Broken Couplers Yet Again :(

I know this topic has been discussed as much as anything. But I just can't seem to overcome it. I changed every break value to 5.1e7N. But alas my Southwest Chief consist still manages to break apart.What could be wrong? My consist is 4 P42's, a baggage car, (Superliners)transition, 2 sleepers, diner, lounge, 3 coaches, 7 express boxcars, an adapter, 6 roadrailers, and another adapter.My suspicion lies with the roadrailers. But again, where it says break in the wag files, I put 5.1e7N. I should add the route is Cajon.Any tips? Suggestions? Is there another section in the wags and engs that needs to be fixed in addition to the break values?Thanks

http://pages.sbcglobal.net/fdoti/index.html

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

RE: Broken Couplers Yet Again :(, jtr1962, Oct-07-03, 03:42 AM, (1) RE: Broken Couplers Yet Again :(, Swissie, Oct-07-03, 04:12 AM, (2) RE: Broken Couplers Yet Again :(, SouthwestChief, Oct-07-03, 11:52 AM, (3) RE: Broken Couplers Yet Again :(, Grimace780, Oct-07-03, 03:55 PM, (4) RE: Broken Couplers Yet Again :(, DTA, Oct-07-03, 05:28 PM, (5) RE: Broken Couplers Yet Again :(, kevarc, Oct-07-03, 05:37 PM, (6) RE: Broken Couplers Yet Again :(, Swissie, Oct-08-03, 05:45 AM, (7)

Conferences | Topics | Previous Topic | Next Topic

I've played with other people's "fixes", tried my own, studied the track data base, etc. looking for a fix to this problem. Bottom line, there is none other than using CouplingHasRigidConnection() for all your equipment(which is what I ultimately did). While I no longer have any coupler slack, I can set my break values to a more realistic 1.7e6N or even less. The truth is that I'm completely disgusted by this problem and Kuju's failure to issue a patch for it. I know my "solution" isn't really that, but at least now I can run freight trains without worrying that they'll split in two whenever I come to a node(i.e. switch). And before purchasing MSTS2 myself I plan to see read these forums to see if this problem has been fixed. If not, I don't plan to purchase it.

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

Hi SouthwestChieflet me ask for the record: Did you also change all of the coupler damping values to the correct unit of measurement? One of the "inherited" problems from MSTS' default files is that basically all the US freight cars and diesels, i.e. all US2 rolling stock files, have a physically wrong unit of measurement for the damping of coupler slack in their respective .eng and .wag files. Here's a sample to compare: First the correctly damped metrocarriage.wag (Metroliner coach from the default NEC), and then the faulty US2freight6.wag. Both are taken from a virgin copy of MSTS:Correct (metrocarriage.wag): Coupling (
Type ( Automatic )
Spring (
Stiffness ( 5e6N/m 0 )
Damping ( 1e6N/m/s 0 )
Break ( 2e7N 0 )
r0 ( 15cm 1e9 )
)
Velocity ( 0.1m/s )
)Wrong (us2freight6.wag): Coupling (
Type ( Automatic )
Spring (
Stiffness ( 1e6N/m 5e6N/m )
Damping ( 1e6N/m 1e6N/m )
Break ( 1.1e7N 1.1e7N )
r0 ( 20cm 30cm )
)
Velocity ( 0.1m/s )
)Note: Where the unit of measurement for damping should read N/m/s, it reads only N/m in all of the US2 equipment. Which makes the damping useless - after all, common sense dictates that this function must be variable over time, isn't it? Unfortunately, many repaints were released with this error in it before the MSTS wizzards detected the reason for part of the broken coupler problem.Apart from editing this incorrect physical setting in all your .eng and .wag files, there is no immediate cure. I recommend the following points be observed:- Get EngMod to do all of the changes (saves a terrible heap of work), allowing you to do all changes in one go, fleetwide.- Change all .eng and .wag files fleetwide to the same or very similar values for coupler strength and damping. I have noticed that if I had one car in a long consist which had totally different settings (weaker break and/or damping figures), occasional breaks still occurred despite of the corrected damping physics, interestingly always before or aft of said car.- When doing the changes, get & implement the values from Bob Boudoin's latest physcs update - I think it stands at physics 4.6 for cars currently - available in the file library. Since I have implemented all these points with my trainset folder half a year ago, coupler breaks are as good as gone from my MSTS.For the record: The coupler breaks are not a problem of the consist you use, but depend on the distance and the speed the train travels between track nodes. For a simplified explanation, consider track nodes as synonymous term for switches. Now when your train (or any train, including AI trains, for that matter) travels down the line, it's position is always calculated by MSTS as distance travelled from the last track node (switch). Naturally, the calculation process is prone to some rounding of figures a couple of digits behind the comma, and the further away the train travels from the last node until it hits the next one, the more noticeable these small rounding errors become. As a rule of thumb: After 8 miles without a switch, problems start to occur when the train reaches the next node a little earlier than calculated and expected by MSTS. The train then "jerks" to its new position, car by car, and this is when your couplers are put to undue stress. By including the higher coupler break value and the corrected coupler damping units of measurement, your train should make it to most of the problem zones, although you will see cars jerk and juggle when going over the switch. I hope you can live with this solution. The rounding errors and thus the jerking cannot be overcome in another way than by spacing nodes closer together, which would include putting non-prototypical spurs and switches into a route. Lukas a.k.a Swissie

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

Thanks for the detailed instructions I have not corrected any of the damping values. This should probably help. And I know that several of my express cars are US2 variations. I think a while back that I did read the earlier posts on coupler breaking and I think I changed the damping values. I didn't have many coupler breaks in the past. But, as usual, after a reinstall everything gets "fixed" again. It is definitely happening on the swithches. The whole train jerks going over it and the break occurs somewhere near the back of my train.I'll give the damping correction a try, and I'll post my results.Thanks again As a side question, how come the AI trains never break their couplers? I have 15 or so separate AI trains in one activity and none ever break.

http://pages.sbcglobal.net/fdoti/index.html

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

>As a side question, how come the AI trains never break their
>couplers? I have 15 or so separate AI trains in one
>activity and none ever break. That's an easy one. The simulation cheats and doesn't do nearlyh as many calculations when it comes to the physics of the AI trains. As long as they appear to be going the correct speed and appear in the proper places, that is all that matters. Doing all of the correct calculations for AI trains would simply reduce frame rates and be a waste of processing power.Cheers,
Gary C.
Grande Prairie "The last civilized stop before Alaska," AB
Canada

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

Depends what you want in the sim. If you don't want to be technically correct, just set all break values to 9, and you will never see another train disconnection. As Kuju got it wrong in the first place, I see no reason to compensate for the awful points.David

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

Once you do the N/m/s correction, you can lower the break value to 3.1 or lower and not have the problems.Kevin Arceneaux
DEFAULT IS FOR DUMMIES
Cry havoc and Let slip the dogs of war.
Lone Wolf Activities Forum and Activity Packs are on www.3DTrains.com

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

Hi SouthwestChief,you're welcome - it's an old issue for me too, that has me bugged for a while, so I know 101% how you feel when that "activity ended" screen pops up while you rumble along peacefully...> As a side question, how come the AI trains never break their couplers? I have 15 or so separate AI trains in one activity and none ever break.As long as an AI train is out of your field of view (which is determined by your distance to the AI train and the value you have set in the sim's properties for the range in which objects are to be displayed), MSTS does not really run the Ai train over the tracks. Not knowing the hard code of the game, we must assume that it just uses the calculation process "distance from last switch travelled" to determine the AI train's progress, and goes onto the next "leg" (i.e. the next node) when the calculation tells the program that the AI train has arrived at the node. Thus the point in that assumption is that MSTS is never faced with the phenomenon of a distant AI train being a few feet closer to the next node than calculated. Note the word "distant AI train".Things become a little different once the AI train comes into being, i.e. when it is "created" at the limits of your viewing range. The location where it pops into existence is of course calculated from the distance travelled from the last node, including the rounding errors. And if coincidence wants this to happen close to the next node on one of the troubled sections of the route, the AI will face the same fate as your player train as soon as it hits the next node prematurely. Remember that specific default activity for the default NEC which always ended about 2 minutes into the game with an "activity ended - train derailed" message? (Is it "Morning Commute"? Don't recall what its English name was as my MSTS default stuff is in German, but Kevin mentioned it recently, too.)Well, after long hours of reasearch, someone of the wizzards (probably Sniper?) found out that this was casued by a nearby AI train hitting one of the nodes with the distance problem...Lukas a.k.a Swissie

Alert | IP

[ Home | MSTS Forum | MSTS FAQ | File Library ] Copyright © 2002 by Nels Anderson. All Rights Reserved.