[Prev][Next][Index][Thread]
Re: Rubber toroids [ graphite ]
From: Wes A Brzozowski[SMTP:wesb-at-blue.spectra-dot-net]
Sent: Tuesday, October 28, 1997 5:14 AM
To: Tesla List
Subject: Re: Rubber toroids [ graphite ]
On Sat, 25 Oct 1997, Tesla List wrote:
>
> From: Harri Suomalainen[SMTP:haba-at-cc.hut.fi]
> Sent: Saturday, October 25, 1997 8:45 AM
> To: Tesla List
> Subject: Re: Rubber toroids [ graphite ]
>
> On Fri, 24 Oct 1997, Tesla List wrote:
> > The problems of this method stretch beyond mere dependability. It's fine
> > as a classroom demonstration in a beaker, but that's about it. Since the
> > rubbed graphite isn't actually attatched to the work peice in any way,
>
> I have a bottle of commercial sprayable graphite. It contains some sort
> of binder (not specified), colloidic graphite and the solvent making
> it thinner.
>
> The manufacturer tells it has good adhesion to glas, plastics and other
> even surfaces. It is primarily used for repairing cathode ray tubes
> (tv's, monitors). Manufacturer also mentions "excellent solution for
> galvanotechnics".
With all due respect, this description *sounds* like the sales pitch that
would have been printed on the bottle; it makes it sound good without
commiting them to details. For example, what does "good adhesion" mean?
If it's going to be stuck to an item mounted inside a cabinet, like a
monitor, and never touched after that, then "good adhesion" may just mean
that it won't fall off after being handled a few times. Consider though,
that a toroid will get bumped, jostled, scratched, and will suffer
numerous tiny hot spots from the electrical discharges, which will cause
severe stresses if the coefficients of expansion don't match very closely.
This is not to say that commercial conductive paints are ineffective, but
that there are a lot of unanswered questions to their use.
> >From the experience I can tell the graphite is as *very* fine powder.
> The amount of binder seems pretty low. There is just enough binder
> to make adhesion good. Main component is definately graphite. Hope
> this tiny bit of info helps all the people not buying the commercial
> stuff.
This sets off all sorts of alarms in my mind. I've experimented with lots
of conductive laquers, and have found that if the graphite loading is too
high, you get a surface that's as soft as, well, graphite. This means that
every little scratch, bump, or ding will dig a chunk off the surface.
And it does.
This is fine for an experimental setup that's going to be used for
taking a single set of data, but a little disappointing if it's to be
used for long-term service.
I've gone to a mix of graphite and conductive lampblack. The neat thing
about lampblack is that it's usually the pigment in all regular black
paints. It's already of sufficient particle size that it requires no
further milling to get it to pigment quality. I've worked with both
heavy loading of the two (soft surface, easy to start plating) and mixes
that go heavy only toward the lampblack (tougher surface, takes some
finesse to get the plating started, though it picks right up, once a
seed-layer of copper is down on the surface). BTW, conductive lampblack
is a specially treated form of garden-variety lampblack. It's been
heated quite hot for a reasonable stretch of time to cause it's internal
twisted-and-broken structure to "heal" into something a bit more
graphite-like. The fact that it's of pigment quality produces a smooth
paint-like surface, rather than the grainier surface that comes from
using actual graphite. I suppose graphite could be ball milled down to
pigment particle size, but I've never seen it available.
As you might guess, there are a lot of variables that you don't get to
control with a commercial coating that's made for other purposes. There's
certainly nothing wrong with trying those commercial coatings, as the
payoff is sufficiently big. But as with most manufacturing processes that
have a lot of variables, there are often just one or two "sweet spots" in
the variable space where the process is actually practical,
and much of the development process is dedicated to finding those spaces.
As Tesla Coilers, we have the luxury of working with a system that's
fairly hard to make not to work. And once it's working a little, we can
diddle and tweak,
keeping those changes that give improved performance. In so many real
world systems, if you just try a few things blindly, you get poor results.
If you change them a little, you get poor results. If you change them a
lot, you get poor results. Then all of a sudden, a little tweak here or
there gets good results. Then you spend a lot of time in ever decreasing
tweaks, trying to optimize that sweet spot. If you find you've lost just a
little control over one of your variables, suddenly you're producing
batches of scrap and don't know why. The point here is that a great many
manufacturing processes require much much tighter controls to get anything
to work, because the sweet spots are so very, very, very tiny. Once again,
try the commercial coatings, by all means, but don't just conclude that
the concept is impractical if it doesn't work well for you; you will in
fact have tried only a tiny fraction of the available possibilities.
> > Also, since the graphite isn't attatched to the work peice, neither will
> > the layer of plate that you build up. Expect it to chip off readily. If
>
> Good point. However, in surfaces like a ball it is good enough to have
> a surface with full conductive (metal) layer. If it has no cracs and
> covers completely the surface it cannot fall off. However, I'd presume
> it is easy to damage.
I guess now you'll have to define "good enough". If you mean it'll last
for one usage, maybe more if I handle it like glass, then maybe you're
right. Personally, if it can't take the stresses and shocks inherent in
ordinary handling, I'd consider it unacceptable. It's a matter of what
you're willing to settle for, I guess...
> I just descovered one good point in using copper chlorides: if you make
> a lot of PCBs you'll get lots of copper chloride for free. You're able
> to recycle the copper dissolved from from PCBs. I actually did that some
> years ago when I just happened to have no other copper salt around :)
There's a commercial process that does some similar things and might
be of interest to some folks on this list. Copper has two chlorides.
Cuprous chloride is only slightly soluble in water, but soluble if there's
also some HCl. If you blow air into this solution, it changes into the
more soluble cupric chloride, consuming the HCl in the process, and
"recharging" the solution; that is, it can once again etch copper. In the
etching process, it becomes the less soluble cuprous chloride. This is
recharged again by adding HCl and blowing in air. This process is used in
commercial etching operations, because it reuses the same chemicals over
and over. In fact, all the etched copper becomes more copper chloride,
and the problem is in the disposal of all the additional etchant that
the process *produces*.
> I think I've also seen copper plating with *no* copper solution when
> the copper came actually from the copper electrodes dissolving slowly.
> Great if you have pieces of copper to be distroyed I guess. :)
There are many ways to plate metals; some are practical, others aren't.
I'd expect what you're describing to be somewhat slow, and to waste lots
of energy, electrolyzing the water in the bath. In all practical plating
operations a copper electrode is destroyed, though. It's how the plating
bath is replenished, and replacing the plating bath alone is much more
expensive than just replacing the copper electrode. If you add in the
cost of modern disposal of the depleted plating bath, it would become
totally unacceptable. But replacing the copper electrode is not done
all that often, and it's neither hard nor time consuming. No one minds.
Wes B.
*****************************************************************************
* wesb-at-spectra-dot-net * "It's a magical world, Hobbes ol' buddy... *
* * ..Let's go exploring." - Calvin *
*****************************************************************************