Date: Mon, 12 Jul 2004 22:42:34 -0400
Reply-To: David Beierl <dbeierl@ATTGLOBAL.NET>
Sender: Vanagon Mailing List <vanagon@gerry.vanagon.com>
From: David Beierl <dbeierl@ATTGLOBAL.NET>
Subject: On thinking, was Re: simple question
In-Reply-To: <vanagon%2004071211205820@GERRY.VANAGON.COM>
Content-Type: text/plain; charset="us-ascii"; format=flowed
Since people more practical than I quickly gave useful answers...here are
some thoughts, if you can call them that, about getting to an
answer. Since it says a good bit on a number of levels about how I think
-- and write, and teach -- I'm extremely interested in any remarks or
discussion on any aspect of it. And if it turns out to be genuinely useful
to anyone that would be pretty neat too. Oh yes -- there's one factor I
know I left out in considering the flatness of the cover. Anybody spot
it? Or the ones I *don't* know I left out?
:)
d
At 11:20 7/12/2004, Bike Florida wrote:
>nonetheless. Any ideas what to try here? Have I opened the proverbial can o
>worms?
No. Those covers hold the oil in and keep idle hands and hyperactive
pebbles out. No structural function IM(semi)HO that's worth noticing. The
flange and cover have to be flat (or matched) to within the limits of the
gasket's conformability, and any scratches on either surface have to be
wide enough and flat enough ditto. The spring has to compress the whole
package enough to get compliance from the gasket. The gasket has to meet
its own specs. It all has to stand up to hot oil but not any noticeable
pressure. It mustn't shed crumbs or strings into the system. That's what
it had when it left Wolfsburg...either supply the lack, i.e. flatness or
surface finish or compliance or pressure, or compensate for it by messing
with the gasket dimensions. Oil is unforgiving because it has no surface
tension to speak of -- hmm, don't want to saw that limb off just
yet. Well. It is. Unforgiving, i.e. will ooze and seep through tiny
places, because -- yep, not much surface
tension. Cra-a-a-ackkkk!!! <g> Of course compared to say liquid
helium...not going there, thank you. Welcome to this little glimpse of my
mind, hope you're as amused as I am.
Actually measuring flatness is a lot harder than you might expect and can
lead into metaphysics if your mind runs that way. Carpenter-level
understandings of "close enough" -- well, fine joiners are starting to use
machinists' measuring tools, so one end of carpentry intersects the other
end of machining. But Watt took ten years of agony and boatloads -- entire
fortunes -- of other people's money to get one lousy full-size model built
of a simple steam engine that he could already show you running in the palm
of his hand (so to say) -- because nobody could build him a piston and a
cylinder that fit together, or quantify why they didn't etc etc. At least
that's Andrew Carnegie's view on it and I defer to his
judgment. </blither> Have to get better at judging how much of this stuff
is actually worth writing down...
If you don't have an accurate "feel" for it from experience and don't have
the relevant design specs or access to other methods that give an actual
answer you can get to someplace usable:
Gasket compliance is X many thou per inch of thickness, takes Y psi to get
there. A Post-It note is about four thou (.004 inches) thick. Cork-rubber
gasket hmmm, google cork gasket physical properties, ok, it says here
cork/rubber 25% compression at 2.8N/mm^2, what's that in something I
comprehend, google? -- ok, a Pascal is a Newton/m^2, hmmm ok for feathers
maybe, so changing to that from mm^2 would be 1000 x 1000 x 2.8 or 2.8
Megapascals, lessee MPa to psi is times 150 roughly so now we've got 25%
compression at 450 psi loading, sanity check <poke at gasket material in my
mind> yes, feels plausible.
Pressure is total spring load / area of the sealing surface,
lessee. Spring load hmmm, translate muscle-memory of working it...ok, 50
seems on the low end of possible and 300 seems darned unlikely. WAG of 150
lb for now? Sealing surface 3/8" wide or so by <shape the cover with my
hands and look at it, maybe four inches on the two diagonals by eight and
ten on the long sides close enough> 34 inches boy that seems a lot, sanity
check, shape it in hands again and deform into a circle yep, pi * d seems
to work so 3/8 x 35 is 105/8 is about nine square inches. 150/9 is 15 psi,
really, that little? 300 lb spring and quarter-inch rim makes it x2 x1.5
or 45 psi, seems to be an outer limit. So maximum compression on the
gasket something like 2.5% and minimum noticeably less than 1%. Sanity
check, how deep was the indentation on the old gasket and why did I do all
this figuring anyway if I could get there from measuring the old
one? Beats me, dumm I guess.
Well I don't actually have an old one, so keep on WAGging. Gasket
thickness abt 3/16 or .2" -- half per cent of that .002 OUCH, it better be
more than that!! No way even the Germans would work that hard just because
they can. I insist on their allowing at least +/- .005 on the cover,
probably more -- that's a total maximum allowance of .020 or five Post-Its
thickness. Sanity check, that's ten per cent of a 3/16 gasket, would need
150 psi compression and I figured 45 psi max. So...either the cover and
case each need to be flat within +/- .002 or better -- nah, I refuse to
believe it -- or I'm wrong about the spring pressure (or made some
mistake). Back to memory lane, unh, unh, SNAP! Ow, darn it! mmmm...10:1 or
so on the lever, 300 lb would still be 30 lb and I need lots more...ah Ha!
the cover is an inclined plane exerting the outward force, I'm just prying
it along the ramp with the screwdriver. Ok, I conditionally accept that
the spring could be exerting 1000 lb of force. Maybe. Intuition rebels,
says the wire is too thin and deflection not enough. Accept that for now
-- implication is they really do have to be pretty darn flat. To make sure
(from where I am now) I'd better be able to gauge it to +/- .001 which
seems outrageous. I could actually do that, easy enough as such things go
(i.e. not *that* easy) for the case but the stupid cover has a lip around
it, bah. Ok, search for crumbs (straws?)...would it be close enough
measuring the outside? -- depends on the stamping, but maybe. But -- maybe
a cover is way out? Twisted, say? If I could set it on four equal-height
points it would rock if it were twisted...equal? Not so easy when you're
working in thou, that's why they make surface plates and dial gauges and
stuff...
I've gotten to where all the alternatives are expensive...more information
will help clarify things. There's one piece that should be readily
available given a big pool of people -- how deep is the indentation on a
used gasket? I'll ask the list and plug that back in. Mphm...even that's
not utterly trivial to measure but if I can get an answer for how thick the
compressed part is that will work. And so forth...
Ok, exercise over, discussion period follows. All remarks welcome,
including "What are you, nuts?"
--
David Beierl - Providence RI USA -- http://pws.prserv.net/synergy/Vanagon/
'84 Westy "Dutiful Passage," '85 GL "Poor Relation"
| 
