Hump yard operation

[csantucci]

I don't kick the cars, I try to run the engine at constant speed (see the second part of the video with the view from above), even if this is not easy, switch turnouts and uncouple cars (if I would uncouple cars in advance, they would begin to brake and the whole game wouldn't work. Engine speed varies between about 0.5 to 2.5 mph. In the first part of the video I have to stop the engine every two groups of cars to allow for the camera following the cars. That of kicking is mybe only an impression, because OR nicely implements car slack forces.
True that cars arrive at a bit high speeds against the parked consists, but this can be adjusted by putting the parked consists farther away or by increasing the MaxApplicationRate parameter.
Thanks for appreciation!

Carlo

[sniper297]

Maybe "kicking" isn't the right word, what I meant was they have to be rolling when uncoupled, you can't just hang them over the hump and let go and watch them start rolling just by gravity, correct? What percent is the hump grade? Do the brakes on the modified cars work as train brakes when they are coupled on?

It's been several years since I tinkered with MSTS, but the one time I tried the modified "coasting physics" (which were available at the time, what you're talking about is probably different) the cars were okay (not good, but okay) for kicking or humping, worthless otherwise because in a train they had no air brakes. Got kinda hairy trying to stop a 50 car train with only the independent brake, so you kept the specialized "coasters" separate and didn't add them to a train.

Unmodified physics, Roger Von Ahren's 100 ton coal hoppers on a 5% grade;

That's what I'm talking about, when the train is sitting on the grade or going up the grade, if the hose is disconnected, air bled or cock shut off, on a 5% grade it SHOULD do the same thing as the train when the brakes are released, start rolling down by gravity alone picking up speed as it goes down. Default MSTS physics never did take that into account, ORTS 360 won't do it either with cars that haven't been modified for it.

[csantucci]

sniper, I haven't tested if the cars start rolling also from 0 speed when they are decoupled on the grade.
I don't know the grade. I tried to look at with the RE, but it gives me only 1 FPS on my computer, and so it was impossible to reach the point where the hump is to check about the grade.
The cars brake a bit (but only a bit), because the MaxApplicationRate parameter is greater than zero. The cars parked in the yard have standard braking parameters. This way, even if you add at the back of them some cars with reduced applicationrate parameter, the resulting consist can be braked quite comfortably.

Carlo

[OTTODAD]

The cars parked in the yard have standard braking parameters. This way, even if you add at the back of them some cars with reduced applicationrate parameter, the resulting consist can be braked quite comfortably.

If they had the same brakes settings as the humped wagons. i.e. brakes released and then getting hit by them could they then roll into the buffers ?

The loose consists not having brakes applied could they roll in yards where tracks are slightly inclined / declined, I have noticed in some yards where after releasing the player train's brakes it starts to move.

O t t o

[sniper297]

"The cars parked in the yard have standard braking parameters. This way, even if you add at the back of them some cars with reduced application rate parameter, the resulting consist can be braked quite comfortably"

Comfortably but not REALISTICALLY, I think what we're looking for here is something that don't require kludges, workarounds, or compromises.

Dunno how many of you remember the prepatch MSTS back in 2001, but before the first patch when you coupled onto a cut of cars (for you Brits a "rake of wagons") with the front coupler of the engine, it took full power to drag half a dozen cars out of a facing point spur. That was due to the "holding brakes" (or whatever they call that subroutine) not releasing, the game was designed so that the back coupler of the engine engaging the front coupler of a car would release those "holding brakes", they forgot to program that into the front coupler. Prior to the patch people tried that type of workaround, hack the physics in some cars to reduce or eliminate the air brakes, then you needed special cars for switching with the front coupler different from cars used in a 70 car freight over Marias Pass. In other words, unsat. Ideally we're looking to keep the air brakes and hand brakes while getting rid of the unrealistic holding brakes.

[conrail1]

It sounds like this is becoming more difficult than it has to be.

In the real world a road train arrives at a terminal and if it has to be classified it is put in the receiving yard.

The road crew cuts off the road power and the train goes into emergency with the road power heading for the house (fuel plant).

Sometime after that the car department bleeds off the train and preps it for classification over the hump. Most receiving and classification yards I’ve worked in are in a bowl so the cars do not run out. When the cut of cars is prepped for classification two or three cars might be left with air on them for brakes (some operations/rules of course require the use of a sufficient number of handbrakes). When the hump crew ties on they just bleed off the 2-3 cars before shoving it over the hump.

The hump unit then ties on to the train and shoves it over the hump.

As each car gets cut off and rolls into the classification yard (again in a bowl) the retarders and the bowl itself slows the car eventually to a rest or couples up to cars already in the track.


A simple “prep for classification” feature in the sim that will show the cars “bled off/no air/brakes” (whether it is for flat switching or a hump operation) could be implemented.

The rest of it is up to the physics model, and if it accurate then the cars will gain and lose momentum in a realistic manor. Of course there are a lot of other variables such as resistance caused by the track (crossovers and turnouts), environment (wind/temp) and the equipment itself (roller/friction bearing, car length as an example).

[sniper297]

In a perfect world...

Right, but it's not really about the physics, that's adjustable in each freight car - the problem is brakes on/brakes off.

BrakeEquipmentType ( "Handbrake, Triple_valve, Auxilary_reservoir, Emergency_brake_reservoir" )
BrakeSystemType ( "Air_single_pipe" )
MaxBrakeForce( 50kN )

MaxHandbrakeForce( 35kN )
NumberOfHandbrakeLeverSteps( 100 )

TripleValveRatio( 2.5 )
MaxReleaseRate( 15 )
MaxApplicationRate( 25 )
MaxAuxilaryChargingRate( 5 )
EmergencyResCapacity( 7 )
EmergencyResChargingRate( 5 )
EmergencyBrakeResMaxPressure( 90 )
BrakeCylinderPressureForMaxBrakeBrakeForce( 50 )

That's all the data for a typical freight car brake system, with those numbers it works okay for normal operations IN A TRAIN. As in coupled to an engine, or coupled to other cars coupled to an engine, the train brake in the engine controls brake application in the car. Problem is for whatever reason MSTS was originally programmed so that car not coupled to engine = brakes on full, from what I see here this version of ORTS does the same thing - uncouple car, brakes on car go into emergency. In other words what happens if the cock on the train side is shut off, the cock on the last car is left open, uncouple last car and the air brake hose separates dumping the air in the pipe on the car that was uncoupled. Program in something additional that does NOT automatically apply the brakes on a car when it's uncoupled, any car should roll realistically after that.

[conrail1]

Problem is for whatever reason MSTS was originally programmed so that car not coupled to engine

I would hope that using anything from MSTS would just be temporary for OpenRails until the physics portion of the program is worked on and up and running.

Program in something additional that does NOT automatically apply the brakes on a car when it's uncoupled, any car should roll realistically after that.

A feature just like I mentioned above that shows the consist/cut of cars “bled off” would solve this problem. If the programs physic engine recognizes the car as not being on air and bled off the brakes obviously would not apply whether it’s coupled to other cars or not.

[csantucci]

If someone likes to try the hump yard activity that I created, he can download it here:

http://www.4shared.com/file/etNvOize/Ablaufberg_Act.html

A small readme file is inside. You will find it difficult to be fast and precise enough to switch every car to the right track!
If there are problems, pls. report them here.

Carlo

[conrail1]

If someone likes to try the hump yard activity that I created, he can download it here:

http://www.4shared.com/file/etNvOize/Ablaufberg_Act.html

A small readme file is inside. You will find it difficult to be fast and precise enough to switch every car to the right track!
If there are problems, pls. report them here.

Carlo

Thanks for taking the time to come up with the activity.

At this point without hump retarders and power switches in the class yard being implemented yet, I wonder if it would be better to just experiment with trying to flat switch in OpenRails instead

[OTTODAD]

If someone likes to try the hump yard activity that I created, he can download it here:
http://www.4shared.com/file/etNvOize...fberg_Act.html

That link gave me the ZIP for the activity alright, the link in your e-mail did not.

Where can I get the route from ? It was released in 2008 and posted about at UK-TS and says something about using an old Tsection.dat ?

O t t o

P.S.: Found it and it's update: http://www.msts-streckenbau.de/index...cid=41&start=0
being over 600 Mb will take over an hour to download and hope that it will be worth it !

[GaryG]

Right, but it's not really about the physics, that's adjustable in each freight car - the problem is brakes on/brakes off.

Hi Sniper and other readers

With the "percentage" braking, you DO still have normal in-train braking forces as well as the free-rolling uncoupled capability. I just looked at the files provided in recent versions of RouteRiter. The "percentage" I mentioned refers to the values provided by Rich Evarts and are defined as 1% of the real values but end up actually providing the expected real brake forces when braking in a train.

One item that shows "percentage" braking is being used is to look at the BrakeCylinderPressureForMaxBrakeBrakeForce value. Percentage braking uses 0.50 or so while the original value would have been 50 or higher.

GaryG

[OTTODAD]

Now here is something amazing !

Have downloaded this route I am calling HUMPING, installed it into a MSTS MINI because it uses custom tracks and a custom Tsection.dat and then did my usual customizing to my liking.

Did not like the trackbeds and changed them all.

Running it in MSTS their textures looked awful from mid-distance on before and after my mod and could not re-mip them. But running it in ORTS the change was incredible !

Now tomorrow we are going to do some humping !

O t t o

[conrail1]

If someone likes to try the hump yard activity that I created, he can download it here:

http://www.4shared.com/file/etNvOize/Ablaufberg_Act.html

Carlo

Well I decided to give it a try, entertaining to say the least.

It was just as much fun flat switching (kicking) the cars as it was humping them. I'd like to give it a try with the Feather River Route in Oroville yard.

The Rubelandbahn route itself is also excellent and performs very well in OpenRails. Without a doubt worth downloading.

Thanks again for sharing this with us Carlo.

[longiron]

This information should be of help to the community as it explains the braking behavior in Open Rails when uncoupling cars.

Nothing special is done to set the brakes correctly after uncoupling. There is code that ends up setting the new loose consists brake pipe pressure to 64 PSI. Physics calculations are not performed for loose consists unless they are already moving, so the brake cylinder pressures won't change to match the pipe pressure unless you give the consists a push, or they are moving.

You can push a loose consist if you just uncoupled from it. You need to get more than .5 meters away to re-couple to the same consist.

If you dump the air, | on my keyboard, before uncoupling, the loose consist will have no brakes.

We are investigating how the physics might be modified to provide better hump yard and car kicking, including the ability to actively modify the braking of a loose consist.

chrisvw