Marx 1998 Diesel ALCO S-3 Switcher Performance Testing

I have a couple of Marx 1998 diesel locomotives - they resemble the ALCO S-3 diesel switchers that were produced in the 1950s. Back in the day, the S-3s weren't especially powerful, could only output about 660 hp, so it was mostly used as a yard goat - to move cars short distances around a rail yard. The Marx 1998 is like this - not an especially capable puller, but they are still fun. One of my 1998's has the Union Pacific livery, and has, for lack of a better term, a "Pullmor" motor - with aluminum side plates and two rubber traction tires. The other 1998 has an AT&SF livery, and this locomotive doesn't have the Pullmor motor. So the side plates are made of steel, and the drive wheels do not have any rubber traction tires. I thought it might be fun to do some performance testing of these locomotives and share my results with this forum. I will begin with a description of some salient mechanical/electrical features of the Marx 1998 locomotive.

The 1998 locomotive is 9" long. The real S-3 was 45 ft long, so, the Marx 1998 works out to be 1/60th scale, based on length alone. However, for purposes of my testing, I am going to consider the 1998 to be a standard O gauge locomotive, i.e. 1/48th scale. The locomotive without the Pullmor motor weighs 1 lb, 11.8 oz, while the Pullmor loco weighs a bit less - 1 lb, 10.1 oz. The difference probably a result of the steel vs aluminum motor side plates. The locomotive appears to have both front and rear four wheel trucks. However, it has a 2 wheel, single idler axle front truck, a 4 wheel motor, and a 2 wheel, single idler axle rear truck. It is really clever the way Marx designed this. It looks like an AAR (Association of American Railroads) type A wheel configuration, but would actually be considered an AAR "1-D-1" meaning there are three trucks under the unit. At either end are trucks with one idler axle; and the center truck has four powered wheels. The Pullmor motor unit has two of the four drive wheels with rubber traction tires. This makes a pretty substantial difference in pulling power. Using an Ohaus 8004-ma pull-type hanging spring scale, I measured the drawbar pull of each of these 1998s, and the Pullmor variant could draw 8 oz, while the non-Pullmor unit could only draw 6 oz. I removed the shell to take a look at the mechanical and electrical components. The motor is a series-wound, universal (AC/DC) motor. Both locomotives have the same gear train. The gear ratio is 6.4 to 1. I was able to determine this by counting the gear teeth and doing the math. The diagram below shows the gear train.

I bench tested the Pullmor locomotive with a 90 watt Lionel 1033 transformer and measured the drive wheel rotational speed using a Cen Tech digital photo sensor tachometer. At 13.5V the wheels where spinning at 1115 rpm and the locomotive was drawing 1.8 amps. At 8V, the wheels ran at 260 rpm and the motors drew 1.6 amps. Since I don't have the ability to directly measure the rotational speed of the motor, and since this locomotive has a gear ratio of 6.4 to 1, I calculated a motor rpm equal to 7,136 rpm at 13.5V and 1,664 rpm at 8V.

Likewise, I bench tested the non-Pullmor locomotive, and at 13.5V the drive wheels were spinning at 1150 rpm and the locomotive was drawing 1.9 amps. At 8 volts, the drive wheels were spinning at 615 rpm and the motor was drawing 1.75 amps. I calculated a motor rpm of 7,360 at 13.5V, and 3,936 rpm at 8V.

Now let me describe the test track and testing equipment. I have a loop of 0-31 gauge tubular track, and the total center-rail length is 13.4434'. I have a 90 watt Lionel 1033 transformer, and I have power applied in two separate equidistant points along the line. I have a digital multimeter to monitor the track voltage. I have a clamp-style amp meter to monitor the amperage from the hot wire between the transformer and the track. For the time trial, I used the stopwatch feature of my Android phone, and timed the train over 5 laps. I tested just the locomotive at the lowest and highest speeds possible without stalling or tipping over. Outside of this very narrow band of voltages, the locomotive either would not move or would move so fast I was afraid that I'd derail and damage it. I then connected the consist and repeated the time trials at the lowest and highest voltages that I could safely operate at. Below are the velocity calculations and power draw calculations:

For the locomotive without the Pullmor motor, and no trailing consist:

For the locomotive without the Pullmor motor, and a trailing consist of 5 boxcars:

For the locomotive without the Pullmor motor, note that at 9 VAC, the locomotive alone was running at 81.2 MPH (scale), drawing 2.2 amps, 19.8 watts. When I attached the consist (five boxcars) to the locomotive and ran it at 9 VAC, it slowed down to 33.1 MPH (scale) drawing 2.3 amps and 20.7 watts. I also bench tested the locomotive at 9 VAC, and it was drawing 1.75 amps and 15.75 watts. So, I prepared a pie chart showing the power profile at 9 VAC. As you can see, the majority of the power (76%) was consumed overcoming mechanical/electrical inefficiencies of the locomotive, while 20% of the power was consumed moving the weight of the locomotive. The remaining 4% was consumed by pulling the cars. See the chart below.

For the locomotive with the Pullmor Motor and no trailing consist:

For the locomotive with the Pullmor Motor and a trailing consist of 7 freight cars:

For the locomotive with the Pullmor motor, note that at 8 VAC, the locomotive alone was running at 81.6 MPH (scale), drawing 1.9 amps, 15.2 watts. When I attached the consist (seven boxcars) to the locomotive and ran it at 8 VAC, it slowed down to 49.5 MPH (scale) drawing 2.04 amps and 16.32 watts. I also bench tested the locomotive at 8 VAC, and it was drawing 1.6 amps and 12.8 watts. So, I prepared a pie chart showing the power profile at 8 VAC. As you can see, the majority of the power (78%) was consumed overcoming mechanical/electrical inefficiencies of the locomotive, while 15% of the power was consumed moving the weight of the locomotive. The remaining 7% was consumed by pulling the cars. See the chart below.

Below is a video of the Marx 1998 without the Pullmor motor. As you can see, this locomotive is really struggling to pull a consist of 8 freight cars. In order to get it moving, I really had to jack up the throttle and then once I built up some momentum, I rolled off on the throttle. You can see the wheels slipping as the locomotive struggles to get up to speed. One other thing worth mentioning is that the locomotive is very light, and since it doesn't have as much inertia as a heavier locomotive would have, it has a tendency to stop very abruptly. I have found that the abrupt stops can sometimes cause derailments in the consist. Also worth mentioning is the tendency for the these locomotives to go fast. As you can see from my calculations, the slowest I was able to run the locomotive was 19.3 mph (scale). This speed is probably much too fast for real-world shunting or switching duties.

Below is a video of the Marx 1998 with the Pullmor motor. As you can see, this locomotive is pulling a consist of 12 freight cars. It gets up to speed with very little drive wheel slippage. This locomotive is also very light, and since it doesn't have as much inertia as a heavier locomotive would have, it also has a tendency to stop very abruptly. I have found that the abrupt stops can sometimes cause derailments in the consist. Also worth mentioning is the tendency for the these locomotives to go fast. As you can see from my calculations, the slowest I was able to run the locomotive was 19.4 mph (scale). This speed is probably much too fast for real-world shunting or switching duties. At a fairly low voltage (9VAC) the Pullmor locomotive hauling 7 freight cars was running at 99.8 mph. I didn't want to run the locomotive at this speed for too long lest it fly of the rails on a curve.