Car Physics - Friction?

[scottb613]

Hi Folks,

I've been running the B&O Mountain Sub again - hands down my favorite route. It's simply gorgeous east coast coal railroading and an incredible site to behold.

I was running one of the canned activities - Empties to Austen - and it just brought this to the forefront again in my mind.

I'm still pretty much a newb at TS2022 - but is there any - friction - in this sim? Davis Coefficients? The train is forty-five empties with three SD-35's out of Grafton - even so - I'm toggling between idle and notch 1 - so I don't exceed the speed limits as I'm climbing 0.5 to 0.8 grades. The train seems to roll forever without adding power. I could probably drop two of the SD-35's and still make the climb.

If there is friction - are there any examples of rolling stock that would behave more realistically that I could try?

Thanks.

Regards,
Scott

[GreatNortherner]

Hi Scott,

First of all, thanks for the kind words about the route.

In which scenario were you encountering this? I couldn't find a train with fourty empties and three SD35s in the scenarios or QDs but I could have just overlooked one. Or is that a consist you've made yourself? In any case, I've just started the Off the Patterson Cutoff (piggyback) scenario which you mentioned and, though it gives you a ton of power for the flat sections at the beginning, my train did loose speed slowly but steadily when idling the throttle. I didn't measure it, but maybe at a rate of something like 0.1 mph every two seconds or three at 15 mph and at a quicker rate at 30 mph on the level track past McKenzie.

One related thing to note, drag settings and maybe also friction are a bit all over the place in this sim (including on my own older models). So it's easily possible that two trains of the same consist make-up but assembled from different models/sources can behave quite differently.

Cheers,
Michael

[scottb613]

Hi Michael,

My apologies - I corrected my original post - it happened to be - Empties to Austen - and it was forty-five empty hoppers. The scenario and rolling stock were all original - as issued with the route. While I am by no means an expert - it just seemed like the rolling resistance for friction bearings seemed more akin to futuristic bearings using electromagnetic levitation rather than steel axles sitting in a blob of grease - hence the grades were hardly an issue.
:)

I had done some searching and reading on friction - here - so I didn't think it was an issue with the product - just an inherent nature of the sim. I was kind of wondering if someone had made a more advanced car physics model addressing friction - - - that I could test to see if it behaved more realistically. I don't know anything about how the car physics are calculated - here.

The other thought I had was trying to leave some handbrakes on to add some drag to the train - but that's less than ideal.

Thanks so much for the interest.

Regards,
Scott

[GreatNortherner]

Hi Scott,

Thanks! Apologies from me, too: I've found the train and scenario. I'll blame it on having so many scenarios installed for the route that I missed this one with empties out of Grafton.

I've just run that train to past Hardman in updated TS and it felt to me like it has always done (before the update). The track is more or less level up to past Thornton, where it starts going gently uphill. The train does loose speed pretty slowly on those grades, almost as if the grade wasn't there at all. I admit that I never noticed this before (but now I probably won't be able to "unsee" it ever again). Though I should also add that I don't know just how much a 0.2 - 0.5% grade *should* slow down an empties train in reality. I'm not an expert about train (sim and real) physics and I always rely heavily on input from others for my models.

I'm afraid this half-answer is the best I can come up with.

Cheers,
Michael

[scottb613]

Hi Michael,

No worries - appreciate your input - I just didn't want to spin my wheels if it's impossible.

I just looked at your site - WOW - nice work...

Have a good day.

Regards,
Scott

[kris120]

Please take into account that TS (or is it the Physix engine ?) does not calculate speed correctly in every case.

When you convert a wagon blueprint with serz.exe and look for the value left of </RollingFrictionCoefficient> you see often 0.0015 which is common usage and not bad, but on longer trains you feel that this is too much.
You go generally better with 0.001, but don't try to enter a loop which is not levelled and slide without any power (this is only possible in a freeroam). You will accelerate from lap to lap.

These are only my personal general remarks as I do not (yet) have this route.

[GreatNortherner]

Hi Kris,

Thanks much for your comments on this (to me, somewhat "mystical") topic.

I quickly checked a bunch of rail vehicle BPs on Mountain Sub and the value of 0.0015 seems to be most common (which also looks very familiar to me and makes me wonder if this isn't the old "Kuju standard" for this parameter). Crucially, the 55 t and 70 t hoppers on the route have their friction set to the 0.001 which you mentioned, so it makes sense I guess that the coal train would roll farther in idle.

[Edit] -- I've just done a comparison run with the same empties hopper train, once with the 0.001 and once with the 0.0015 rolling friction setting. SD35 + 70 empties in QD out of Grafton, idle off at 35 mph around GN Tower, wait for the slack to settle and then count seconds between 0.1 mph reductions. I could not detect a difference. There probably was one, but so minor that it was undetectable with "verbal" counting of seconds.

This is what the Techdocs say about this: https://sites.google.com/a/railsimdev.c ... n-Elements

What is stated there suggests that the development was all based on UK coaching stock, which might explain why long American freight trains are trickier to set up physics-wise.

Cheers,
Michael

[scottb613]

Hi Michael - Kris,

I ran the full activity again last night leaving 12 handbrakes set - for the duration. While not a real solution - the train did feel more realistic when letting off the throttle - there was actually some desire to slow down. In general - the train seemed to behave in a more plausible manner. I also ran TS Tools and I think I have my - very first - "blueprint" open for the physics. Obviously - I have much to figure out but I was looking for the friction values in question. This sim has been around for a very long time - so I'm sure I'm not the first to head down this path - and since there's no obvious fix already - perhaps it's not possible. The HIS team seems quite capable.

Thanks again - appreciate your interest, patience, and insights.


Regards,
Scott

[BNSFdude]

It's not necessarily the cars at fault here. Locomotives without scripted power control tend to be either far over or under powered depending on how the blueprint is set up.
You can typically tell which way it is leaned towards if the locomotives twitch around as you adjust your throttle position, which would be an indication of tevsthrottle being ridiculously high.

[scottb613]

Hi Anthony,

Good point. I think I'm probably going down the rabbit hole - further than my current knowledge level and skill set will allow. I do have your SD40-2 pack and that seems to behave very well - LOL - I'm certainly not up to scripting work. I wish your model had some paints available that would be appropriate on this beautiful route.

I spend most of my time on the other side of the fence - where basic friction modification is as easy as using FCALC to crunch some numbers and updating four variables:

Code: Select all

   ORTSBearingType ( Roller )
   ORTSDavis_A ( 4442.98lb )
   ORTSDavis_B ( 4.0380lbf/mph )
   ORTSDavis_C ( 0.024000lbf/mph^2 )

Thanks for the help.


Regards,
Scott