New Coupler Syndrome Perhaps?

Think we may have two things at work on us with this switch node problem. Example:Sometime ago I downloaded an activity that runs from Whitefish to Shelby with 62 default double stacks. Made it over Cutbank bridge no problem (though I did ease across it at 10 MPH.)Currently creating an activity where a train also leaves Whitefish to Shelby. By the time it reaches Cutbank bridge, it consists of about 40 default cars. Moving across the bridge at 20 MPH: Broke coupler. This is with 5 coupler values and /s on the damping. The only significant difference is: I picked up loose consists along the way.I begin to smell a rat when after picking up a helper at Essex to add to the engine lash-up, there seemed to be a bounding box issue once the new loose consist unit was cut-in. That is, there was now about 4' of airspace between all units. Yet, they remained coupled under load going up Marias. I picked up two more loose consists of default hoppers/grain cars... and when I got to Cutbank bridge: Broke.I suspect that a loose consist somehow is accounted for differently within the sim, and thus the broken couplers at a previously crossable "dubious" switch node section. This seems to have evidence in that I have another downloaded activity that runs from Spotted Robe to Shelby, picking up hoppers and grain cars along the way... and I can't clear Cutbank bridge with that one either.My suspicion: Loose consists compound the switch node timing issue.Input? Observations? Andre Ming

Alert | IP

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

RE: New Coupler Syndrome Perhaps?, Bing, May-31-03, 08:41 PM, (1) RE: New Coupler Syndrome Perhaps?, Coonskin, May-31-03, 11:39 PM, (2) RE: New Coupler Syndrome Perhaps?, Bing, Jun-01-03, 01:21 AM, (3) RE: New Coupler Syndrome Perhaps?, CWA13000, Jun-01-03, 01:34 AM, (4) RE: New Coupler Syndrome Perhaps?, Coonskin, Jun-01-03, 02:44 AM, (5) RE: New Coupler Syndrome Perhaps?, CWA13000, Jun-01-03, 09:11 AM, (6) RE: New Coupler Syndrome Perhaps?, Bing, Jun-01-03, 10:56 AM, (7) RE: New Coupler Syndrome Perhaps?, CWA13000, Jun-01-03, 05:57 PM, (8) RE: New Coupler Syndrome Perhaps?, Coonskin, Jun-01-03, 11:05 PM, (9) RE: New Coupler Syndrome Perhaps?, CWA13000, Jun-02-03, 00:00 AM, (10) RE: New Coupler Syndrome Perhaps?, sniper297, Jun-02-03, 00:59 AM, (11) RE: New Coupler Syndrome Perhaps?, Coonskin, Jun-02-03, 10:29 AM, (12) RE: New Coupler Syndrome Perhaps?, Coonskin, Jun-02-03, 10:37 AM, (13)

Conferences | Topics | Previous Topic | Next Topic

Its not the Cutbank Curse, Andre. That occurs at the crossover, east edge of town. So it has to be something else. Have you increased Break Foce and Derail BufferForce values for all the cars and the loco(s) involved in the activity? The Curse is alwways westbound anyway. Eastbound far as I know the problem does not exist. Never had anything happen at that bridge, going either way, been through there dozens of times. Bing

Alert | IP

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

Right you are about it not being in the right location... this is switch node timing quirk, pure and simple. It's the same problem that
any route can suffer from if the switch nodes are too far apart. As for break force, they're all at 5. I've not touched the buffers, but perhaps it's time to up that too.Andre

Alert | IP

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

Just a suggestion here: Lately I have seen more and more cars set for 9.1 BreakForce; I know that seems high, but it comes down to the reality of trackwork in some routes not being the best. Not talking Cutbank which is a particular, rather technical problem. I had one foreign route where the cars literally jumped up and down at a certain point. Derailing, of course. The other is to set the DerailBufferForce to 1000kN. Reason for this is those rascals get to banging back and forth violently as they go through a bad section of track. Try to get stationary by the Cutbank Crossover and watch the fun. When you see what is going on, maybe you will want to change the DerailBufferForce. Solved a lot of problems for me. Then you can concentrate on coping with Switch Node Timing. Sounds like a nasal problem, doesn't it? Code in the node? (That really, truly is an awful pun.) Bing

Alert | IP

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

In reference to the "Cutbank Curse", I was looking at the railroad crossing East of town in the RE and I noticed that when I put the crossing into wire frame that the road underneath was partially exposed on the West side of the road and appearing to be under turf on the 2/3 of it toward the East side. It looked like the road was level but the tracks were starting to run down hill. I put the section of road into wireframe and called up the properties box and it said that it was Shape File Road2lstrt50m.s and collition = Bounding Box. Kip Crawford in his Track Fix doc. included in the Hoodoo Pass Rt. suggested editing the .sd files of some track. I started fooling around with the road file and thinking I could recover easier if I just edited the road file as he suggested for the track files. I deleted the minus sign from both the .s and .sd I backed up these first. It didn't seem to make a difference in the sim. I was going back to either eliminate the bounding box line as he suggested or make the negative value less because it didn't make sense to me to make it positive since I thought that positive was supposed to mean above the rails. I could not do that because now when I try to open any .s or .sd files it comes up like that shorthand computor code instead of abreivated english and I can't edit. Have any of you guys investigated this possibility???

Alert | IP

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

Thanks for some input fella's.CWA: I don't have any experience with the situation you're describing, so can't comment. However, I think the reason you see the hyroglphyics is because the .s file is compressed, it will need to be decompressed before Unicode can read it.Andre

Alert | IP

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

Thanks for the tip. Do you think it is possible that the edge of the pavement could be protruding up through the tracks. In Bounding Box Editor I tried to raise the box of my PO&N 1883 passenger coaches to +1.5 to see if I could get over what ever I seemed to be running into and that didn't work. Ill try to decompress and maybe jack-up the cars more and see.
Good Luck

Alert | IP

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

CWA: RouteRiter will decompress the .s Shape file so that you can read it in WordPad or equivalent - stick to wodPad or Unicode capable; if you accidentally hit Save on the way out and you are using, say, NotePad in the version which does not support Unicode you have just ruined the file. Research on the Cutbank Crossover Curse indicates it is not solely a fixed object shape problem, though that may have something to do with it. Rather, it seems to have to do with westbound train length, there is also a theory about a geographic "seam", since I haven't yet gotten into building routes from the ground up I can't detail that aspect. I will say this: Some of the best minds we have working on MSTS have tackled the problem. We have seen stand alone versions of Marias in the "MP" series which now I believe stand at four - original MP, MP2.0, Mp 3.0 and currently MP 3.1 - there may be others. The Cutbank Crossover Curse is still there. The shakers and movers believe it always will be there. I just increase CoouplerBreak and DerailBufferForce values and have fun watching the cars go thru the crossover. Looks like your Uncle Ed playing "Lady of Spain" on his accordian. I figure if the brightest and best can't cure the malady, why waste time on it? Bing

Alert | IP

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

"it seems to have to do with westbound train length," Is my train too short; K36 #383 PO&N 1883 baggage and 1 PO&N 1883 coach?
I just raised the B Box lower to 1.5 on engine and tender to match cars, all units have a 1,000 buffer, #6 in break (I guess I have to raise to 9.1) and I just shortened all tenser #s by .4. After doing the 1.5 on engine and shortening tensers I went back and instead of my coupler at rear of tender breaking at the west edge of RR xing, I derailed at that west side of the x over switch east of the RR xing. This is the first time I had a problem there and I was only going about 3mph getting ready to watch what happens at the other spot. "I just increase CoouplerBreak and DerailBufferForce values and have fun watching the cars go thru the crossover. Looks like your Uncle Ed playing "Lady of Spain" on his accordian. I figure if the brightest and best can't cure the malady, why waste time on it? "I whole hartedly agree,I wonder if the antique rolling stock can take the punishment! Just to check, in the break statement I am supposed to change the first number, is that correct? Sometimes instructors give advice assuming that us dummies actually know which numbers are the variables discussed.
Thank you very much!

Alert | IP

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

"I will say this: Some of the best minds we have working on MSTS have tackled the problem. We have seen stand alone versions of Marias in the "MP" series which now I believe stand at four - original MP, MP2.0, Mp 3.0 and currently MP 3.1 - there may be others. The Cutbank Crossover Curse is still there. The shakers and movers believe it always will be there."Not exactly. I've fixed it, but it requires trackage changes to cure the true culprit: The switch node timing bug. Unfortunately, "my" modified Marias was unique because of the changed .tdb, so none of my activities would work with everyone else's Marias, so I deleted the modified Marias and reinstalled stock MP3. You can once and for all fix the Cutbank Curse by adding a non prototyical "Team Track" at Ethridge on the westbound (north) main. (That is, simply adding a short spur so as to place a switch node.) This breaks up that long stretch of track where the cumulative timing errors pile up that's to be readjusted at the next switch node. Westbound on the north main, that next switch node happens to be the first switch at Cutbank, and thus the Cutbank Curse. As is, there's something like 18 miles of uninterupted track (i.e. no switch node) between the switch nodes at Cutbank and to the east on the north main. Much too long, and the train will "snap" at the next switch node as the MSTS timing is readjusted to be "on target" at the switch nodes. See it this way: MSTS times/moves the train according to the data of each track section in the .tdb. These are not "exact" figures, but must be rounded off to the nears "x" amount. Therefore, the calculating routine accumulates error as it takes its timing from the track pieces. When it gets to a switch node, it "adjusts" the train to the fixed point of the switch node. The jerking of the cars is the result of each car's location being adjusted to compensate for the cumulative "drift" of the running calculations neccessary between switch nodes. Bottom line: On a straight route, you can go up to about 10 miles between switch nodes. On a curvacious route, only about 5 miles. Any thing more: Broken couplers. The Cutbank switch node distance is so great that it requires astronomical break/derail values to traverse successfully. The above switch node at Ethridge reduces the distance between the switch nodes to about 8 miles, well within acceptibility range for a straight type route. Clear as mud?Andre Ming

Alert | IP

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

"You can once and for all fix the Cutbank Curse by adding a non prototyical "Team Track" at Ethridge on the westbound (north) main. (That is, simply adding a short spur so as to place a switch node.) This breaks up that long stretch of track where the cumulative timing errors pile up that's to be readjusted at the next switch node."Does it have to be a spur and if so does it have to be a particular length? Would X-overs suffice? I would guess that the spur would be easiest to place assuming one knew how to rip up some track and get it back together again. The fanciest thing I've done is place some coal and water towers and they don't look good cause I don't know how to lay track into the supply in side of the tipple. I still have not figured how to delete them when I decide that one is in a stupid place. Sometimes I abandon them and sometimes I drag them down the tracks till I find a better place. But it's all in fun, right, common work with me! It's all in fun, right!
Have a nice day! I'm going over to the Hoodoo for a while and if that doesn't make my day I'll go back to PO&N where everything works and I belong!

Alert | IP

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

Hee-hee, everything works cuz I doubt there's anyplace on the route where there's more than a half a mile run between switches!Andre, your conclusion about the loose consists may have something. Prior to the infamous late-coming half-measure 1.1 patch, there were a bunch of us trying to figure out the front coupler bug ourselves. Prior to the 1.1 patch the "holding brakes" on loose consists would not release using the front coupler, so it took notch 8 and a good tailwind to drag 6 cars out of a spur against the locked brakes. We never paid much attention to the switch window problem since you couldn't drag more than that anyway. (Now you know why the PO&N has so many wyes for reversing engines!) One experiment that was tried and tested by many was to make a consist with the engine reversed, and you could run at 100mph in reverse, hit the brakes and stop, push forward, back up again, run in reverse again, all day long or until you uncoupled. Once you uncoupled, coupling back on again with the front coupler, the brakes wouldn't release and anything over 10 cars you couldn't budge it even going downhill.So yes, loose consists are different. One other thing that might be a factor:http://www.train-sim.com/dcforum/DCForumID3/18341.htmlThe whole reason for that first PO&N activity patch (not the recent update, altho the patch is included in that) was to change the US2chemicar (buggy coupler box values or something) to the US2freightcar3, because I found that starting in Texas Tea, assembling all the tankers, going to Johnsburg and adding all those tankers, then running up the South Pass / Fiddletown-Combs Landing branch cutoff to North Pass / Newcastle /Armstrong / Stevens Point, drop the train outside Stevens Point and get all the tankcars in Stevens Point yard, suddenly I'm getting lockup/crash/send-don't send with the front coupler, and if I reverse the engine on the wye and use the back coupler the cars refuse to couple together unless they're in a certain order. Why?Unless all the cars are "facing" the same way, the ones with incorrect BBox values will only couple together in a certain order, and even then the Moon has to be in conjunction with Jupiter in the 3rd house or something. Might be worth a look, check that all the loose consists are facing in the direction of travel. If that's not it a call to the Psychic Friends Hotline might be in order.

Alert | IP

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

Sniper:I too, have gone back and reoriented any loose consists that were inadvertantly placed in the "wrong" direction.I was in the process of test running this lenghty/involved activity and got the the yard throat crossover at Shelby, only to get a red light that wouldn't budge. No loose consists, no AI traffic headed my way, but still a red light. Got to figure that one out now.Andre

Alert | IP

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

Easiest fix is a short spur. Remember this formula: One 50m straight = One 10d switch and one 10m straight. There's a single piece of 50m straight at Ethridge where I cut-in the new "Team Track" spur. The spur had a westbound connection to work well with the typical flow of traffic.BE ADVISED... that IF you MODIFY the track of ANY route, you have just cut yourself off from any activities that are created for it, nor can your activities be shared. They won't be compatible.To delete a section of track: Click the "Select Object" button (first button on left, top row). Place the cursor on the track piece you want to delete, it should turn red (highlighted). Click on the track piece while it's highlighted. It should turn into a wire frame. Hit the Delete button: It's gone.IF the track section is a dynamic track piece, it will NOT highlight. Look for a little green diamond sitting between the rails. Highlight and click on THAT. The green diamond will turn wire frame. To remove that dynamic track piece, now hit "Delete". Be reminded that you CANNOT simply plop in another track piece if you change your mind after deleting a dynamic track piece, you will have to RELAY the piece. (Or exit without saving to retain the previous dynamic track section.)Andre Ming

Alert | IP

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