Now that it's been out for a while, I'd like to hear opinions from buyers
about this model. I'm interested to hear about the "level of detail" vs. the
amount of cast-on parts, if there are any really horrible features, etc.
I read the article and today I went down to the Train Shop in Santa Clara,
CA, and picked mine up (UP 3985, price $259.99 + tax).
The article is correct, especially about the quick "attack" on the chuff
sounds. But I find that some of the one time sounds (air release, brake
squeal, sand release, water injector, blower hiss) release to quickly at the
end of the sample, like they just get cut of too fast. It's a minor quibble.
The sound volume has two positions using CV52, loud or soft. You make finder
adjustments using a small screwdriver in the tender. (no disassembly
required, just remove a hatch). Since I don't change the sound much, other
than on/off, that's OK, but many folks might like a CV for the individual
volumes like the QSI controller which BLI uses. The Genesis decoder is from
Model Rectifier Corp and only overall volume can be controlled. The other
feature I'd like is the two position volume control to be one of the twelve
functions. There are two on/off buttons (double click F0, or F12), so one
could be used for volume control.
in article BDD09891.66E8% email@example.com, Edward A. Oates at
firstname.lastname@example.org wrote on 11/29/04 9:13 AM:
I'm not a rivet counter, but it all looks good to me. The MR article
mentioned some detail problem with the fact that both the front and rear
loco drivers swivel (to accommodate down to 18" curves), and that prevent
some steam pipe or another from connecting as they should. But it all looks
good to me.
An operating front coupler is included as well.
in article pWRqd.1422$ email@example.com, DCC Models at
firstname.lastname@example.org wrote on 11/29/04 7:39 PM:
The 'extra' articulation of the Athearn Challenger is typical of all the
mass market articulated models from BLI, P2K, Lionel, and Rivarossi.
The Bowser and Mantua models are/were correctly articulated, as are
'brass' models. This makes them look better, but they require larger
Think about this:
A 33 inch radius curve on an HO model railway is a reasonably broad model railway
curve. Nearly all locomotive models will negotiate such a curve with no problem.
On a prototype railroad this represents a curve of 479 foot (146 M) diameter.
curve can be expressed in its degree of curvature as a 24 degree curve. This is
approaching the limit for 4-axle North American locos. Such a curve would only be
found in extremely tight and restricted areas such as very old industrial
areas like docks and wharves and in places like engine terminals where a track
encircled the roundhouse, for example. The locos would not operate coupled in
cases and would be restricted to "walking" speed.
I would venture to say that there is no prototype articulated steam locomotive
for mainline, class-1, North American railroading, that could negotiate such a
Someone else can reveal what the minimum curvature was for a Yellowstone, or a
or a 4-6-6-4 Clinchfield Challenger, or a Southern Pacific AC-12, or any number
articulated machines, But at anything beyond ten degrees, they started getting
restricted as regards to what they might be coupled and how fast they may travel.
Now; A ten degree curve, which is regarded as a moderately sharp curve on the
prototype, equals an HO scale curve of 79 in (2 M) radius. Rather broader than
model railway curves.
Hence the need to double articulate large, multi-engined, HO scale steam
The CPR had a 23 degree curve at Boston Bar in the Fraser Canyon,
British Columbia, for a while. This was the tightest mainline curve on a
N. American RR. Can't recall when it was broadened, but it's gone now.
The need for tight curves is the reason the B&O Docksider 0-4-0T was
built. IIRC, it could negotiate street-car curves (ca. 50ft radius), but
then the coupler issue appears. On very tight curves, cars were coupled
with draw bars inserted into the knuckles of the couplers. If you
examine older pictures you will see that the knuckle had a slot in it
for this purpose. I recall seeing such couplers as late as the 1950s in
IIRC, some long-wheelbase locos actually spread the rails and derailed
when they were run on curves approaching the alleged design limit.
Recall seeing a photo of a UP 4-12-2 in such a predicament.
The correctly articulated model engines will generally run OK on 36" and
larger curves, but they don't look good IMO.
OTOH, a train gliding around a curve is one of the great sights of
railroading. Hence, John Armstrong advocated "cosmetic curves" of 72"
radius and larger wherever possible. He also pointed out that a yard
built on a wide curve not only looks better than a straight one,
operationally there is no problem if the radius is 72" and higher, and
if restricted to shorter cars and engines, you can get away with a yard
on a 36" curve.
On the prototype, the tracklayers could rarely manage to get access to the
center of the curve. Even if they had, they rarely had a compass big enough,
or even a long enough piece of string line to draw the curve, so they worked
out other means of creating continuous curves. This "degree of curve" seems
to be a US invention and someone, somewhere has a table of equivalents or a
formula of conversion.
20 degrees would seem to equal about 290 feet of radius from the example
A one degree curve has a circumference of 36,000 feet.
One degree of angular displacement for every 100 feet of distance along the
circumference. 360 Degrees X 100 feet/degree = 36,000 feet.
Use the known radius in inches of any model curve to determine the circumference
the complete circle:
2 X the radius, X Pi
divide by 12
multiply by 87.0857 ( for HO scale)
This number (N) is the circumference of the prototype circle in feet that
to the model curve
Divide 36,000 by this number (N) and you have the degree of curvature
the model curve.
A 33 inch radius curve:
33 X 2 = 66 inches
66 X Pi = 207.3451 inches
207.3451 divided by 12 = 17.2788 feet
17.2788 X 87.0857 = 1,504.7329 feet
36,000 divided by 1,504.7329 = 23.9245 degrees, which is
another way of saying 23 degrees 55 minutes 28 seconds
just 24 degrees
There are much "shorter" ways to do it, but this way shows all the steps and the
reasons for taking them. There are also several other ways to get the same
but this, I think, is the simplest because it involves nothing more esoteric than
simple, four-function math. A calculator with a "Pi" function makes it a
If you want a real no-brainer try this:
divide 789.5083 by the curve radius in inches
789.5083 divided by 33 = 29.9245 which approximately equals 24.
If all you care about is the answer, this will do fine.
In fact, it is a tiny bit off because, although highway curves are measured with
100 foot arc, railway curves are measured with a 100 foot chord. the
a one degree railway curve is actually a tiny bit longer than 36,000 feet. For
railway purposes though, it is an insignificant amount and the figure of 36,000
plenty close enough. Especially when you consider that we do not lay our curves
a transit and compass, but most often with a ruler or marked stick.
I'm sure someone will give you the exact formula. The degree of
curvature is the angle of offset (the amount you swing the transit) to
determine the next staking point on the curve using a 100ft chain. It's
a North American thing, since it cuts to the chase: how to lay out the
curve if you haven't got a large enough compass. :-)
The Brits and Yurpeens use radius or diameter, but they have to convert
those figures to - you guessed it - how many degrees to offset the next
length of chain as the curve is staked out on the ground. Takes a fair
amount of arithmetic. I did it one late spring in a surveying course,
years ago when I thought I wanted to be an engineer.
John Armstrong listed degrees of curvature and corresponding radii in
actual feet and scale equivalents on page 42 of "Creative layout
Design". If there's enough interest, I'll post the table here.