OT : looking for info on "axis avoider" or "cable crossing" mechanism

I'm looking for links (info, patent) on a mechanism inventend on
World War 2 to let cables cross others, to avoid submarines mines to
be dragged. It is also called "axis avoider" (translated from french)
as it allows a rod to move "through" another one.
I learned about it at university some years ago, but could not find
any info on the net (it is probably called differently...)
I want to design the mechanism in SolidWorks. It consists in a star-
shaped wheel inserted in curved rails like:
-(*)-
when an obstacle (cable, axis) hits the star *, it rotates freely in
the ( ) rails, ensuring continuity of the rod and letting the
obstacle cross...
Did you see this somewhere ? Thanks!
Philippe Guglielmetti - www.goulu.net

 

I only heard of this one time. The context was in Richard Feynman's
book, "Surely You're Joking, Mr. Feynman". Engineers were joking about
a "shaft passer" to allow gears to overlap. Then some mention was made
of the mine warfare application.

You may need to dig deep into military history to find it. Perhaps
start inquiring with the naval types.

 

Thanks to those who pointed me to Richard Feynman's book where it is
mentioned as "shaft passer". However, If you search for "shaft
passer" on Google, you won't find many useful info, so I expect my
new page about it to rank first very soon ;-)
I made a model of it with SolidWorks 2004, available on http://www.
goulu.net/article.php3?id_article=42 at the bottom of a small article
with pictures. I'll try to drive the entire construction from the
blue shaft position and diameter, as it looks like the ratio between
the wheel diameter and the shaft's has to be fairly high. The
tolerances analysis and manufacturability of the whole is another
thing...
But now I meet a simpler problem as I expected to use the "physical
dynamics" SW feature to animate the model, but it fails : when I try
to move the wheel, I get an error saying physical dynamics can't
solve it, even if I only specify one red star and the blue shaft as
components... Looks fairly simple though... any idea ?
Philippe Guglielmetti

 

So it looks like a mechanism that allows a fixed position shaft to be placed
inside the periphery of a large wheel and the wheel still be able to rotate.
But, I don't see what is supposed to attach the hub of the wheel to the
outside rim. Yes, I see the 6 small star shaped devices, but how do they
transmit torque to the outer rim? Am I missing the obvious?

WT

 

Am I missing the obvious?

It's easier to envision if you replace the "stars" with solid disks?

That is interesting. I'm trying to envision a similar device placed in a
mine anchor cable and having a bit of trouble with that. 8~)

Cool stuff. I once saw a series of books (Ingenious Mechanisms?); wonder
if they have anything on it there?

 

...... similar device placed in a mine anchor cable ...

Oops! Wasn't paying attention to the details of the channel sectors.
Wonder if they were ever employed.

 

But now I meet a simpler problem as I expected to use the "physical

Looks like the problem is in the component pattern for the stars.
Suppress those and just the parent works fine with Physical Dynamics.
Dissolve the pattern, and fix the subs and it the whole thing
simulates great on 2005.

SC

 

I'm not sure that this is the intended way of using this design,
but theoretically, it will work if the 'shaft passer' just transfers
motion from one gear to another (rather than driving it).

Imagine a three gear scenario with the shaft passer in between
the two. The hub would be stationary. Very comical!

Now all you need to do is add some electromagnets for some
real silliness.

This reminds me of some attempted "perpetual motion" devices
I've seen.

Mike Wilson

 

Yes, thanks to some unrealistic mates. In reality, the only thing
that should be mated is the inner hub center hole to a shaft,
assuming the rails are welded to the spokes. Gravity would
not be kind to this assembly.

Upon further inspection, I see that the stars actually do transfer
the load. The images on Philippe's website made me think that
only one point of each star was within the rails. If two star points
are in the rails at all times, then it will work.

I'm guessing that the contraption would be more stable if it
were wider?

Pretty cool Philippe!

Mike

 

30.09.2004 23:55:51

no, no! the "goal" is much simpler !
Read Feynman's story and watch the original minesweep avoiding
mechanism on http://www.goulu.net/article.php3?id_article=42
FYI, I know a watchmaker who does very compact mechanisms and
"complained" designs required "bars" to separate layers of wheels.
(see http://www.zvisuel.com/montrespassion/english/index.html for
nice 3D anims of watch mechanisms)
I mentioned the "shaft passer" as a joke, saying there was always a
more complicated way to solve a problem...
Please downlaod the model from there and tell me how to make it
"work" using "physical dynamics" : normally, if you select the red
and blue components, you should be able to rotate the wheel, but
SolidWorks can't do it. Why ?

 

When I opened the assembly, I had mate errors in 2005. There are
some in-context features and flexible subs. Your best bet is to
make the this all one assembly (no subs). Don't use any patterns
and use only three stars instead of six.

Try and simplify it as much as possible. SolidWorks is trying to
calculate too much in your assembly.

I started to do that here...
http://www.mikejwilson.com/misc/avoider.zip

The simulation is already set up. I eliminated many mates to give
real world results, so don't drag any parts around! Let Physical
Dynamics do the work.

Mike

 

I opened the "shaft passer demo" and let the physical dynamics do the
job.....worked like a charm from start in SW 2004

Krister L

 

Sorry....should have been SW 2005....even the small stars move as the thouch
the passing shaft (which they don't do when I try Mikes fixed assy)

Krister L

 

OK, I simplified it and made a motion picture (rated 'G')...

http://www.mikejwilson.com/misc/shaft-passer.zip

Mike Wilson

 

Thanks. Philippe posted a picture of the device in a cable (heck of a
deal). Now, I wonder if Paul Ratajczak had that 1946 copy of Popular
Science and remembered seeing the article? 8~)