I bought a locomotive as an add-on DLC. It is one of many pushing and pulling a long US freight consist. We are decelerating by reducing throttle. We steadily descend from 35 mph towards 29. At 30 i see the reason: a tight curve ahead. My lead loco enters the curve at 29. During our passage, i emerge, but the remaining long train traverses that loop of track, still at 29. We remain 29. Like a Jungfrau of old. Now a slight force can cause a large deflection, but this is a heavy freight train caught in a curve. Only the track elevation and curvature deflect the curved path of what had been a train on a straight track. The bogies pivot for a good reason, but the axles are arranged about those pivot points, typically fore and aft. Imperfect, but relative to the track curvature and gauge, effectively aligned withe the curving track, but not centred on the track. The curved track has an outer edge where an inner track surface carries wheel rim and flange. However, in the cab, it appears that this massive mechanical deflection exacts no visible cost in momentum. In my mind i hear the wheels screeching as the heavy long freight train rounds the tight curve. Why is this at no cost to the very heavy long train as might show on the speedometer ? Is the wheel noise NOT a reflection of increased friction along the curve? By Newton alone, deflection requires a force. This train is not a photon along a geodesic jet bundle traversing a gentle gravity well in space-time. Hence my doubts. Now about those Toronto screaming trams ... witches or physics ? I choose Newtonian mechanics. And then there is physics in the 20th century: heat and black-body radiation. The turning wheels are deflecting, Kosten-frei. Entrée gratuite pour nos 電車。Entering curves, caught in a climbing curve, however you look at it, room for doubt. Try it in a rented Cessna 150, however gently, simply bank with ailerons and rudder tracking the ball. Observe the cost on the instruments. Airspeed or altitude or both. See ? Oh. A train is not a plane on a plain ? Is a virtual train World in chunks a greater expanse than a bogus Train Planet ? hmmm ... 1/10th train life ? ... where marketing trumps geometry and physics ... so it's back to drifting in Assetto Corsa GT's, where the tyres scream for real ... sorta.
The track will be canted in a corner to keep speed rub off to a minimum, otherwise you'd burn 4 times as much fuel by having to constantly accelerate through curves.
How heavy was the train?, you can quite easily work out the friction needed to be able to reduce the speed of the train (of that weight), I bet it's HUGE.
Usually the flange squeal noise you hear is actually not from the flanges touching the rail. It's simply one of the wheels slipping on the rail, since the two are connected by an axle. That being said, it's usually not an issue with regular banked curves. But TSW uses the sound effect way too much anyways. To answer your original question, the flange is not in contact with the rail as often as you'd think. Therefore there is not much additional friction. That being said, I don't know if TSW simulates any form of additional friction in curves even if it would be applicable.
A very long winded way of asking whether curves slow down the train in TSW as they would in the real world and the short answer is no. The longer answer is no they don’t as it isn’t simulated. Should it be simulated? Probably. Maybe one day. They did introduce air drag effect in TSW3 but I think that was quietly removed or at least wasn’t added to many trains or routes.
I quickly looked up the formula in a nice little book that i have and found that the resistance form passing a curve is calculated as 6500/(R-55) with R obviously being the radius of the curve. Resulting units are N/t (newtons per metric ton) so if you know anything about physics, 1 N/t will give you just 0,001 m/s of deceleration. Now if I for example had a 300 meter radius, that would be 6500/(300-55), which is 26,5 N/t or 0,0265 m/s of deceleration. For comparison, acceleration and deceleration from a grade is calculated as 10N/t for every promile so a 1% grade will be 100N/t or 0,1 m/s of acceleration/deceleration (which TSW does seem to simulate properly).
This is not correct. Air resistance has been in the game at least as long as I have had access to the editor (early 2022 or late 2021, can't exactly remember) and it is enabled by default, but needs tuning per vehicle of course. What was added for TSW3 is "side drag" that is a parameter used to simulate transverse force on the wagon by wind. You can actually set it too high or have wind strength too high (in scenarios) resulting in the train being blown off the tracks. But this was mostly a feature relevant to Kassel-Würzburg (if even there, imho). Curve drag is not implemented as a core physics feature, but can be implemented by developers on each vehicle relatively easily since curve radius is an "exposed" variable (used for audio for example) and iirc the rolling resistance coefficient is editable at runtime so you can simply add a curve term to it. I haven't done this myself yet, but this thread made me tempted EDIT: I lied. You can't change rolling resistance at runtime. Maybe something DTG can add in the future (or just add curve drag)
The reason is that the method of turning is different for each of them. Train turns by leaning on the track, and if everything was 100% rigit, then turning wouldn't loose any more speed than going straight. The loss for train comes mainly from track deformation, and suspension movement... and as Tomas9970 wrote, that deceleration is not that big of a deal (10seconds of turning equals around 1km/h loss). And is kinda detached from the trian, becasue it has a lot do with the track flexibility, not core train physics. Airplanes are different, they have no hard surface to lean on. They turn by producing lift, but lift causes drag.. airplanes are also lighter than trains, and move faster (so drag affects them more). When you move your ailerons, and bank, you move the lift vector of your wings, which is no longer pointing straight up.. that's why you start loosing alltitude. So then you compensate with elevators, to pitch-up, you increase your AOE to increase the lift produced, but that also increased drag, not only by the increased lift, but also becasue more air can now be, potentially, hitting the body of the plane. Then there is also side slip, which you need to correct with rudder, which causes additional drag... that's why you, in the end, need to compensate with more throttle... These aerodynamic interaction just doesn't exists on any hard surface vehicle... be it train, or a car.
Thanks for the corrections and explanations. The drag effect does seem to be most noticeable on Kassel-Würzburg and the ICE 1 but not so much elsewhere. The higher speeds are probably needed to actually experience its effect in game so that’s why I thought the was no air resistance elsewhere as I hasn’t noticed its effect and thought they were the same thing.
