| What is going on when your train won't go around the | | | | If you suspect weight is dragging your train down, try |
| track? Can you make a lighter train or a different | | | | to make sure you are always pulling the train from the |
| locomotive go around the loop? If so, then you've likely | | | | front rather than pushing from the middle or rear. If the |
| pushed the limits with the heavier train that won't. | | | | slack is not pulled tight from the front, all of the cars |
| BrickTrainShop examines these limits to help you pick | | | | ahead of a pusher motor will wobble, creating extra |
| and choose where to focus your efforts to get your | | | | drag, as they are forced by the rails to go forward. |
| trains moving. | | | | While friction is your enemy in train cars, it can be your |
| Let's begin with a little perspective; take a look back at | | | | friend in the locomotive. Increasing the weight on the |
| most of the 9v line and you will see lightweight train | | | | motor keeps your wheels from spinning. A purist can |
| sets with two to four cars, and often the cars did not | | | | use the LEGO train weight (part 73090) or simply build |
| have bogies. A good example is at the beginning of | | | | your locomotives as solid as possible. Or if you don't |
| the line in 1992. The Load N' Haul Railroad (set 4563, | | | | mind concealing non-LEGO within your model, you can |
| profiled in this issue) had three cars, each with only | | | | use coins or other metal as ballast (here in the US I |
| two axles. This tradition continues to the remote | | | | prefer nickels since copper pennies are more likely to |
| controlled Cargo Train Deluxe (set 7898) released in | | | | oxidize). But be careful not to over do it, since in the |
| 2006. Another trick you see in the 9v train sets is the | | | | ballast is still weight the motor has to pull. |
| fact that they often make use of panels and windows | | | | You may also encounter friction in some unexpected |
| to increase the volume while keeping the weight down | | | | places. The train buffer beam with plow (part 45708) |
| (e.g., sets 4559 and 4561). The LEGO train motors and | | | | introduced in 2003 has very close clearance with the |
| controllers were made for these lightweight trains. The | | | | track. On a perfectly flat layout it is not a problem. But |
| choice made sense since the sets only came with a | | | | as soon as you encounter uneven track, the bottom |
| small oval of track and many kids would not add to it, | | | | of the plow can drag across the top of the rails. At |
| so the trains would be short and the motors did not | | | | best, it will simply result in a high-pitched squeak, but it |
| need to be powerful. | | | | can also result in a derailment. After balancing all of the |
| A notable exception to the trend is the Super Chief | | | | various weights and frictions, once you assemble the |
| locomotive and cars (sets 10020, 10022, 10025) | | | | train together, perhaps the most critical point in your |
| released in 2002. In comparison to the 9v trains that | | | | heavy train is at the rear of the last locomotive. Just |
| came before them, they weigh a ton. After these sets | | | | like real trains, the longitudinal force on the drawbar is |
| came out we set up a small layout in the corner of the | | | | the largest here. Likewise, with a long train, you will |
| warehouse. While the floors looked level, the Super | | | | need cars and locomotives with sufficient longitudinal |
| Chief made it clear that they were not: uphill slow and | | | | strength to withstand such forces. While the front cars |
| downhill fast. To pull five cars and two locomotives we | | | | need to be strong, you can still use the weaker cars, |
| needed two motors and they seemed to be straining | | | | but they'll have to ride toward the rear of the train, a |
| at their limits. | | | | technique that is also employed by real railroads. One |
| Since those days we've gone on to pull very long and | | | | advantage of LEGO trains is the fact that it is so easy |
| heavy trains using the 9v system. Our personal record | | | | to swap out bogies (unless they've been carefully |
| is 51 bogied cars and four locomotives over uneven | | | | integrated into the model). So you can motorize a few |
| track at NMRA 2007. There are many challenges to | | | | cars and thereby distribute motors throughout a train |
| running such long trains, the first of which is simply | | | | to reduce the longitudinal forces. For e.g., returning to |
| having enough cars and track to do it. But along the | | | | the Super Chief for a second, I often wondered if the |
| way we had to overcome many other challenges that | | | | extra space under set 10025 was provided to allow |
| you'll likely encounter with just six or seven cars. In fact | | | | you to insert another motor. In any event, distributing |
| at home, we can only comfortably run trains with 6 or | | | | motors throughout the train does add the risk that if |
| 7 cars. | | | | the locomotive derails, the rear of the train will continue |
| COLORING WITHIN THE LINES | | | | pushing cars off the track. This fact might not be |
| No matter what the size of the layout is, weight will | | | | important if your layout is on the floor but it could be |
| always be among your top concerns. If your surface is | | | | disastrous if your track hugs the edge of a table. In the |
| even a little uneven, as it was back in my apartment, | | | | end, everything comes down to power and the need |
| the motors will have to pull the weight of the train uphill. | | | | to get the electric power to the motor to move the |
| So generally speaking, the lighter you can make the | | | | train. If you are having problems, first check to make |
| train cars, the happier your motors will be. Even if you | | | | sure there are no breaks in continuity either due to an |
| are able to find a completely level surface, weight still | | | | unplugged wire, switches being lined incorrectly, or two |
| comes in to play by increasing friction and inertia, which | | | | track segments pulled apart. The controller is supposed |
| we will get to in a moment (no pun intended). Keeping | | | | to put out a fixed voltage, Vcontroller. Each segment |
| the weight down is always a good starting point. But | | | | of track the current has to travel through before |
| generally, the more realism you strive for, the heavier | | | | reaching the motor will drop the voltage seen by the |
| your cars will be. As a result, we personally prefer 6 | | | | motor. There is a miniscule voltage drop along an |
| wide cars for operational reasons, but we drool at the | | | | individual track segment, with a greater loss at the joint |
| detail you can put into 8 wide. Sometimes you can | | | | between two track segments, and a net resistance |
| figure out tricks that give you the realism with little or | | | | per segment, Rtrack_segment. The greater the |
| no weight penalty; it just becomes one more factor in | | | | number of track segments between the controller and |
| your designing process. Some of my first MOCs were | | | | train, the greater the power loss. If the voltage at the |
| boxcars, and as a result of my linear thought process | | | | motor is too small, the motor will not move. After n |
| the sides wound up being composed of alternating | | | | track segments, the power reaching the motor is |
| rows of plates and bricks. Needless to say, the density | | | | roughly: pmotor = Vcontroller 2 ⋅ Rmotor Rmotor + |
| of plates is higher than bricks and much higher than | | | | n⋅ Rtrack _ segment ( )2 In other words, power |
| panels, so these original cars weigh a lot. Later, when | | | | roughly drops inversely proportional with the square of |
| we wanted a few more boxcars, we redid the design, | | | | the number of track segments between the motor |
| and built the walls out of panels. The new cars are | | | | and power connection to the track, i.e., 1/n2. Be sure to |
| about two-thirds the weight of the originals, but look | | | | see things from the electrons' perspective. If there is a |
| virtually identical from the outside. While balancing | | | | switch lined in the opposing direction then current can't |
| aesthetics, weight, and functionality, it is hard to beat | | | | flow that direction around your loop. A diverging switch |
| the train base for its ratio between weight and | | | | next to the power connection can make for a very |
| longitudinal strength. Friction increases with weight and | | | | long distance that the electricity has to travel before |
| it crops up in several locations on a LEGO train - most | | | | reaching the motor; it has to flow all the way around |
| importantly the wheel-sets, and secondly bogie rotation. | | | | the loop to get to the train, losing power with each |
| Looking through the Lugnet archives, we clearly were | | | | track segment. So make sure to check that all of the |
| not the onlyone having problems with the weight of the | | | | switches are lined correctly. |
| Super Chief cars. The wheel-sets (part 2878) are | | | | Does your heavy train stop in spots? Can you |
| designed to have a needle bearing, riding only on the | | | | improve performance at these spots by moving the |
| points at the end of the axle and thereby minimizing | | | | power connection closer to them? If so, you are |
| the friction surface. In a discussion started by Reinhard | | | | probably losing too much power along the track. When |
| "Ben" Beneke, various AFOLs quickly found a design | | | | I have a choice, I try to put the power connector on |
| flaw in the wheel holder. Apparently at some point a | | | | the up-hill side of the layout, to ensure the least power |
| third party manufacturer of the train wheel holder | | | | loss when the train needs it most. If you have a large |
| changed the design without telling LEGO, and as a | | | | enough loop, there might simply be too much of a |
| result, the wheel flange would rub at the 10 o'clock and | | | | power drop to overcome through conventional |
| 2 o'clock positions. Older versions of the wheel holder | | | | methods. You can do some quick experiments to |
| were measured to be 0.9 mm thick, but by 2002 the | | | | determine where the problems lie. While adding a |
| thickness had grown to 1.1 mm and thesethicker | | | | headlight on the locomotive is extra power loss from |
| holders were the source of the problem. | | | | the motor, it is a great indicator as to whether the |
| Reportedlythe design problem has been fixed, but | | | | motor is getting power and the intensity of the light |
| even in a brand new LEGO set a given brick may | | | | should show you just how much power. Next, does a |
| have been manufactured many years ago. | | | | single locomotive make it all the way around the |
| As already evident, the train wheel-set has evolved | | | | track? Then you should have continuity. Does it do so |
| since 1992 in small but important ways. Another big | | | | at slow speeds? If not, you might have dirty track (that |
| change came in 2006. Prior to that year, a metal axle | | | | darn inertia helped you get past at faster speeds). A |
| passed through the plastic wheels and provided the | | | | pencil eraser should help clean the track but be sure to |
| needle in the needle bearing. So the exact position of | | | | clean the right part of the rail. The 9v motors are a little |
| the wheels on the axle was variable. With these older | | | | odd compared to most model railroad motors. They |
| wheels you may have to adjust the spacing to make | | | | do not take power from the top of the rail, they take it |
| sure the wheels do not rub on the wheel holder when | | | | from the inside of the rail. From the shape of the |
| they spin. If you notice any problems, you may also | | | | motor wheels, the most critical spot is the inside- top |
| want to check the wheel spacing to make sure the | | | | corner of the rail. I've found a single sweep with an |
| wheel-set rides well on the track, particularly if your | | | | eraser across this corner on each rail is usually |
| layout includes switches or crossovers. The easiest | | | | sufficient to clean the track. If your eraser leaves a lot |
| way to check is to simply put the wheel-set on the | | | | of dust and droppings, follow it up with a soft cloth to |
| most restrictive type of track you have, i.e., crossovers | | | | clean them up. On a side note, I have also found that |
| and switches if you have them, otherwise, | | | | my locomotives with two motors under one baseplate |
| straighttrack is fine. Starting in 2006, LEGO eliminated | | | | seem to dirty the track quicker than two motors under |
| the "floating wheels" on the axle. Now the metal axle | | | | separate locomotives. Assuming it is not simply due to |
| stops at the backside of the wheel and the needle is | | | | a small sample size, my hypothesis is that this problem |
| molded into the plastic on the front of the wheel. This | | | | arises because when two motors are under a single |
| change should eliminate the positioning problems on the | | | | baseplate they are rigidly fixed together and fight one |
| axle, but the plastic on plastic bearing will likely have | | | | another more than when there is the extra slack in the |
| different performance over its lifetime than the old | | | | couplers between two locomotives with one motor |
| style metal on plastic. Whether you have old or new | | | | each. |
| wheel-sets, inevitably the needle bearing will wear | | | | BURSTING OUT OF THE LINES |
| away the plastic in the wheel holder. As this | | | | So far the discussion has been straightforward. But I |
| wear-and-tear occurs, the wheel flanges are more | | | | was not able to pull that 50 car train without bending |
| likely to start rubbing on the wheel holder. So every | | | | the rules, or <gasp> literally cutting corners. Any |
| now and then flip your cars over, give each wheel-set | | | | deviations from LEGO guidelines are done at your |
| a spin to see how long they keep spinning. A new | | | | own risk, and most of what follows deviates from |
| good wheel-set will continue spinning for up to 10 sec. | | | | LEGO guidelines. So exercise proper judgment and |
| But even after a little use the duration of spin on a | | | | precautions. First, let's return to the wheel holder. |
| good wheel-set will drop to a few seconds. If you get | | | | Whether you have a new wheel-set that drags from |
| almost no residual spin, then you know it is time to | | | | the first day or an old one that has worn out, the |
| repair or replace that wheel-set. And of course keep | | | | AFOLs also devised a solution, namely using a hobby |
| an ear out for rubbing sounds that might be easily | | | | knife to notch out the plastic where the wheels would |
| fixed by repositioning the wheels. If you do have a | | | | otherwise rub on the wheel holder. I've used this trick |
| sluggish wheel-set, don't throw it away. As of this | | | | on almost of my rolling stock. If the train gets too |
| writing, you can purchase individual black wheel holders | | | | heavy, the LEGO magnets can pull apart. Assuming |
| in the United States from the on-line Pick-A-Brick. But | | | | you are running on a loop and don't catch it in time, the |
| even if you don't do anything to a sluggish wheelset, | | | | front of the train can smash into the rear. You can |
| you can always use it under cars in a shorter train | | | | eliminate the magnets altogether and use drawbars or |
| where friction is less likely to be a limiting factor, for | | | | shared trucks (e.g., the center of the TTX car, set |
| static displays under a car that you don't run, or even | | | | 10170), but assembling the train becomes a lot more |
| detailing as train parts around the shop building or | | | | difficult. Another alternative is to use rare-earth |
| payload on a flatcar. Some builders do away with the | | | | magnets. On Lugnet, Mathew Clayson suggested using |
| wheel holder and build trucks that are more | | | | D61 3/8" x 1/16" from K&J Magnetics. Noting that |
| aesthetically appropriate for the given car or | | | | "this size works very well, and isn't too fragile. On |
| locomotive. From everything I've read and seen first | | | | many occasions I've inserted these magnets between |
| hand about such custom trucks, the friction is higher | | | | the standard LEGO magnets on adjacent cars to |
| than the LEGO wheel-sets. I've found that a simple | | | | reinforce the coupling. When placing all of the motors |
| bogie consisting of two train wheel-sets, a 2x6 plate, a | | | | at the front of the train, the forces drop off as you get |
| bogie plate, and buffer is hard to beat for longitudinal | | | | further from the locomotives because there are fewer |
| strength (though its use may mean putting function | | | | and fewer cars being pulled by that coupling. So you |
| above form). But experiment and see what works | | | | only need to reinforce the couplers in the front of the |
| best for you. Returning to the entire train car, curves | | | | train, e.g., my 50 car train had these magnets between |
| slow LEGO trains down just like they slow real trains | | | | every car for roughly the first 30 cars. Rumors of |
| down. There are two forces acting on a train car in a | | | | other AFOL's using glue to stick the magnets together |
| curve, the first being momentum trying to force the | | | | have floated around, but "glue" is a four letter word. |
| train car straight ahead, pressing against the outside rail | | | | Now let's return to that power equation. There are |
| that is forcing the car in a new direction. The second | | | | other ways to keep n small without shrinking the size |
| force being the friction on the bogie plates as they | | | | of the loop of track. If the problem is simply a long loop |
| rotate. In my nonscientific experiments I can't say | | | | of track, and not a heavy train, two or more power |
| which of these forces dominates, but both appear to | | | | connectors (part 5306) from the same controller to |
| contribute. Aside from making your trains lighter, there | | | | opposite ends of the loop can shrink n in the |
| is little you can do about momentum, since you have to | | | | denominator of the equation and reduce the power |
| turn sometime. In fact, at times the momentum will help | | | | lost to the track. But care must be taken to get the |
| your motors past dirty spots in the track. For reducing | | | | polarity correct between the two power connections. |
| the impact of the bogie plate friction, you want the | | | | Using multiple power connections also helps keep the |
| contact between the rotating truck and the car body | | | | train speed more even around the loop. When the train |
| to be as smooth as possible, e.g., using the bogie plate | | | | is really heavy or has more than two motors, a single |
| (part 4092) or tiles for your contact. Based on my | | | | controller probably cannot supply enough power. So |
| experiences we've found that LEGO trains slow more | | | | instead of using multiple power connectors from a |
| in "S"-curves than they would in an equal number of | | | | single controller, two (or more) separate controllers on |
| curve sections all bending in one direction. Since LEGO | | | | a single track will increase the available power. But it |
| track has fixed radius curves, this rotational friction only | | | | becomes that much more important to have the |
| comes at the junction between curve track and | | | | polarity correct with the power connectors, and all of |
| straight track, or curve track in opposing directions. | | | | the controllers should be set to the same level and |
| Reducing the number of these transitions will also | | | | same direction. All four of the engines pulling my 50 car |
| reduce the drag. For both momentum and bogie plate | | | | train had a single motor and power was supplied by |
| friction, you can reduce the impact of curves on your | | | | two controllers connected on opposite sides of the |
| train simply by reducing the number of curves your | | | | layout. While the train made it around the loop under its |
| train might be in at any given moment, spacing the | | | | own power, I had even more operational success |
| curves far apart, with long straight-aways in between. | | | | when I cut back to 47 cars and added a fifth |
| If you think your track layout is causing significant slow | | | | locomotive. In between, I added a third controller to |
| downs, a good rule of thumb for shorter trains is to | | | | supply enough power to the motors. The 47 car train |
| never have the angle between the front and rear of | | | | ran for an hour before we replaced it with another |
| the train exceed 180 degrees at any time, and for | | | | train. |
| longer (heavier) trains, try to get it down to 90 degrees. | | | | |