-----Original Message-----
From: STMFC@... [mailto:STMFC@...] On Behalf Of Jim
Betz
Sent: Tuesday, November 07, 2006 9:30 AM
To: STMFC@...
Subject: [STMFC] Derailments, Coupler Swing, and related topics ...



Denny/et al,

I suspect that the size of the head of the coupler is the least
significant issue with respect to reliable coupling/uncoupling.

Has anyone on this list ever seen anything that talks about the
relationships between coupler shaft length, amount of swing in the
coupler (as determined by the width of the coupler box), the distance
of the swing point of the coupler from the mounting point of the
truck, car length, and track curvature ... as all of those relate to
reliable operations. (Yes, I know that wheel profile and flange
depth is related as well ... for the purposes of this discussion
I'm willing to defer that part of the equation to a later time when
it makes more sense ... at least to me.)

For instance, it is common knowledge that 90-foot cars with body
mounted couplers wont go thru an 18-inch radius curve without
putting so much side force on the wheels that they will derail.
Similarly, if your layout has no turnouts smaller than a #10 and
no curves smaller than 30" radius you can reliably operate a lot of
equipment that simply won't do on a layout with more restrictive
curvatures.
The real railroads all had operating rules that 'prohibited' certain
types of equipment on certain routes - and I've seen the crew of 4449
literally walking-and-watching it thru a section of track to ensure
that it didn't derail due to tight curvature. We need the same thing
for our model railroads as well. An example of which is that on the
club I'm a member of we have a section of track known as "the branch"
and that track was built to be reliable with 4-axle diesels and does
not support 6-axle trucks ... and we have designated it as such.

A lot of the products we use are 'compromised' (ie. designed) based
upon the desire to operate on track such as Atlas 18" snap track. Some
of those design decisions are built in to our 'legacy' products. If
we had some kind of table/guidelines that addressed the issues with
respect to a variety of curvatures we'd be able to determine just how
much scale accuracy our layout(s) will realistically tolerate.

So has any one ever seen a table/formula that describes this
geometry in 'realistic fashion' ... ???
- Jim in San Jose