Locomotive Power Testing

[jtlan]

Ever wonder how much power you're actually getting out of your locomotives? Commander Wolf and I did!

Of course there's Philo's motor characteristics page, but that only tells you the motor power ratings under the load he tested with, under his conditions. We wanted to see how much drawbar power we were actually getting, particularly with Power Functions locomotives.

Here's our setup:

A string is attached to a pin inserted where the magnet would normally go on the locomotive. The string is attached to a weight, which dangles off the balcony into the floor below (a 3-meter drop).

Here's what we tested:

AGEIR

Power source:

PF rechargeable battery

Motor:

71427

gearmotor

2096

Power source:

2-cell 20C Lithium-Ion battery (custom)

Motor:

2x PF M motor

Baggage Car 999

Power source:

Wall wart

Motor:

2x 9V train motor

HH1000

Power source:

AA battery box with Eneloops

Motor:

2x PF M motor

U30

Power source:

AA battery box with Eneloops

Motor:

2x PF L motor

We did two tests.

Traction Test

We ran each locomotive at top speed. If it moved the weight, we increased the weight until it no longer moved, then recorded the result. We then divided the max weight pulled by the weight of the locomotive to get an effective traction ratio (for 2096, we divided by the weight of the center powered section only). Unfortunately, since our weight was made up of two battery boxes, the minimum granularity in our measurement was the weight of a AA battery.

Results: The locomotives with the PF train wheel, as well as the 9V motor unit, all had about the same traction ratio, from 0.25 to 0.27. No. 2096 had a much higher ratio of 0.337. I'll note that 2096 uses the large drive wheels with custom O-ring tires, rather than the official rubber band. Some locomotives had different traction ratios going forward and going backward.

Power Test

In this test, we fixed the weight at 98g. We then measured how long it took each locomotive to raise the weight a fixed distance. From this we could calculate the power output by each locomotive against that particular load:

mass of weight (kg) * acceleration of gravity (m/s^2) * distance (m) / time (s)

Results:

Note the high values for 2096 and the baggage car -- this is due to their high top speed (large drivers on 2096, speedy 9V motor on the baggage car -- we ran the 9V motors at top speed). We ran the test on our top three winners with 2 more AA batteries of weight:

Note that the power ratings for 2096 and U30 went up more than the power rating of the baggage car did. Interpretation: PF motors can continue to haul heavy loads at decent speeds, whereas 9V motors are gradually slowed by the addition of more weight.

How much is 98 grams of load equivalent to though? The bonus round:

Measuring Rolling Resistance

We hooked up some unpowered rolling stock to the weight, and increased the weight until they began to move. Turns out the normal 9V wheelsets, lightly lubricated, have neglible resistance. The weight of a single 2x4 brick was sufficient to move a piece of 2-axle rolling stock. Of course, there are things with a lot more rolling resistance:

- Commander Wolf's PRR T1: 52 grams worth of resistance

- The entire rake of P54 cars (3 total): 32 grams worth of resistance

Conclusions

• Effective traction is about the same regardless of your power source. If you want more pulling power, increase the weight of your locomotive ... or try to find some tires with more traction. U30 weighs 1020 grams and could lift ~ 250 grams.
  
• 9V train motors produce a huge amount of power ... if you want to go really fast. For pulling heavy loads at more sedate speeds, use the PF motors. We haven't tested the 9V train motors at a lower speed, but observational evidence suggests that their effective pulling power is drastically reduced at lower speeds.
  
• The rolling resistance of lubricated stock wheelsets is almost nothing compared to much lower than the rolling resistance of technic axles. Be prepared to fight lots of resistance if you make custom rolling stock that doesn't use the stock wheelsets.
  

Until next time!

Edited March 7, 2015 by jtlan

[Commander Wolf]

As the "co-author" I'd like to add some additional points:

• The actual weights of the units and their maximum tractive effort:
  

• These maximum tractive efforts are limited in different ways for different units: for the PF locomotives, the limiting factor is typically the amount of friction between the wheels and track, while for the 9v motors, the limiting factor is typically the output torque of the entire drive train.
  

• Units with uneven weight distribution may be able to pull more in different directions. For example, an off-center battery pack gives the U30 a max forward tractive effort of 256g, but a max backwards tractive effort of 282g.
  

• Battery charge level will obviously affect the power output of the engine, but will typically not affect max tractive effort - unless the tractive effort is limited by drivetrain torque instead of the usual friction between the wheel and the track.
  

To me the more interesting calculation is the one that comes after this: determining the power draw at the battery given the power output of a locomotive pulling a certain load at a certain speed, and therefore determining how long the batteries *should* last under those circumstances. Unfortunately that involves the more difficult task of measuring entire drivetrain efficiency, which at least I haven't thought of a good way to do.

Edited February 13, 2015 by Commander Wolf

[edsmith0075]

Thanks for taking the time to post this info.

" The rolling resistance of lubricated stock wheel sets is almost nothing compared to the rolling resistance of Technic axles. Be prepared to fight lots of resistance if you make custom rolling stock that doesn't use the stock wheel sets. "

I feel like an idiot asking this but could you help me identify the best setup for rolling stock. I know with the locomotives the fake/unpowered bogies use Technic axles/pieces and, from my experience, have a lot of resistance. What do you recommend as an alternative?

Ed

[pirzyk]

Thanks for taking the time to post this info.

" The rolling resistance of lubricated stock wheel sets is almost nothing compared to the rolling resistance of Technic axles. Be prepared to fight lots of resistance if you make custom rolling stock that doesn't use the stock wheel sets. "

I feel like an idiot asking this but could you help me identify the best setup for rolling stock. I know with the locomotives the fake/unpowered bogies use Technic axles/pieces and, from my experience, have a lot of resistance. What do you recommend as an alternative?

Ed

Using the assembly like this http://www.bricklink.com/catalogItem.asp?P=2878c02

[Heppeng]

Good stuff! Wait till you try a 4.5V train...

[Dan-147]

If you're going for a look that's different from the standard Lego wheelblock (for example, North American freight cars) and are not a die-hard purist, I would suggest a design by Cale Leiphart that is available on the Railbricks site. http://railbricks.co...freight-trucks/ I use a simplified version on my freight cars. Here is a picture of it beside a regular Lego truck. https://www.flickr.com/photos/14163824@N07/5726592488/in/photostream/. Please overlook the fact that in the picture it uses Kadee couplers. It also works very well with the older Lego magnets. Besides the couplers, the only non Lego parts I use are K+E Engineering 1/8" tubing cut to 33mm length and K+E Engineering 0,081" brass rod cut to 40mm length. This rod fits snuggly on the regular Lego wheels without any glue so it can be disassembled at will. The rolling resistance of this setup is comparable to Lego wheelblocks. You can also reduce even more this resistance by using graphite powder.

Dan-147

Edited February 13, 2015 by Dan-147

[jtlan]

• The rolling resistance of lubricated stock wheelsets is almost nothing compared to the rolling resistance of technic axles. Be prepared to fight lots of resistance if you make custom rolling stock that doesn't use the stock wheelsets.
  

This is what I originally wrote. I woke up this morning and realized this wasn't entirely accurate, as the P54s use popped-out wheelsets, and the T1 is a heavy locomotive with a lot of axles. I replaced the two stock wheelsets on the test car I had measured with technic axles and remeasured:

• Two axles with Big Ben Bricks small wheels took three 2x4 bricks to move.
  
• Two axles with the Power Functions train wheels (small) took four 2x4 bricks to move.
  
• A 2x3 brick is actually sufficient to move two stock wheelsets.
  

So it's not quite "almost nothing". But a significant difference.

[LoneBrickerSG]

If you're going for a look that's different from the standard Lego wheelblock (for example, North American freight cars) and are not a die-hard purist, I would suggest a design by Cale Leiphart that is available on the Railbricks site. http://railbricks.co...freight-trucks/ I use a simplified version on my freight cars. Here is a picture of it beside a regular Lego truck

Dan, do you have any of those trucks in use? do they also have a low rolling resistance like normal lego wheelsets, or does the brass axle housing create some?

This is what I originally wrote. I woke up this morning and realized this wasn't entirely accurate, as the P54s use popped-out wheelsets, and the T1 is a heavy locomotive with a lot of axles. I replaced the two stock wheelsets on the test car I had measured with technic axles and remeasured:

• Two axles with Big Ben Bricks small wheels took three 2x4 bricks to move.
  
• Two axles with the Power Functions train wheels (small) took four 2x4 bricks to move.
  

So in theory it would be beneficial to use BBB wheels in all the dummy trucks that need technic axles?

Did you guys happen to test any wheels on technic axle pins? The ones that snap into technic holes?

Edited February 13, 2015 by LoneBrickerSG

[jtlan]

So in theory it would be beneficial to use BBB wheels in all the dummy trucks that need technic axles?

The official Lego wheels probably have more rolling resistance because I tested them with the tire on. I suspect if you remove the traction tire they will have similar rolling resistance to BBB wheels.

Did you guys happen to test any wheels on technic axle pins? The ones that snap into technic holes?

We didn't test any Technic axle pins, but I expect the rolling resistance of those will be higher than a Technic axle. This is because the pin is not as well supported as an axle, and the resulting cantilever forces will increase the resistance.

Edited February 13, 2015 by jtlan

[Commander Wolf]

I'd assert that the rolling resistance of the P54s with the popped wheels is significantly higher than what it would be with full stock wheelsets. It's better than Technic for sure, but it's not miles better... the cars don't actually roll that easily.

• Two axles with Big Ben Bricks small wheels took three 2x4 bricks to move.
  
• Two axles with the Power Functions train wheels (small) took four 2x4 bricks to move.
  
• A 2x3 brick is actually sufficient to move two stock wheelsets.
  

I'd also note that the difference in rolling resistance between BBB smalls and PF smalls with tires might not seem too big here, but in practice the difference will be exaggerated (maybe by a lot) in corners. The same can probably be said for the resistance difference between PF smalls with and without tires.

[Dan-147]

Dan, do you have any of those trucks in use? do they also have a low rolling resistance like normal lego wheelsets, or does the brass axle housing create some?

The brass axle on brass tubing has very little friction compared to a plastic technic cross-axle on plastic technic hole. Just make sure that there is no flashing left on the cut of the tubing. I use a round needle file to grind it off. I've also tried the original 9-volt wheels with a steel axle that went through the wheels. This also worked well but these wheels are getting harder and more expensive to get. A bit of graphite powder lubricant reduces friction even more without making any mess.

The inner shoulder on the Lego wheel is flush with the side of the wheel. This does not create a problem when the wheels are used as originally intended as there is nothing that can rub on the inside of the wheels in the wheelblock. Initially, I cut the tubing at 32mm (exactly 4 studs) but the sides of the wheels would sometimes rub on the sides of the bogie frame (creating quite a bit of friction). That is why I now cut the tube at 33mm so there is a ½mm play on each side.

Dan-147

[legoman666]

I tested one of my own locomotives in the same way; I had it pull a weight up. I kept adding weight until it couldn't pull any more. The locomotive I tested has steel wheels, so the traction is kind of low for its weight. On metal track, it could pull 340g. On plastic track, 390g. The locomotive itself weighed 1152g.

If this locomotive had rubber bands on the wheels, I'm pretty sure it would tear itself apart because the wheels would never slip.

[baard]

Interesting topic, thanks for posting

My experience is that locos should only have technic axels where there is no option for anything else. For two bogey trains I replace the non-motor bogey new train wheels on technic axels with one that has the http://alpha.bricklink.com/pages/clone/catalogitem.page?P=2878c02#T=C part. Include some snotting and no one will notice the difference in apperance.

For slow moving shunters traction is the limiting factor, not speed. Good O rings and weight seem to play a bigger role than motor power for my purposes at least.