 | | | | | | |
| |
|
In lugnet.trains, Brian Kendig writes:
> Are there any web pages which have a study of how to line up Lego train
> layouts with diagonal track?
>
> That is, I want a little more freedom in my layout than forcing it to
> always use ninety-degree curves and parallel/perpendicular tracks, but
> I've found that if I try to be *too* freeform, nothing lines up and I
> end up with two-inch gaps between the pieces I'm trying to connect.
>
> If someone else has already done the math for how to make curves using
> one, two, or three pieces of standard 22.5-degree curved track and then
> get the track to line up properly and snap together firmly without
> having to rely on any 'wiggle room,' that would be a great help to me!
> Thanks!
If you're trying to keep things squared up, the two magic numbers are 7 and 11.
which is to say, a corner that keeps straights lined up with the normal Lego
geometry can be one of 3 things:
4 curves
1 curve, 11 straights, 3 curves
2 curves, 7 straights, 2 curves
There are similar combinations for coming off of switches, but they have the
same drawback as the corners: it takes lots of space.
thanks,
James
| | | | | | | | | | | | | | | |
"James Brown" <galliard@shades-of-night.com> writes:
> If you're trying to keep things squared up, the two magic numbers
> are 7 and 11.
>
> which is to say, a corner that keeps straights lined up with the
> normal Lego geometry can be one of 3 things:
>
> 4 curves
> 1 curve, 11 straights, 3 curves
> 2 curves, 7 straights, 2 curves
>
> There are similar combinations for coming off of switches, but they
> have the same drawback as the corners: it takes lots of space.
This is very interesting... Is there a similar formula for a dog-leg?
1 curve, X straights, 1 curve the other way? What equations/formulae
would you use to compute this? It's been way too many years since I
studied trigonometry...
--Bill.
--
William R Ward bill@wards.net http://www.wards.net/~bill/
-----------------------------------------------------------------------------
Consistency is not really a human trait.
--Maude (from the film "Harold & Maude")
| | | | | | | | | | | | | | | | | | | | |
William R Ward wrote:
> This is very interesting... Is there a similar formula for a dog-leg?
> 1 curve, X straights, 1 curve the other way? What equations/formulae
> would you use to compute this? It's been way too many years since I
> studied trigonometry...
I don't have an equation, but I think one curve, 12 straights, one curve
will do a dog-leg that is eight 32-stud baseplates long and 88 studs
wide (3 baseplates minus 4 studs each side).
-chris
| | | | | | | | | | | | | | | | | | | | | | |
Christopher Tracey <ctracey@enviroweb.org> writes:
> William R Ward wrote:
> > This is very interesting... Is there a similar formula for a dog-leg?
> > 1 curve, X straights, 1 curve the other way? What equations/formulae
> > would you use to compute this? It's been way too many years since I
> > studied trigonometry...
>
> I don't have an equation, but I think one curve, 12 straights, one
> curve will do a dog-leg that is eight 32-stud baseplates long and 88
> studs wide (3 baseplates minus 4 studs each side).
THANK YOU!!! This works perfectly! I just tested it in Track
Designer....
--Bill.
--
William R Ward bill@wards.net http://www.wards.net/~bill/
-----------------------------------------------------------------------------
Consistency is not really a human trait.
--Maude (from the film "Harold & Maude")
| | | | | | | | | | | | | | | | | | | | | | |
William R Ward wrote:
> Christopher Tracey <ctracey@enviroweb.org> writes:
> > I don't have an equation, but I think one curve, 12 straights, one
> > curve will do a dog-leg that is eight 32-stud baseplates long and 88
> > studs wide (3 baseplates minus 4 studs each side).
>
> THANK YOU!!! This works perfectly! I just tested it in Track
> Designer....
Does it work perfectly or does it go over in length by at least a half a
stud? I couldn't tell from Track Designer.
-chris
| | | | | | | | | | | | | | | | | | | | | | |
Christopher Tracey <ctracey@enviroweb.org> writes:
> William R Ward wrote:
> > Christopher Tracey <ctracey@enviroweb.org> writes:
> > > I don't have an equation, but I think one curve, 12 straights, one
> > > curve will do a dog-leg that is eight 32-stud baseplates long and 88
> > > studs wide (3 baseplates minus 4 studs each side).
> > THANK YOU!!! This works perfectly! I just tested it in Track
> > Designer....
>
> Does it work perfectly or does it go over in length by at least a half
> a stud? I couldn't tell from Track Designer.
It's close enough that TD considers it to be connected, and that's
good enough for me. I used it for the yard design I just posted (q.v.)
--Bill.
--
William R Ward bill@wards.net http://www.wards.net/~bill/
-----------------------------------------------------------------------------
Consistency is not really a human trait.
--Maude (from the film "Harold & Maude")
| | | | | | | | | | | | | | | | | | |
| |
|
"James Brown" <galliard@shades-of-night.com> writes:
> If you're trying to keep things squared up, the two magic numbers
> are 7 and 11.
>
> which is to say, a corner that keeps straights lined up with the
> normal Lego geometry can be one of 3 things:
>
> 4 curves
> 1 curve, 11 straights, 3 curves
> 2 curves, 7 straights, 2 curves
>
> There are similar combinations for coming off of switches, but they
> have the same drawback as the corners: it takes lots of space.
The 1-11-3 design isn't very close at all, according to Track
Designer. Here's what I did: straight, curve right, 11 straights (5
switches and a straight, but that is equivalent), 3 curves right,
cross-track. Then from the cross-track, a bunch of straights, 4 right
curves, some more straights, and another cross-track. But it doesn't
line up...
However, when I tried the 2-7-2 design I found it to be much closer,
but still not close enough for Track Designer to consider it
connected: 2 rights, 7 straights (3 switches and a straight), 2
rights, and a cross piece, connecting back up to the beginning in the
same manner as above.
--Bill.
--
William R Ward bill@wards.net http://www.wards.net/~bill/
-----------------------------------------------------------------------------
Consistency is not really a human trait.
--Maude (from the film "Harold & Maude")
| | | | | | | | | | | | | | | | | |
In lugnet.trains, William R. Ward writes:
> The 1-11-3 design isn't very close at all, according to Track
> Designer.
What does work is 1-13-3. It's close enough that Track Designer will
consider it a closed loop.
You can turn it into a triangle with a 5-13-7-12-4-5 pattern (starting with
curves and alternating with straights). Then it's easy enough to see that
it forms a lesser know Pythagorean triplet, 5-12-13. The angle formed by
short leg and the hypoteneuse is 22.62 degrees, which matches very close to
the curve track being 22.5 degrees.
John
| | | | | | | | | | | | | | | | | | | | |
In lugnet.trains, John Gramley writes:
> The angle formed by
> short leg and the hypoteneuse is 22.62 degrees, which matches very close to
> the curve track being 22.5 degrees.
Oops. That should be the long leg and the hypoteneuse.
John
| | | | | | | | | | | | | | | | | | | |
| |
|
In lugnet.trains, William R. Ward writes:
> "James Brown" <galliard@shades-of-night.com> writes:
> > If you're trying to keep things squared up, the two magic numbers
> > are 7 and 11.
> >
> > which is to say, a corner that keeps straights lined up with the
> > normal Lego geometry can be one of 3 things:
> >
> > 4 curves
> > 1 curve, 11 straights, 3 curves
> > 2 curves, 7 straights, 2 curves
> >
> > There are similar combinations for coming off of switches, but they
> > have the same drawback as the corners: it takes lots of space.
>
> The 1-11-3 design isn't very close at all, according to Track
> Designer. Here's what I did: straight, curve right, 11 straights (5
> switches and a straight, but that is equivalent), 3 curves right,
> cross-track. Then from the cross-track, a bunch of straights, 4 right
> curves, some more straights, and another cross-track. But it doesn't
> line up...
John is correct; it's 1-13-3 that works. Blame my faulty memory, it's been
a bit since I played in TD.
> However, when I tried the 2-7-2 design I found it to be much closer,
> but still not close enough for Track Designer to consider it
> connected: 2 rights, 7 straights (3 switches and a straight), 2
> rights, and a cross piece, connecting back up to the beginning in the
> same manner as above.
The 2-7-2 is close enough that it connects physically very soundly. The
offset is roughly 1 stud. Easy enough to cover with slop, with this many
track sections involved.
The dogleg number is like the large corner. switch, 13 straights, curve
(back to parallel), then close the loop. This will generate a
close-enough-for-TD connection (which warns you about the short-circut).
A siding can use switch-6straight-curve-curve-6straight-switch to be close
enough for slop to absorb; not quite close enough for TD.
Hope that helps, sorry about the confusion caused by my faulty memory!
thanks,
James
| | | | | | |
