Making Dual Gauge Switches

Hello all. I am working on narrow and dual gauge track pieces to complement the standard and Xtracks pieces, and to make life easier without dynamic tracks. They may be headed for the unified tsection.dat at some point.My question is about building dual gauge switches. I've defined the track sections and shapes to include four paths per switch (straight and diverging, standard and narrow). They work great when on the narrow gauge, but I cannot even set up an activity path across the standard side without a "failed to load path" error. The activity editor shows two black dots for the two points as expected, but only the narrow gauge one is ever used (whether the standard or narrow paths were used). I suspect MSTS is being confused by the two sets of points and is only using the first (narrow) one, hence all standard gauge paths break.Does anyone have any hints or better yet knowledge on how to get this done in a single shape (plan B is to just make the route builder lay separate narrow and standard switches overtop of each other, but it's not as elegant, and can have some rendering issues).For completeness, below is the left 10deg, left-common dual gauge switch from my tsection.dat: TrackSection ( 9034
SectionSize ( 0.9 1.51059 )
)
TrackSection ( 9035
SectionSize ( 1.5 40 )
)
TrackSection ( 9036
SectionSize ( 1.5 0 )
SectionCurve ( 163.7769 -10 )
). . . TrackShape ( 9013
FileName ( D1tPnt10dLftMnlLC.s )
NumPaths ( 4 )
MainRoute ( 1 )
SectionIdx ( 2 0.287 0 0 0 9034 9036 ) #narrow diverging
SectionIdx ( 1 0.287 0 0 0 9035 ) #narrow straight
SectionIdx ( 2 0 0 0 0 183 175 ) #standard diverging
SectionIdx ( 1 0 0 0 0 179 ) #standard straight
ManualJunctionShape ( )
)Thanks for any help,
3footer

Alert | IP

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

RE: Making Dual Gauge Switches, OkrasaGhia, Aug-17-02, 04:03 PM, (1) RE: Making Dual Gauge Switches, 3footer, Aug-17-02, 05:40 PM, (2) RE: Making Dual Gauge Switches, OkrasaGhia, Aug-17-02, 06:03 PM, (3) RE: Making Dual Gauge Switches, Emu Farmer, Aug-17-02, 08:41 PM, (4) RE: Making Dual Gauge Switches, davidbeach, Aug-20-02, 00:19 AM, (10) RE: Making Dual Gauge Switches, OkrasaGhia, Aug-18-02, 02:42 PM, (5) RE: Making Dual Gauge Switches, 3footer, Aug-18-02, 05:01 PM, (6) RE: Making Dual Gauge Switches, OkrasaGhia, Aug-19-02, 02:14 PM, (8) RE: Making Dual Gauge Switches, 3footer, Aug-19-02, 03:33 PM, (9) RE: Making Dual Gauge Switches, Emu Farmer, Aug-18-02, 10:54 PM, (7)

Conferences | Topics | Previous Topic | Next Topic

I'm not sure this is at all accomplishable, have never tried it.
The two switches are completely seperate thus you would need two different main routes (but it could work anyway). One narrrow gauge as you have defined and one standard gauge. The problem is somewhat similar to the slip switches which in XTracks are done as one section with independent switches in it. It works (sort of) but not without problems. Come to think about it the problem is very similar to the single slip.Your TrackShape looks fine and I take it you get the blue & red poles in RE?A different issue with your TrackSections, it does not look like the sim cares but all of them should have the gauge set to 0.9. The last one which defines a curve MUST have a second entry directly after it defining a similar right hand curve (even if not used by a TrackShape).

TrackSection ( 9037
SectionSize ( 0.9 0 )
SectionCurve ( 163.7769 10 )
)

Strange things will happen when you flip the switch using the 'T' key if this entry is not added.Nice idea with the 'D' prefix, I'll adopt that in the XTracks update I'm finishing now. What does the 'LC' suffix stand for?Keep us updated and don't hesitate to ask questions.


//Okrasa
Chief Engineer
okrasaghia @ yahoo.com
http://www.geocities.com/okrasaghia
http://www.geocities.com/okrasaghia/XTracks



Steam: 4-4-0 8, 0-4-0ST 9, 0-6-0 12, 0-8-0 15, 2-6-0 25, 2-8-0 27, 2-6-6-2 37, 2-8-2 40, 4-6-2 56
Diesels: SW8 16, F3 53, F7 54, FP7 57, CPA 58, RDC 61, GP7 81, GP9 82, RS11 86, H16 87, TM 88

Alert | IP

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

Yes indeed I get the blue and red poles (a double set) as one would expect. The gauge error in the track section entries are a goof, now fixed. The missing right hand curve to complement the left hand was in there, I just clipped it out in the post. Also noticed the 0.287 offset should be -0.287. In the nomenclature, "LC" is for "left rail common" to both gauges. "RC" will be for right common. This only
pertains to curves/switches, as the straights can just be flipped as needed.I rechecked and noticed that of the two black dots for the switch points in the AE, the one on the *standard* gauge side is always highlighted, whether making standard or narrow activity paths
(not the narrow gauge dot as I had written before).Having two mainroutes is an interesting idea. I tried forcing a second one in the obvious way (two entries either both at top,
and again with one above each pair of paths) but that simply made things less usable... AE refused to lay a path into the points, only
from the trailing side.In rerunning another test, I restored a clean version, put in a
dual switch and layed an activity path thorugh the standard side first. The AE reported happy, but when I reloaded it the activity path had been relocated over to the narrow gauge track (!), but the green "start" dot was still on the standard gauge track. Now that's a wierd state.I looked at the slips and turntable for help, but it seemed to me that they all have the property that there is only a single valid path at any one time, so the choice is 1 from 4, or 1 from 9, etc. With the dual gauge, the choice is 2 from 4. Yet it is reasonable to treat it as only 1 of 4 since who cares if the unused points are set. The (unknown) assumptions built into MSTS will make the difference. If you could provide more insight into how the slips may hold the answer, that'd be great.Somewhat off-topic, the other sections I have up and running fine
are 1 degree versions of the narrow gauge curves, a short 0.25m straight (0.2000m is too short, MSTS will balk, but 0.20001 etc is fine... go figure), and straight dual gauge in various lengths. Curved dual gauge will be easy enough, and some specialty pieces.... getting the switches is the tough part. 3footer

Alert | IP

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

I'll do some tests and let you know if I find something out.Although it might have sounded like it I was not suggesting making two mainroute entries. I was pretty sure it would not work, you can leave it out entirely on the other hand.I think the single slip is the section that comes closest, the choice being 1 or 2 from 3. The turntable is nothing but straights on top of each other, no switching at all.I did just send my update to David Beach for inclusion in the tsection.dat, amongst it are 1 & 2 degree 300r narrow gauge, 0.3m straights standard, narrow and dual gauge (the only dual gauge section I have).
Looks like a little duplication but nothing serious.I wish you luck and I'll be back with my findings regarding the dual gauge switch.


//Okrasa
Chief Engineer
okrasaghia @ yahoo.com
http://www.geocities.com/okrasaghia
http://www.geocities.com/okrasaghia/XTracks



Steam: 4-4-0 8, 0-4-0ST 9, 0-6-0 12, 0-8-0 15, 2-6-0 25, 2-8-0 27, 2-6-6-2 37, 2-8-2 40, 4-6-2 56
Diesels: SW8 16, F3 53, F7 54, FP7 57, CPA 58, RDC 61, GP7 81, GP9 82, RS11 86, H16 87, TM 88

Alert | IP

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

Hi that was very interesting as I live in AU with duel gauge it could make building a lot easier, one thing Okrasa is it possable to make an incline and decline rail sections to save having axle breaking bumps at the joints,thanks if you could,

Frank in AU

Alert | IP

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

Okrasa,Your email Saturday missed me by about 45 minutes, I have been out of town since then, trying now to get caught up. It may be a couple of days before I can give enough attention to your file, but I should have the next build posted no later than Thursday evening.David Beach.Palouse & Lewiston

1% Completed

Alert | IP

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

3footer, tried a double switch and had no probles at all with it. Both switches operatable and no problems going where I want either track. Made paths over both in the activity editor, no problems.
If you could mail me a copy of your tsection.dat I could look it over and try to see if there is something wrong in it.Frank, I belive you're asking for a vertical curve section to even out the transition from flat track to a slope. This is not possible to make. You could do the visual shape but there is no way to get the sim to understand it neither is it necessary. When making a slope you should do it in smaller increments. Using the mouse and/or 'End'-key together with keys '2' & '8' you can change the slope of track sections in much smaller steps than if only using '2' & '8'. Use short track sections and increment the slope gradually and you will have a transitions that won't be noticable.


//Okrasa
Chief Engineer
okrasaghia @ yahoo.com
http://www.geocities.com/okrasaghia
http://www.geocities.com/okrasaghia/XTracks



Steam: 4-4-0 8, 0-4-0ST 9, 0-6-0 12, 0-8-0 15, 2-6-0 25, 2-8-0 27, 2-6-6-2 37, 2-8-2 40, 4-6-2 56
Diesels: SW8 16, F3 53, F7 54, FP7 57, CPA 58, RDC 61, GP7 81, GP9 82, RS11 86, H16 87, TM 88

Alert | IP

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

Your dual switch works? Hmmm.I have continued beating on mine, teasing out the differences between your working single slip and my broken dual switch. I found that there was a limit on how close the points could be within a shape. For example, my dual switch had the points offset by 0.287m (the difference in gauge), and the AE was continually confused (though manifested in different ways).When I separated the two switches by a greater distance (ie two side-by-side switches) it worked perfectly. A 'quick' binary search showed that a separation of 0.5m is the minimum that will work. I suspect that the automatic point detection algorithm used by MSTS (note that we never have to explicitly specify the location of the points) is using < 0.5m as a test criteria... at least at some level (they're always given separate dots in the AE, but the interpretation breaks at < 0.5m).And who would ever need two sets of points closer together than that on one shape? (chuckle)So, the slips work fine because their points are further apart than 0.5m. I'm anxious to hear if this hypothesis fits your working dual switch.I have tentatively fixed this by relocating one of the points by 1.5m (possible because of the #183 straight section prior to the diverging track). Unfortunately, this requires an extra end piece to mate up with the disparate points and get them flush so dual track pieces can then be added on. I'm going to try some more to get this end piece folded into the switch shape. 3footer

Alert | IP

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

Damn, you're correct. I did not have them close enough together and when I moved them in to 0.287 from each other the path can't be laid anymore in the Activity Editor. It still works in explore mode but not being able to lay a path over it is a big drawback.
My guess is it would not help to make them two seperate sections either as it seems to be the distance between the points that is the problem.
It's not true that we don't specify the point location, you do that when you specify the SectionIdx, two of them have the same origin and thus form a point.
The relocation sounds like the best option at the moment, maybe you could do the third rail as a seperate shape which then is used with the original a1tpnt10d to make a dual gauge switch? Then it would matter less that their ends are not even as the third rail section would only have the 3' path and will be matched with other third rail parts. Such shapes would be easier to make also as you don't have to bother with the track bed, that you get from the normal gauge piece used.Well, I wish you the best of luck!


//Okrasa
Chief Engineer
okrasaghia @ yahoo.com
http://www.geocities.com/okrasaghia
http://www.geocities.com/okrasaghia/XTracks



Steam: 4-4-0 8, 0-4-0ST 9, 0-6-0 12, 0-8-0 15, 2-6-0 25, 2-8-0 27, 2-6-6-2 37, 2-8-2 40, 4-6-2 56
Diesels: SW8 16, F3 53, F7 54, FP7 57, CPA 58, RDC 61, GP7 81, GP9 82, RS11 86, H16 87, TM 88

Alert | IP

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

Okrasa writes:
>It's not true that we don't specify the point location, you
>do that when you specify the SectionIdx, two of them have
>the same origin and thus form a point. Well, yes and no. The information is there because the two paths share a point, but this is a rule built into MSTS in interpreting the paths. We do not have a special marker or field to explicitly say that path 0 and path 1 make up a switch point.It also seems that MSTS does not like having points anywhere other than the END of a path. Case in point, I tried my switch with the striaght path having two straight sections, and the diverging path having one section that met at the intersection of the straight path's two sections. After placing this beast the RE would run for a while and then crash. The fact that the MS/KUJU switches have a rather obvious avoidance of this otherwise common sense construction indicates that they are conforming to some requirement.I also noted in your single slip that you don't have a single path (the curved one) with a point at each end, but rather open a gap so that they are two paths and use the special end piece to bridge that gap. Is having a point on both ends of a single path another no-no?>maybe you could do the third rail as a seperate shape which
>then is used with the original a1tpnt10d to make a dual
>gauge switch? Well, since it looks like it has to be made from two shapes anyway, I'm opting for a single shape dual switch with a special end piece to bring out the recessed point. The main shape can then have properly animated points and whatever fiddly bits are desired, and it can use a separate trackbed texture (D1cleantrack1.ace) that has the three tie plates and the two lines of oil drips, etc. The special end piece will just be a rail or two to make it visible when NOT in use (in use, it will be hidden by the main switch shape). Bonus is the special end piece is identical (1.51059m straight) for all four switch versions (left/right and left-common/right-common rail) so only one is needed.On to building the animated .s shape files! 3footer

Alert | IP

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

Emu Farmer Okrasa. thanks for the reply, yes I all ready do them with the small sections, but it makes me reach for the glass of wine.

Alert | IP

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