|This review appeared in O Gauge Railroading
Magazine, February, 1999. Reproduced with permission of The Myron J. Biggar Group.
ATLAS O EMD SW-8/9 SWITCHERS
2 and 3 Rail Versions Are Visual/Operational Delights
Review and Photos by Fred M. Dole with assistance from
Bill Culliton and Barry Lewis
EMD SW-8/9 switchers for 2 and 3 rail operation. Engines
feature a die-cast chassis, body and truck side frames with separately applied plastic and
metal trim pieces. Engines feature horizontal motors with flywheels. 2-rail versions
feature directional lighting, Kadee compatible scale couplers, NMRA DCC socket, insulated
wheels with 8 wheel power pickup. 3-rail versions include Lionel compatible operating
couplers, directional lighting and an electronic reversing unit and sound system
designed by Dallee Electronics, Inc. Sounds include horn, bell, prime mover sounds
including air release sounds and compressor air pops. Available in a number of road names
(the version [SW 8 or 9] selected for a particular road name is dependent on which model
ran on that railroad with tow different road numbers available for each road name.
Suggested retail price is $349 for the 3-rail version with sound and $249 for the 2-rail
version. Manufactured in China for Atlas O LLC, 603 Sweetland Ave., Hillside, NJ 07205.
One of my favorite locomotives has always been the die-cast
frame NW-2 switcher produced by Lionel back in the early 50s. It was a good looking
close to scale engine that ran great and was perfect for my not-very-large childhood
layout. Several of them, in various custom painted roadnames, continue to run on my layout
today. So, when I heard that Atlas Os first locomotive was to be the SW-8/9 switcher
I was delighted. Now I could add more switchers to my motive power roster and have the
latest in electronics and sound.
By the way, to prevent lots of mail, I know the NW-2 and the
SW-8/9 are not the same engine. The NW-2s were produced from 1939 through 1949 and the SW8
and SW9 switchers were built roughly from 1950 to 1954. The SW8 was an 800-horsepower
unit, and the SW9 had 1,200 horsepower. Units that replaced the SW8/9s were the SW600,
SW800, SW900 AND SW1200. The only variation in the bodies seems to be in the number of
louvers in the battery box behind the cab. The SW8 had a full-length front radiator grille
and a single centered stack; the SW9 also had the full-length grille but two centered
stacks. Predecessor switchers from EMD, the NW series, had a half-length front radiator
From my 3-rail tinplate heritage, though, the two models are
very similar in terms of appearance and size, and since Im not a fanatic, it
doesnt matter if it is a NW-2 or an SW-8/9. I just like these switchers.
For most model locomotives (especially 3-rail), the design
process is one of compromise. You start with two lists: the things you have to have, and,
secondly, the things youd like to have. For most 3-rail O gauge diesel locomotives,
the "have-to-haves" seem to be a pair of vertical can motors and an existing
sound system, and the "would-like-to-haves" (the areas where the compromises
occur) are scale dimensions and prototype realism.
For its first locomotive, Atlas O turned this process on its
head. The must-haves in the Atlas design process were absolute fidelity to scale
dimensions and scale detailing. Initially, the designers planned on using a standard
3-rail drive train with twin vertical motors and an off-the-shelf sound system. However,
the EMD SW-8/9 diesel switcher is a locomotive with very little space in which to hide a
mechanism. Early on, it became clear that Atlas would have to do some "thinking
outside the box" (as they say today in corporate America) to shoehorn a mechanism
under the hood of this switcher. Because this engine has full cab interior detail, even
down to lighted gauges and a brake hand wheel inside the cab, that space wasnt
available for motors or electronics.
Whats fascinating is how Atlas pulled it off. Rather
than compromise prototype looks or dimensions, they elected to make their compromised in
three other areas: motor/drive train, electronics, and speaker size.
For the motor/drive train, Atlas chose a single horizontal
motor with two flywheels instead of the normal two vertical motors. This may result in
some loss of pulling power, but if you keep in mind that this is a switcher and not a road
engine, there is plenty of power available for switching chores.
For the electronics, tight space required a custom design
incorporating the reverse unit and sound system on a single small board. There also
wasnt room for a 9-volt battery, which most other systems use to keep the
sound/reversing components alive during direction changes when the track voltage goes to
zero; instead the designers used the largest capacitors that would fit into this space.
They arent, however, large enough to keep the system alive for very long (this is
what accounts for the need to tap the direction button very quickly; more about that in a
The third area of compromise was the speaker for the sound
system. Most diesels have a 2" speaker in the belly of the fuel tank. However, a
switcher has a smaller fuel tank to begin with, plus Atlas has made the air tanks and
piping on either side of the fuel tank. However, a switcher has a smaller fuel tank to
begin with, plus Atlas has made the air tanks and piping on either side of the fuel tank
fully 3-dimensional, so the tank is too small for a speaker. Therefore they went with a
tiny 1" oval speaker facing up through the grilles on top of the front of the hood.
Although a 1.5 watt rated speaker was chosen to maximize the output from such a small
space, there is only so much sound you can get out of the speaker. Again, more about that
First and most importantly, one has to say that this is a
good looking little engine. As I said earlier, Lionels NW-2 has always been one of
my favorite engines. If I were to give the Lionel engine a mark of 70 or 80 for looks, the
Atlas engine gets a solid 100. The overall appearance and attention to detail makes this
engine a visual winner. Atlas has also done their homework in designing the little details
into this engine. For example, the switchers are equipped for multiple unit (MU) operation
only if their prototype ran them that way. Thus, the Texas and Pacific engines are
MU-equipped with MU cables as well as working drop steps and proper end railings to permit
an engine crew to walk between engines. The non-MU Boston and Maine engines, on the other
hand, have solid end railings and lack the MU cables and drop steps.
The end railings, with full bolt detail, are a flexible
plastic, which will take more handling than metal without getting bent, and are molded in
the correct color for each engine. These railings have to be installed by the purchaser
(an easy job). Replacements will be stocked by Atlas just in case.
The detail is state of the art for the 1990s with see through
grilles for the grids on the top of the hood (which provide an outlet for the sound on the
3-rail models) and the same feature on the steps and the footboards. The cab has
windshield wipers both front and rear, so the two little enginemen who inhabit the cab can
see on moist days.
Paint jobs have also been well researched for maximum
realism. For example, for the background color of the "Minuteman" logo on the
Boston and Maine switcher, Atlas found the original paint specs called for imitation
aluminum which is what they used. This color tended to fade to white, which is what you
often see in photos. Their attention to prototype paint schemes is also why you wont
find the NYC switcher in lightning stripe colors (even though many NYC fans have requested
it). The NYC didnt paint its SWs in that scheme. If you really want one that way,
Atlas does offer an undecorated version that you can custom paint.
Overall, the paint jobs on the several different roadnames
weve looked at are crisp with clean letters, numbers, and heralds; even the EMD
builders plates are readable.
If there is any place where one could find fault with the
paint job it is in the area of the radiator grilles. The grilles themselves are black
plastic but the area behind them is painted whatever the body color of the particular
locomotive is. And this color, of course, shows through the open mesh of the radiator
grille. This background color is especially apparent on our sample Western Pacific engine.
It is hardly noticeable on the dark green Pennsy and the maroon B&M engines. It would
have been nice if the area behind the radiator grille could have been painted black.
Quality of construction is superb, and well thought-out for
easy disassembly (witness the easy traction tire replacement mentioned elsewhere). One of
our engines also had a drooping front coupler when delivered, and this was easily
corrected by removing and tightening the coupler; a feat that Barry Lewis accomplished
with a small Philips screwdriver without removing the body or front truck. A full
exploded view of the engine is included in the instructions.
Both 2-and 3-rail versions of the engine have excellent
pulling power due to all wheel drive, 4 traction tires (3-rail only) and hefty weight from
the die-cast frame and hood.
The maximum load in the tests that Barry performed on Marty
Fitzhenrys layout was 13 lighted aluminum 70 MTH passenger cars. That would
equal at least 25 or more typical 3-rail freight cars. And it can really CREEP with a full
load. When overloaded, the 3-rail engine threw a traction tire which, however, was very
easy to replace because the truck side-frame comes off by removing two screws, giving easy
access to the wheels without any other disassembly. The dual flywheels on the single motor
make this engine operate very smoothly and coast beautifully.
Bill Culliton tested the 2-rail version on the Stamford,
Conn. O Scale Club layout. A random 20 cars of various vintages from the freight yard was
tied on the SW9. All of these had a minimum weight of 12 ounces. In running over most of
the train drew 1 amp of current but stalled on a 2-percent grade. So five cars were cut
off, and the train made it up all grades, provided it entered the grade with some speed.
After 45 minutes the current draw dropped to ½ amp.
One area that is of some concern is the way the reversing
unit on the 3-rail engine works. We tested the engine using the MTH Z-4000 transformer,
the Right-of-Way transformer, the Lionel Trainmaster system and a ZW transformer, and we
had the same problem with all these power supplies.
That problem is this: the normal sequence of the reversing
unit is forward-neutral-reverse. When you bring the engine to a stop using the throttle
handle, the reversing unit should cycle into neutral. Then when you cycle the handle
again, it should cycle into reverse. What we found was that many times the engine would
skip the neutral-reverse cycle and start again in forward. At first we thought this was a
defective unit but more experimentation led to a different conclusion.
Once we began using the direction button to control the
engine instead of the throttle handle and once we learned to tap the direction button
quickly and not hold it down the engine performed flawlessly.
Whats causing this is that lack of a battery that we
mentioned earlier or the lack of space inside the engine for larger capacitors (which
Dallee calls a "microcontroller system"). The capacitors which are installed
dont have enough current storage capacity to keep the electronics alive for more
than a second or two and if you hold the direction button down too long the
capacitors juice drains off and the engine reverts to the startup position, in this
case, forward (you can also set the system to start in neutral but this doesnt help
solve the direction changing problem. The only difference would be that the system would
recycle back to neutral instead of forward). The instructions that come with the engine
suggest boosting the power in neutral for a moment before sequencing the direction. Doing
this, Dallee says, gives the capacitor a charge that will get it through the direction
Now this light touch requires some getting used to but once
you get the feel of the engine and whatever throttle system youre using, youll
be delighted with the way this engine operates. It will provide you with hours of slow,
realistic, back and forth switching.
Barry Lewis has these comments about the various power
"MTH Z-4000: Meters made this an easy
throttle to use with this engine. I was able to get pretty consistent reversing by taking
care not to throttle down below 6.0 volts running (or 7.5 volts while in neutral
same handle position) and use ONLY the direction button for reversing. This allowed nice
slow-speed switching. The one thing you cant do is use the throttle handle to get a
really slow move-out from neutral. You need the direction button. If I tried to use the
handle up/down for reversing, it was difficult to find reverse.
"Right-of-Way transformer: This unit does
not have a reverse button and it was nearly impossible to get engine into reverse by using
throttle up/down movement.
"Lionel Trainmaster: With some
experimentation, I was able to find a slow speed that would allow reliable/consistent
reversing with the direction button, provided you tapped it very quickly; below that
speed, operation of the reverse button was inconsistent; but the engine is capable of
beautiful slow speeds with this throttle, which goes down smoothly to zero, unlike the MTH
which drops from about 6 volts direct to zero. Another thing: Once you get the feel of
this, you can intentionally use the direction button to go from forward to forward (or
neutral-before-forward back to neutral-before-forward) by holding down the button rather
than tapping it."
I got pretty much the same results with my own testing using
the Z-4000 and the Trainmaster systems. The key to success is to hit that direction button
quickly and momentarily. Do that, and the engine goes from forward to neutral to reverse
every time. In addition, I tested the switcher with a ZW and found that it was almost
impossible to get the forward-neutral-reverse sequence to work properly using the throttle
handle. You just cant move it quickly enough to keep the unit from cycling through
to forward. But, using the direction button, I was able to get the engine to reverse most
of the time. The ZW direction button was not as reliable as the direction button on the
Z-4000 but this was more due to the age of my ZW and the way the direction button operates
sliding down, if you will, instead pushing down. I wasnt unhappy with the
engines performance on the ZW, however. The clue to success is operator ability and
that comes with practice.
I guess the fix for this operating touchiness would be to
somehow program a longer cycle time into the reversing unit or increase the size of the
storage capacitors. We already know there probably isnt room for larger capacitors
and since programming is way out of my league, I dont know if that answer is even
possible. So I guess learning how to work with the engine is the best solution. And with
some practice, thats very doable. Hey, no one ever said being an engineer was easy!
When you do get it down pat, this little engine will provide hours of fun doing switching
The engine has directional constant-voltage headlights with
lit number boards. For total realism there is no light spillover.
Unlike most 3-rail locomotives, there is no cab lighting.
Again, this was done for realism. Its like an automobile at night: the only interior
light in a locomotive cab comes from the instruments and they ARE lit on this model.
The sound system in this switcher is quite different from
what we have become used to in the 3-rail industry (i.e. LOUD sounds). One of the things I
often experience when I go out to photograph layouts is that many operators have their
sound systems set at the highest possible volume and after a few minutes of listening to
several engines running around those pikes I have a splitting headache. The truth of the
matter is that we dont need to have such loud sounds. But because that, rightly or
wrongly, LOUD has become the "norm", some operators are going to feel the Atlas
O switchers sound is too low. It is definitely not a screamer like RailSounds, and
indeed the engine (prime mover) sounds may be inaudible over the train sounds when the
engine is moving with several cars, especially if your track is not insulated from the
table top by roadbed.
I think Atlas has struck a nice compromise: the engine sounds
are at a pleasant but not intrusive level at idle, and the horn and bell sounds are loud
enough at running speeds. Using the Trainmaster throttle, Barry found that the sounds were
especially nice in slow-speed switching: The motor revs up hard to get the engine moving,
then settles down as the engine reaches switching speed.
The sound volume is adjustable (the screw is at the bottom of
the exhaust stack) but even at its highest point, you are going to magnify the direction
sequencing problem Ive already talked about because the higher volume requires use
of more of the capacitors stored current and will cause it to discharge even
quicker. Dallee recommends setting the volume at about the one-half mark.
The horn is a typical single-chime yard diesel. It was not
supposed to be melodious like a 5-chime road diesel but rather provide a very audible
warning. One nice thing about this horn is that it will blow as long as the horn button is
pushed. And when you let go of the button, it will stop. This means that you can operate
this horn prototypically, using really short blasts to duplicate the way an engineer would
signal his crew about train movements.
The bell is excellent, and always stops at the end of a ring,
never in mid-ring, no matter how you operate the bell button. If you have an older
transformer without a bell button, you can install the standard Lionel No. 6-5906 bell
Another sound included in this system are the air release
sounds heard whenever a locomotive pulls out from a stop as the air brakes are released.
Barry felt that this sound wasnt as effective as it could be and attributed this
mainly to the small speaker. It is that speaker size, without any doubt, that does limit
the effectiveness of the sound system. Dallee has done a fine job with the sounds, but no
matter how good they are, you just cant expect a 1" speaker to perform as well
as a larger one would.
Compressor air pops can also be heard at random during slow
speed operation as moisture is released from the air storage system.
It would have been nice to have a larger speaker. But even
with the 1" unit that designers had to settle for, the sounds produced by this little
switcher are very realistic and, quite frankly, a lot more soothing to my ears than the
cacophony of sound I so often hear on layouts where every engines sound system is
set so high the lighting fixtures vibrate.
Bill Culliton feels strongly that his little piece and other
2-rail versions like those from MTH and Sunset Models (aka 3rd Rail) will be
the salvation of the 2-rail scale genre. He says this:
"In the 1930s many of the men and I say men because
there were almost no women in the hobby then were machinists or people whose trades
required they work with machine tools. They could take rough castings, which today we
would only use for boat anchors, and machine them into running model engines. Today there
are more people in the hobby, but more of us are involved in clerical or service jobs and
dont have the time or the skill to embark on building a locomotive."
"Now I know there are those O scale people who hate
Lionel because it isnt scale (according to their interpretation). By that extension
they will not tolerate any piece of equipment that appears in 2- and 3-rail versions,
since in their eyes this makes it tinplate in a pejorative sense. But as for the rest of
us I think its great. If the importers make their money on the 3-rail market, good
for them. I can live in 2-rail with the slight accommodations made for the
Actually, in the case of the Atlas switcher, those hard-nosed
O scalers will find little accommodation made for the 3-railers. It is the 3-railers who
will find that some things theyre used to are where compromises have been made. This
is, first and foremost, a scale engine and it has been adapted for 3-rail use and not the
other way around. And, by-in-large, Atlas has succeeded with that approach. Being a new
company, Atlas O was not locked into pre-existing tooling and they could approach this
project with total freedom. They have made the decision, rightly I think, to create the
engine around scale dimensions and detailing as opposed to building it around a given
powertrain and electronics. Had they done that, the end result probably would have been a
duplicate of the original Lionel NW-2 in size and detail.
Bottom line: my treasured Lionel NW-2s are going to spend a
lot of time sitting on yard tracks and this handsome little Atlas O switcher is going to
get a lot of operational hours doing switching chores on the Westchester & Salmon
If this first locomotive from Atlas O is any indication of
whats coming in the future, both 2- and 3-rail O gauge operators are in for some