RMD-4, Wallace Branch, Dorsey Trestle

Colleagues,Here is an image of yet another widowmaker on the Wallace Branch of the NP. This one is on a steady 3.6% gradient, and the curve is 90M, made of 3 dynamic pieces, with 7M tangents in between, so the effective radius is about 95M, and it turns through a full 89 degrees of curvature. At the highest point over the ravine it is 135ft. high, but for the most part does not exceed 110ft. In doing this one, I went back and removed all non-visible polygons, such as the top & bottom of each pile, inner sides of part of the deck, and ends of the deck, which should not cause a problem except perhaps at the ends of the trestle, which we will camouflage with something else. This reduced the overall poly count considerably, so that the total is not much over 4000.
I believe this is the last of the big trestles, but we do have a smaller one further down, and then on the lower level, there is another one that may or may not have existed, it is hard to tell. Also at the top of the Pottsville Loop (lower, west side loop), there was some kind of structure over the S. Fork of the Couer d'Alene River, which we will make a low trestle. There are a couple of small ones on the east slope. We can largely assume that most structures on this branch were timber because originally it was planned to build a base tunnel and improve the alignment, but this never done.J. H. Sullivan

Attachments

http://www.train-sim.com/dcforum/User_files/40b7b92d6a09900b.jpg

Alert | IP

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

RE: RMD-4, Wallace Branch, Dorsey Trestle------------, Hack, May-29-04, 04:55 AM, (1) RE: RMD-4, Wallace Branch, Dorsey Trestle------------, landnrailroader, May-29-04, 08:44 AM, (2) RE: RMD-4, Wallace Branch, Dorsey Trestle------------, markhyams, May-29-04, 09:47 AM, (3) RE: why the 7m straights, landnrailroader, May-29-04, 10:25 AM, (4) RE: why the 7m straights, markhyams, May-29-04, 10:34 AM, (5) RE: why the 7m straights, landnrailroader, May-29-04, 03:55 PM, (6)

Conferences | Topics | Previous Topic | Next Topic

That's a heck of a structure there, Jerry. Nice work!Cheers!
Marc - 3DTrains
http://www.3dtrains.comPro Series SuperPacks

Alert | IP

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

After you get the hang of it, it is really quite easy to build these, just repetitive.Ron Picardi fixed me up some X-brace textures with transparent alpha channels. Then I created a group of braced bents, and sufficiently large double sided transparent polygons (actually two polygons back to back) and applied the texture with appropriate tiling. On any straight portion, you can create a polygon pair, and tile the texture horizontal & vertical enough to fill the whole structure with a minimum poly count.Then you start at one end, and work toward the other. The deck is composed of 15ft. sections, and each is rotated slightly from the previous one to fit the curve. In this instance, each one is rotated 2.5 degrees except the ones around the tangent pieces. (7m tangent between each dynamic piece)It took about 8 hours off and on to do this one, about 12 to do the "S" trestle, but I do think the visual effect is neat, even though what you are really seeing is just a bunch of specialized textures fitted together in a (more or less) orderly fashion.J. H. Sullivan

Alert | IP

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

I'm curious about those 7m tangent pieces. Are they there in order to make the trestle pieces fit properly, or was that something the railroad did to make it a bit less of a widowmaker curve?Mark

Moffat Road Tabernash to Phippsburg
All track complete
Currently working on Mtracks

Alert | IP

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

Neither,We learned sometime ago that it was not wise to put dynamic pieces back to back. Likewise on this kind of a grade, 30 degrees is about as much turning angle as you can have, and still have smooth joints at the ends of the dynamics. So we usually put something between pieces of dynamic. In this case, I tried endpoint 10D pieces but they did not work as well as the 7M straights.So it is there to reduce the effects of MSTS oddities.J. H. Sullivan

Alert | IP

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

That's funny because I have every curve on the Moffat Road I've been working on laid with dynamic, and there are several places with more than a 180-degree turn, on a 2% grade in some places. I just laid them back-to-back. I used pieces of 0.200 radians (about 11.5 degrees) to retain smooth joints. I haven't had any problems with hundreds of curves laid on over 200 miles of track. What kind of problems were you getting with dynamic track back-to-back? I'm worried now! Mark

Moffat Road Tabernash to Phippsburg
All track complete
Currently working on Mtracks

Alert | IP

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

In truth, I, personally have had little if any trouble, but there were a number of comments, early on, about the perils of putting the pieces back to back, and I did have a failure or two on doing .tdb rebuilds, so I opted to seperate the pieces. On the Milwaukee Road mainline, I did not have to use much dynamic, because most of the curves in the mountains are 10 degrees, so I used the endpoint 10d of appropriate sex. Not a perfect solution because there is a little tangent in those pieces, but other than a nosing effect as the engine rounds the curve, it is no bother - and most engines nose a bit on curves (weave from side to side, trust me, I have run them, I know).Since going to this method, I have not had problems with dynamic except when a dynamic starts on one tile, and ends on another - I try to note where the tile boundary is, and make sure that I cross it with some kind of fixed piece.I hope that PI Engr. can solve this problem in their effort, as dynamic track is a good solution for matching prototype curves, but it needs to behave more like a spline curve, and there needs to be verticle spline curves as well.Jerry

Alert | IP

[ Home | MSTS Forum | MSTS FAQ | File Library ] Copyright © 2004 by Nels Anderson. All Rights Reserved.