Factorio Forums

I found something odd with train speeds. I started with a bi-directional train:
One locomotive is going forward and one backward. So one powers the train and the other is dead weight that needs to be dragged along.

Next I wanted to make a transfer station to load good from several small trains into a larger train and came up with this:
The spacing is such that one can place train stops on a single track so short trains can stop in the three locations shown. They are then perfectly matched to the longer train shown below. The long train is simply 2 short trains with 3 cargo wagons added in the middle. It has 2 locomotives going forward and 2 going backward and should be heavier than 2 short trains. And then I discovered something odd:

The LCCCLCCCLCCCL is faster than an LCCCL train.

WHY?

Air resistance.

A large fraction of the first locomotive's power is dedicated to fighting air resistance. The second locomotive adds a huge boost to speed because it can focus all its power on pulling wagons.

DaveMcW wrote: Mon Jun 17, 2019 12:53 pm Air resistance.

A large fraction of the first locomotive's power is dedicated to fighting air resistance. The second locomotive adds a huge boost to speed because it can focus all its power on pulling wagons.

Is that coded in the game ? What about the increase of friction due to the extra contact surfaces for extra wagons wheels ? irrelevant ? ( from game or real life perspective )?

It must have diminishing returns then right ?

Did you encounter this when making tunnels ? x)

this answer opens more questions

mmmPI wrote: Mon Jun 17, 2019 11:00 pm What about the increase of friction due to the extra contact surfaces for extra wagons wheels ? irrelevant ? ( from game or real life perspective )?

It must have diminishing returns then right ?

The rolling resitance of trains is infinitesimal when compared to air drag, even at low speeds. I'm not an expert, but if I'm right, it's orders of magnitude lower at low speeds, and at high speeds, the difference further increases, due to the fact that air drag is proportional to the squared velocity, while rolling resistance is independant from speed.

So theoretically, it's always better to add moar locos&wagons than to create a second train, from an efficiency standpoint, and the faster the train, the higher will be the energy economy. However, this might not be practical, so there are always limits to the exercice.

mmmPI wrote: Mon Jun 17, 2019 11:00 pm Is that coded in the game ? What about the increase of friction due to the extra contact surfaces for extra wagons wheels ? irrelevant ? ( from game or real life perspective )?

It is coded in game. Train mechanics is very unrealistic and made for gaming purpose and not to be realistic (or even consistent) simulation.

In real life rolling resistance is very important and it depends on axle loads and rail geometry. Also air resistance of wagon sides are more important than the front surface of locomotive (typically there are tens of freight wagons and they are typically not designed to be aerodynamic).

Koub wrote: Tue Jun 18, 2019 5:40 am while rolling resistance is independant from speed.

I was surprised to discover that it depends on the material used for the moving part and the floor.

In my mind I pictured a curved rail and imagine that there would be more "friction", due to centrifugal power, if the train is faster or heavier.

Before today "friction" and "rolling resistance" was the same for me.

With my wording i described "rolling resistance" which is a surface coeficient that increases to describe the higher deformation of the floor when something roll on it, creating a resistance to the movement (or something like that), i was thinking something else.

Hannu wrote: Tue Jun 18, 2019 6:32 am Train mechanics is very unrealistic and made for gaming purpose and not to be realistic (or even consistent) simulation.

I am pretty sure an empty train or a full train goes same speed and have same acceleration, so yeah i was not expecting precise simulation, i like having many many small trains that looks like angry ants, too bad it's even more ineffecient than i knew because of air drag, i could need less fuel ! ( and less pathfinding CPU usage too but that i knew )

Hannu wrote: Tue Jun 18, 2019 6:32 am In real life rolling resistance is very important and it depends on axle loads and rail geometry.

I read that axle loads is used to establish safety norms and taxes in between other things since the heavier the train or truck the faster the rail or road under it deteriorate. thanks for bringing that up.

I would also picture the heavy train having to slow more than the light train to turn at the same radius,thinking an empty wagon must have his center of mass closer to the rail than a full wagon( unless you load under the wheel ? ), but then what about the size of the ore, small enough it would act like sand , which loaded in bulk act like water if thin enough or something like that.

I'd gladly put that on the back of a perfect magical rail geometry and 'compartmentation' and not deal with any of that unless i want to focus my attention on this aspect of the game and i would try to enhance it with mods.

Hannu wrote: Tue Jun 18, 2019 6:32 am they are typically not designed to be aerodynamic

I see, literaly, and i can hear too, both through my window, they are not aeroacoustic if that is the correct word. They could , maybe.

Thanks for your answers

mmmPI wrote: Tue Jun 18, 2019 9:33 am

Koub wrote: Tue Jun 18, 2019 5:40 am while rolling resistance is independant from speed.

I was surprised to discover that it depends on the material used for the moving part and the floor.

In my mind I pictured a curved rail and imagine that there would be more "friction", due to centrifugal power, if the train is faster or heavier.

Before today "friction" and "rolling resistance" was the same for me.

With my wording i described "rolling resistance" which is a surface coeficient that increases to describe the higher deformation of the floor when something roll on it, creating a resistance to the movement (or something like that), i was thinking something else.

Hannu wrote: Tue Jun 18, 2019 6:32 am Train mechanics is very unrealistic and made for gaming purpose and not to be realistic (or even consistent) simulation.

I am pretty sure an empty train or a full train goes same speed and have same acceleration, so yeah i was not expecting precise simulation, i like having many many small trains that looks like angry ants, too bad it's even more ineffecient than i knew because of air drag, i could need less fuel ! ( and less pathfinding CPU usage too but that i knew )

Hannu wrote: Tue Jun 18, 2019 6:32 am In real life rolling resistance is very important and it depends on axle loads and rail geometry.

I read that axle loads is used to establish safety norms and taxes in between other things since the heavier the train or truck the faster the rail or road under it deteriorate. thanks for bringing that up.

I would also picture the heavy train having to slow more than the light train to turn at the same radius,thinking an empty wagon must have his center of mass closer to the rail than a full wagon( unless you load under the wheel ? ), but then what about the size of the ore, small enough it would act like sand , which loaded in bulk act like water if thin enough or something like that.

I'd gladly put that on the back of a perfect magical rail geometry and 'compartmentation' and not deal with any of that unless i want to focus my attention on this aspect of the game and i would try to enhance it with mods.

Hannu wrote: Tue Jun 18, 2019 6:32 am they are typically not designed to be aerodynamic

I see, literaly, and i can hear too, both through my window, they are not aeroacoustic if that is the correct word. They could , maybe.

Thanks for your answers

Compare a train to a bat. When the boat moves the water at the front is higher than the water at the end. The boat pushes a wave of water in front of it and has to constantly climb upwards. The same happens with the wheels of a train. They deform the train tracks and have to push that deformation along. The heavier the train the larger that deformation is.

As for the center of gravity: An empty wagon will have the center of gravity somewhere below the middle of the car. The wheels are usually heavier than the roof. But it will still be near the middle I guess. But now put in some steel bars. They are incredibly heavy so you only put in a few. Even though the wagon is 3m in height all the steel bars will lie on the floor. Way below the center of gravity when empty. So the center is low. On the other hand fill it with sheep wool. That's rather light so you fill the whole car. The center of gravity of the sheep wool will be above the center of gravity of the empty wagon and pull it up. It really depends what and how you load the wagon.

Proper train tracks are also built so the train tilts inwards in curves, offsetting the centrifugal force of the curve. There will be a set speed for the curve where the combined force will act straight towards the rails. Go slower and you "risk" falling inwards. Go faster and you "risk" falling outwards. Train tracks can't be tilted so much that a standing wagon would tilt over inwards so there is a limit to this.

mrvn wrote: Sat Jun 22, 2019 1:48 pm
Compare a train to a bat. When the boat moves the water at the front is higher than the water at the end. The boat pushes a wave of water in front of it and has to constantly climb upwards. The same happens with the wheels of a train. They deform the train tracks and have to push that deformation along. The heavier the train the larger that deformation is.

That's how I'll explain it if i'm asked that'll be easier.

mrvn wrote: Sat Jun 22, 2019 1:48 pm As for the center of gravity: An empty wagon will have the center of gravity somewhere below the middle of the car. The wheels are usually heavier than the roof. But it will still be near the middle I guess. But now put in some steel bars. They are incredibly heavy so you only put in a few. Even though the wagon is 3m in height all the steel bars will lie on the floor. Way below the center of gravity when empty. So the center is low. On the other hand fill it with sheep wool. That's rather light so you fill the whole car. The center of gravity of the sheep wool will be above the center of gravity of the empty wagon and pull it up. It really depends what and how you load the wagon.

I am not engineer but i imagine they would want the center of gravity of an empty wagon as low as possible to increase the loading capacity and potential speed in curves and would be ready to increase the height of this center of gravity quite high even at the cost of reducing speed when they are loaded. I have no idea how it is done in different countries for who is to decide what is acceptable speed for a train, who owns them and schedule priorities reading stuff about that now.

i had the visual of those wagons in mind https://www.youtube.com/watch?v=uTM1hY_mayQ because the images stuck to my head, it is not hard to guess what happens if you go too fast on a curve with those but most of the time i see some like those one the right side the weird shape that required the picture.
Circuits and consumer goods are like sheep whool contrary to ore stone or coal in my mind.

mrvn wrote: Sat Jun 22, 2019 1:48 pm Proper train tracks are also built so the train tilts inwards in curves, offsetting the centrifugal force of the curve. There will be a set speed for the curve where the combined force will act straight towards the rails. Go slower and you "risk" falling inwards. Go faster and you "risk" falling outwards. Train tracks can't be tilted so much that a standing wagon would tilt over inwards so there is a limit to this.

In case of a wild boar crossing the curves that happens quite often in some region those like biters aren't really scared by the noise. Any emergency stop would be desastrous really if it was too tilted on the inside. That seems logical now.

This also sets a limit on how much you can load on a wagon before the center of gravity will be too high for it not to fall inward on the most tilted curved of its path at speed 0 i guess , also reducing the risk of falling outward when going "regular" speed but with more load.

So you wouldn't put as much steel in volume than you would with whool, yet those coals wagons seems pretty full to me. endless thinking begins, paradox detected ! Maybie those only go straight lane, with limited curved and are running of some dedicated segment with particular rails.


I have been and am still playing rollercoaster tycoon 1, or the openrct version, they teach a lot about trains with the addition of momentum based on height it is a "physical pseudo realistic simulation" of the train too. I like that factorio is fun but also makes me want to know more about realistic things that are not usually the typical thing considered as fun.

The "load" being the "guest", the bigger trains and more filled train are faster in RCT, and can make the ride too rough for the guest in a parameter that is called "Lateral G's" (vs vertical gravity ) meant to represent that if you go too fast on a sharp curve it's dangerous for the neck and unpleasant , you can deal with the problem by having smaller train sometimes instead of rebuilding the part in question.

Now this logic is only partly applicable to factorio that is a almost a flat world, so no momentum based on height and weight of the train; the point i want to develop is the "lateral Gravity" it would be the increased friction that a heavier train would apply on the lateral part of the wheel. The one that keeps it in between the 2 rails. I have seen some pictures labelling the larger and lower diameter of the wheel , some with vertical part of the wheel that is doing this same job.


A heavier trains must lose more energy there right ? ( in real life not in factorio is my guess )

That could be represented when 2 trains are running out of fuel at the same time, the heavier train wouldn't slow down as fast as than the lighter one, (because of air resistance and momentum) but would lose more of its initial speed in % after the first curve ( because of more lateral friction ).

Also i'm learning "technical" english in a way that makes more sense to me here than learning list of vocabulary as like correct me without care in case of wrong use of word, i do appreciate .

mmmPI wrote: Mon Jun 24, 2019 2:17 am Now this logic is only partly applicable to factorio that is a almost a flat world, so no momentum based on height and weight of the train; the point i want to develop is the "lateral Gravity" it would be the increased friction that a heavier train would apply on the lateral part of the wheel. The one that keeps it in between the 2 rails. I have seen some pictures labelling the larger and lower diameter of the wheel , some with vertical part of the wheel that is doing this same job.


A heavier trains must lose more energy there right ? ( in real life not in factorio is my guess )

That could be represented when 2 trains are running out of fuel at the same time, the heavier train wouldn't slow down as fast as than the lighter one, (because of air resistance and momentum) but would lose more of its initial speed in % after the first curve ( because of more lateral friction ).

Also i'm learning "technical" english in a way that makes more sense to me here than learning list of vocabulary as like correct me without care in case of wrong use of word, i do appreciate .

While train wheels have a vertical part that is not what keeps the train between the rails normally. The horizontal part of the wheels aren't really horizontal. They are cone shaped. So the circumference of the wheel is larger on the outside than the inside. This makes the train turn towards the middle of the track. Every time it drifts to a side the part of the wheel connecting to the rail get larger on one side and smaller on the other and it turns back. The normal operation of a train is actually a swaying motion instead of a straight line. This also limits the curves a train can take before the vertical parts would touch the rails.