Re: Freight Car Distribution on smaller RRs

Mike Brock <brockm@...>
 

Dave Evans writes:

"One way to use the N-G distribution model is to assess "plausibility"
for the presence of foreign road box cars on your layout. Needing one
train with 15 SP box cars doesn't instantly drive you into a huge box
car fleet."

Right. I could simply strongly over weight the number of SP box cars in order to be able to generate X4005. Also, the other data shows a much greater representation of SP box cars so it would not be a problem. The 1956 box car data [ 18 trains ] shows:

UP: 157
PRR: 48
NYC: 45
SP: 43
Milw: 22
CNW: 17
GN: 17
SR: 13
B&O:13
NP: 13
ATSF:11

Note the much higher % of SP box cars compared to Pennsy and NYC. Both Pennsy and NYC cars should be double that of SP if the G-N theory holds. BTW, the '49 data supports this relation between SP, PY, and NYC.

"But unless the train ran several times a day, it makes a
strong case for letting the X4005 train stay in staging some days, and
perhaps fiddling the other box cars in that train. If you want to
emulate N-G, then I would simply segregate the SP cars dedicated to
this train from the rest of your fleet balance targets. So now you
need 15 SP cars for X4005, and 100 for the rest of your trains to
maintain N-G."

Which is exactly what I do. Great minds....

"To my thinking, the intent of N-G is make a model railroad visit more
like a stop along the prototype's track. If you randomly stoped along
sherman hill in your era for 3 hours (more if you run a fast clock),
would you expect to see X4005 every visit? IF yes - run it every session."

Possibly...but it wouldn't look the same.

"As for the MWR car, same concept, how may 3 hour visits to sherman
hill before you spot it? One, ten, one hundred?"

Well...each visit would for our purposes see 8 trains...about 240 box cars. If randomly distributed, in each group of 7400 box cars I would expect to see one MWR car. That would be 247 frt trains. IOW, I could expect to have to do 31 visits to expect to see one MWR car. UP, however, is moving 35 frt trains in one day, 1050 box cars, and the MWR car should appear once in each 7400 or once each 7 days [ the shortened time span from 10 to 7 is due to an increase in 1953 to 100 MWR cars from 75 ]. So, while a bum sitting beside the tracks might see an MWR car every 7 days [ and wonder what the heck it is doing there ], our illustrious frt car evaluation team, being whimps and only checking during a 3 hour time period, would expect to observe for 31 days to see what the bum saw in 7. Makes you wonder who should be in charge...

"Stats would suggest it
would be pretty infrequent."

Yep. And if you dozed off...

"Same for the Ann Arbor, D&H and other
small roads (although they would appear much more often than the MWR,
but certainly these roads would not be seen every visit)."

D&H? 2963 cars. In every group of 7400 box cars, 30 D&H cars, 185 trains, 23 visits....To fit the N-G theory I would need about 2 cars if I had a 500 box car population. In my current population....8 trains, about 100 box cars...alas...I could only have a fifth of 2 cars...perhaps everything but the trucks.

"My lessons learned from this thread are:

1) N-G should apply to mainline trunk routes where the nation's
traffic is traversing a layout."

I agree if one applies the "Close Association SP Syndrome" fudge factor.

"6) Having cars from small roads is perfectly acceptable - but they
should be fiddled in - it could be on a dice roll, or more analytic.
It seems that having an extra 30-60 box cars from smaller lines (and
some of these may be well known lines for a smaller fleet), is all
that is required to provide some sense of "randomness" to trains
entering the layout from staging."

I agree.

"I think the concept has some very positive merit. The good news is
that having 50% of your fleet from the 11 major roads is probably
pretty easy to accomplish, unless your fleet is primarily resin, in
which case I think you can do whatever you want ;-)"

"The interesting finding is that because 50% of the boxcar fleet was in
11 roads, the "fiddle" pools do not need to be huge - 30-60 cars
depending on the size of the layout and the length of the PP's memory
(don't let them collect on-layout wheel reports!) And it is quite
reasonable to run that MWR car once every few sessions. Just don't put
it in a string of D&H, B&M, D&LW, AA, and RDG box cars, or no one may
notice it!"

Yep.

Mike Brock

Join main@RealSTMFC.groups.io to automatically receive all group messages.