. You would need what
> we Germans
> call "mu-Metal", which is a special alloy of steel and other things i
> have long
> forgotten but that make it EXPENSIVE.
> It is magnetic, but not a good conductor, which make it ideal for
> shielding magnetic
> fields.
>
> You could try a shielding out of steel sheets, but they might get hot if
> the stray
> currents induced by the transformers magnetic field are too large. And
> the shielding
> effect is not too good...
>
>
> Oh, and by the way: I'm new to this list... 8-)
> Hi, i'm Michael and i am collecting old VAXen.
> I live somewhere in southern Germany, which is not that bad. No,
> really.... 8-)
>
> Cheers
>
> ms
>
>
> P.S.: sorry for the long post...
> Richard Erlacher wrote:
> >
> > I doubt a new transformer will help this very much. If this box ever worked
> > right, then something's changed, and the filter CAPs are a prime suspect
after
> > all these years. I'd look for a failed diode in a 4-diode bridge too, if
that's
> > what they used. However, what's more likely is a problem in the wiring,
since
> > the wiring path changes as you open/close the lid. Since increased
shielding
> > didn't do much, perhaps you should look at the power connections, as the
cabling
> > is perturbed whenever the box is opened. If you check underneath (on the
solder
> > side) the supply connections on the main board and, more likely, on the CRT
> > board, perhaps you'll see a deteriorated connection or possible damaged
trace
> > that moves around when you wiggle the power wiring. It could even be a
badly
> > seated/latched contact in a MOLEX shell if that's what they used. I don't
think
> > Commodore made the CRT board, and they may have used a pretty shoddy
monitor.
> >
> > You've apparently got to find something that's different between when the
box is
> > open versus when it's closed.
> >
> > Dick
> >
> > ----- Original Message -----
> > From: "Adrian Vickers" <avickers_at_solutionengineers.com>
> > To: <classiccmp_at_classiccmp.org>
> > Sent: Saturday, October 06, 2001 1:06 PM
> > Subject: Re: CBM8032 - wobbly screen (The saga continues)
> >
> > > Curiouser and curiouser...
> > >
> > > Today, I happened to have the machine opened up to effect some keyboard
> > > repairs (i.e. a jolly good clean up). Having successfuly removed all
traces
> > > of key bounce, non-working keys, etc., and feeling justifiably please with
> > > myself, it occurred to me to try a couple of things WRT the wobbly screen.
> > >
> > > 1st attempt: "build" some shielding out of aluminium foil to go around the
> > > transformer. Which I did - all rough & ready stuff, with just a bit of
> > > insulating tape over the transformer connections to make sure they didn't
> > > short out. No effect - the screen wobbled its way along as ususal.
> > >
> > > 2nd attempt: Same as above, but this time I made absolutely sure the tin
> > > foil was earthed, by screwing it onto the chassis via one of the
> > > transformer screws. Same effect as 1st attempt: FA.
> > >
> > > 3rd attempt: Getting desparate now, I wrapped a load of tin foil around
the
> > > signal cables (having removed it from the transformer). Decided to leave
> > > the lid up when I switched on and bingo! Steady as a rock (literally no
> > > movement whatsoever. Refusing to get too excited, I carefully brought the
> > > lid down (holding the signal cables so as not to let the tin foil touch
the
> > > circuit board - bah! The wobble returned.
> > >
> > > 4th attempt: Removed all tinfoil, started up with the lid open, no wobble.
> > > The wobble appears progressively as the monitor approaches the transformer
> > > (the lid down position).
> > >
> > > 5th attempt: Put a sheet of tinfoil over the whole of the bottom of the
> > > monitor section. No effect - wobble remained.
> > >
> > > Clearly, therefore, the wobble is induced by the monitor. There must be a
> > > fairly monster field coming off the transformer interfering with the
> > > electron gun aiming circuit(s), or something like that.
> > >
> > > The question is: What *would* shield it? Do I need some thin steel or
> > > copper, for example? Presumably, the shielding needs to be earthed (not
> > > difficult, there's plenty of earth points around there. Or, would it be
> > > easier to simply replace the big transformer with a smaller modern one?
> > >
> > > TIA!
> > > --
> > > Cheers, Ade.
> > > Be where it's at, B-Racing!
> > > http://b-racing.co.uk
> > >
> >
>
> ------------------------------------------------------------------------------
--
> >
> > >
> > > ---
> > > Outgoing mail is certified Virus Free.
> > > Checked by AVG anti-virus system (http://www.grisoft.com).
> > > Version: 6.0.282 / Virus Database: 150 - Release Date: 25/09/2001
> > >
>
> --
> Michael Schneider email (home): ms_at_silke.rt.schwaben.de
> Schleestr.8 http://www.vaxcluster.de
> 72766 Reutlingen Phone: +49 7121 492781
> Germany FAX: +49 7121 479331
>
> People disagree with me. I just ignore them.
> (Linus Torvalds)
>
>
Received on Sat Oct 06 2001 - 18:15:14 BST