[R-390] Autopsy of a bad PTO.

[Jim Miller]

I just noticed some of the photos in my Cosmos PTO article on Dave's page
have become corrupted.  If anyone's interested, the original article is at
the following url.  Copy anything you want.

http://home.att.net/~jamesmiller20/cosmos.htm

I tried emailing Dave but his email bounces.  I guess I like the Cosmos over
the Colling PTO because it appears to be relatively easier to linearize
despite Phil's nightmare.  I spent over a week (couple of hours a day) going
through the stack (trial and error) in a Collins PTO, taking the cover off,
tweaking, putting the cover back on, over and over.  I can see why the
inventor of the Cosmos approach thought he had a better idea.  Sounds like
someone in a past life got heavy handed with a screwdriver and jammed some
of the little screws.

----- Original Message -----
From: "Barry Hauser" <barry@hausernet.com>
To: "Phil Atchley" <k06bb@elite.net>; <r-390@mailman.qth.net>
Sent: Saturday, June 28, 2003 7:28 AM
Subject: Re: [R-390] Autopsy of a bad PTO.


> Hi Phil & gang:
>
> I dunno, there was a time there would have been several posts with
detailed
> "been-there-done-that" posts to assist you.  I've held back on replying
> simply because I've never been that deep into a Cosmos.  Soooo.. .in the
> absence of any experienced replies .... (maybe some were off-list?)
>
> Somehow, I'd also assumed that with all the downloading you mentioned you
> had certainly come across Dave Medley's pages or another site with Jim
> Miller's "A JOURNEY THROUGH THE COSMOS (PTO)".
>
> The direct URL on Dave M's site is http://www.davemed.com/cosmos.html
This
> includes detailed photos, though there's something wrong with one or two
of
> the photos.
>
> Gonna try to shed some light on this, but again, bear in mind that I have
no
> hands-on, so take it for what it's worth.
>
> >Since there is another PTO coming I went ahead and took this one apart
all
> >the way down to the corrector disk to see if I could determine the
failure
> >mode (not sure I can get it back together correctly 8^)
>
> Somehow, I'm not surprised.  I always take your "quit" and "throw in the
> towel" declarations with a grain of salt. (about the size of a football
;-)
> While you were tearing into that PTO, I was probably searching for that
web
> article (or copping some Z's).
>
> >Anyway, I was
> >right.  The 48 tuning screws are rotated one by one past a round spring
> >loaded thingy made of spring steel that has a small piece of steel that
> >looks kind of like a "footbridge" spot welded to it.  As the screw for a
> >particular segment of the frequency range (25 KHz steps) walks over the
> >"footbridge" it presses down on it and pushes the slug under it further
> into
> >the coil.
>
> The web site photo shows this.  The spot welded piece looks more like a
> triangle, so I'm not sure what you mean by a footbridge.  (Maybe like the
> ones in Japanese gardens and Central Park?) Check out the photo for what
the
> shape of it should be.
>
> >The "footbridge" was damaged in two ways.  First it had a BIG dent in it
on
> >one side that would make screws riding on that side of the bridge
(they're
> >staggered) not press down as hard on the bridge. Secondly, the spot weld
> >that holds one end of the bridge is broken loose so that it has lost some
> of
> >its "stiffness" and positive action.
>
> >In trying to fix the "dent" in the bridge I'm afraid that I also weakened
> it
> >some more as spring steel doesn't take kindly to dents and trying to
> >straighten them out.
>
> From the photo, it would seem that bridge is supposed to be rigid, not
> flexible, even if it's made of the same spring steel as the ring. (Just a
> manufacturing convenience, vs. a solid piece of something, also avoidance
of
> dissimilar metals. Also, spring steel is fairly hard and less prone to
> wear.)  Wouldn't it be solid if that other weld weren't broken?
>
> >The third problem I found was that some of the screws had apparently been
> >turned "counterclockwise" too hard, spreading the slot and screw too wide
> to
> >thread down into the disk far enough to reach the damaged bridge.  These
> are
> >the culprits that probably had no adjustment affect as they wouldn't
reach
> >the "footbridge".  These are very tiny screws that are really only a
screw
> >shaft with a "slot" in the end and a flat head on the backside to "walk
> >across the bridge".
>
> Can't you turn them back down again while it's apart, to make sure they'll
> make it through?  Sounds like what you are describing is a false bottoming
> out of the adjustment screws.  The "no adjustment" effect might also be
due
> to the "bridge" being out of shape, partly flattened and flexing flatter
due
> to the broken weld on one end and/or the dent.  Actually, if that bridge
or
> arch was flexing, that would have messed up your linearity adjustments and
> may have contributed to the other problem.
>
> >NOTE:  A couple articles that I read said that there was a flexible
Teflon
> >ring that these screws rode on forming kind of a cam.  This one had no
such
> >ring, only the spring steel thingy that the screw heads walked across.
>
> Was that a Cosmos PTO in the articles?  May have been a different mfr.
>
> Again, keeping in mind the nature of the source, (not
> been-there-done-that) -- here's some partially lame advice:
>
> From the photo, it looks like the bridge should be solid, not flexible.  I
> imagine the screws should just catch the bridge on the rise, not at the
base
> near the spot welds.  If so, then you should be able to repair the thing
> with epoxy, or perhaps a solder repair, if the spring steel will take
> soldering.  One way would be to fill the void in the bridge -- basically
an
> "arch support".  Reshape the arch as best as possible, using the photo or
> another unit as a guide.  Cut the corner off a piece of sheet metal --
> aluminum -- to make a small triangle support with a slightly rounded apex
to
> fit the underside of the bridge and secure the whole business with epoxy,
> using a small clamp to hold in place until it sets.  Basically -- an
> orthothic shoe insert.
>
> Then fix the screws somehow, if they actually need fixing.  The photo on
the
> website shows that quite a few screws are either not present or in the
fully
> backed out positioin.
>
> There's another thing.  I was a bit confused as to how this setup could
work
> if the screws are riding up and down the arch, with gaps in between,
causing
> reciprocating action.  That would mean that it would be nearly impossible
to
> have smooth action in terms of degrees rotation to frequency.  The key is
> that the adjacent screws are straddled so that the tip of the bridge/arch
is
> always riding on a screw or none at all.
>
> But -- what that means is that the screw setting should end up with
> reasonably smooth transition from one to the next.  If there were too much
> of a differential from one (or a pair) to the next, it could jam or cause
> roughness.  I would think that extreme differences between adjacent screws
> might be indicative of a problem elsewhere -- worn lead screw or
something.
>
> Also -- does it seem possible to run a screw down far enough to actually
> crush the bridge piece (or snap a weld)?
>
> The photos on that site are of fairly high resolution.  I blew up the side
> view of the "bridge" and it certainly appears to me that it should be
fairly
> or completely rigid, not flexible, if both welds are intact.  It's
basically
> triangular in shape, but with a smooth bend at the apex. With one weld
> broken, it will flatten under pressure and you won't get a reliable
> adjustment.  That's my theory anyhow.
>
> Of course, as I'm typing the last of this speculative tract, Phil is
waiting
> for the epoxy to cure or soldered the thing, or maybe hand carved a new
ring
> and "footbridge" out of some spring steel in his junk drawer.
>
> "Throw in the towel"  -- yeah, right.  We know you better than that. ;-)
>
> Barry
>
>
>
>
>
> _______________________________________________
> R-390 mailing list
> R-390@mailman.qth.net
> http://mailman.qth.net/mailman/listinfo/r-390