Best Of Tehachapi Layout research help wanted

[bambuko]

Ed, in your earlier message you are saying "...Here is a link to my Loop layout:..."
but there is no link? (just a picture)

[ednadolski]

Ed, in your earlier message you are saying "...Here is a link to my Loop layout:..."
but there is no link? (just a picture)

I don't have a proper web page, just a set of uploaded photos, but here they are:

http://s119.photobucket.com/albums/o153/ednadolski/?start=all

[kiwi_bnsf]

Hi guys,

Excellent to find others interested in Tehachapi, and with similar layouts underway.

Gary - your layout plans are impressive. I like the space you are dedicating to accurate and uncluttered scenes. Your renderings are great!

My plans are much more compressed and are really a compromise to allow more focus on operations and industrial and yard switching at each end of the climb over Tehachapi.

For those of you interested in my track plan, I'm still working on this. However I do have a finalized signaling schematic that should help in understanding what I am modelling.

First up is an overall schematic of the whole layout:




The above Bakersfield yard trackage is very much subject to change - this was the motivation behind my requests for more information. The block and signal labels are placeholders until I have researched accurate ones.


Each location will have a local control panel with block occupancy on the CTC signaled areas (red), turnout indications (green and yellow), and push button turnout controls (black). Below is the mock-up for Mojave:




Again I'd appreciate comments and feedback.

Cheers

Tim

[ednadolski]

I am limited to a similar space for the loop (5.5' by 5') and I have to compromise to 18" radii on the inner part of Walong siding. However it quick eases out to 22" so hopefully with a bit of superelevation it proves reliable.

You're right, the uppermost quadrant of the actual Loop itself is a larger radius than the rest.  I used 21" on that part and 19" on the rest, so you're actually almost right on.  IIRC I uses 1.25" spacing for the siding.  The other thing I did was vary the grade:  I used the WS 3% inclines on the larger radius section and on the straight portion coming out of Tunnel 9.  On the rest I used the 2% inclines.  The extra grade helped to achieve an acceptable-looking clearance over Tunnel 9, even tho the prototype is much more -- 77' vertically.

I didn't use superelevation, since I was concerned about stringlining, and it's only about 2" on the prototype anyway, which scales to 0.012" -- too hard for me to see. 

[ednadolski]

BTW, if you subscribe to MR mag, there are some cool Tehachapi trackplans in the MR database, including the La Mesa club:

http://www.trains.com/mrr/default.aspx?c=tp&id=93&size=&scale=&type=&keywords=tehachapi

Here are some pics of the La Mesa layout:

http://www.pbase.com/intermodal/2008_october_la_mesa_run
http://www.pbase.com/tgcostello/la_mesa&page=all

[GaryHinshaw]

Tim-

I like the schematics you have - it's giving me some good ideas for fleshing out my own terminal plans, an area where I could use all the help I can get.  Is your plan strictly point to point?  For example, what happens to your BNSF line to Barstow?  Maybe you already said in your earlier text...  :  I hope you'll keep us posted on your design progress.

Good MR link Ed.  I don't recall seeing the plan by Les Combs before.  His schematic is similar to what I'm going for in that the SP route is a loop with Mojave morphing back to Bakersfield, while the ATSF route from Fresno seems to be stub-end staging (with no outlet on the Mojave end).  I'm trying to come up with a plan that puts both lines on a similar footing, where the SP(UP) and ATSF(BNSF) are both loops that share the Pass.  Something like this:



where A connects to A' and B to B' in two double-ended yards.

Cheers,
Gary

P.S. I love 3rd PlanIt.  The renderings are incredibly useful for visualization, though it can still be difficult to judge things like duck-under strain, lower level access, etc.  It's also useful for testing ops concept because you can run trains, check siding lengths, see how 89' racks look on S curves, time mainline runs, etc. etc.  Almost as good as running "real" trains...

[kiwi_bnsf]

Hi Gary,

To answer your question, the layout is to be operated point-to-point, but is actually continuous through the staging.

The BNSF and UP lines that diverge at Mojave South Junction rejoin immediately and then descend seven turns of helix, and are then joined by the BNSF/UP "joint trackage" from the lower deck, and then descend another two turns of helix to looped staging on a lower level hanging below the Bakersfield deck.

The staging loops around and has 7 tracks each with room for 30 car trains + 5 locos. The real trick is that the staging runs in a single direction and has four modes:

  1.  Receive train from Mojave
  2.  Dispatch train to Mojave

  3.  Receive train from Bakersfield
  4.  Dispatch train to Bakersfield

Each mode lines the switches correctly into the looped staging so that all trains point in the same direction. This also allows block detectors then detect the inbound trains, select an empty staging track and stop the train automatically (via JMRI).

Modes 1&4 combine to allow continuous running.

Unfortunately the staging helix will be the last component to be installed as this will require a door to be moved   However during construction I can operate trains from Mojave to Bakersfield yards with temporary return loops at each end (loop-to-loop operation).


In other news I have made a research breakthrough today finding that Microsoft Live Earth has a "birds eye" 3/4 angled aerial views of the line from Sandcut to Kern Junction and Bakersfield BNSF yard. The views are incredibly detailed and I now have enough material to refine the lower level track plan. The industries will change a bit as I will now model a tank car loading track at Jastrow refinery west of Bakersfield, and specific packing houses at Edison.

Edison packing houses:




Bakersfield BNSF Yard throat (with old F street station just off picture top right):




Jastrow refinery spur:






Thanks for the MR link, but I am not a subscriber as their mailing to New Zealand is not reliable (I buy from local magazine shops).

Cheers

Tim

[James Costello]

Thanks for the Live Earth news Tim - that will be awesome!

Love those reefer strings  ;D

Oh, just had another look at the BNSF yard photo and I realised you can see where I took some photos from earlier this year.
The photos are here:
http://s284.photobucket.com/albums/ll3/Costellis/Trains%202008/SoCal/
They were taken from both sides of the yard where the access road crosses the tracks.

[ednadolski]

I love 3rd PlanIt.  The renderings are incredibly useful for visualization

Can 3rd PlanIt import graphics (.jpg, .bmp, ...)?  I can do that in Atlas RTS and lay track over the aerial image of the Loop.

Can it generate profiles for making the terrain?  That would be really handy.

I don't recall seeing the plan by Les Combs before.

He hand-built a replica of an SP caboose, and built the layout inside that.  IIRC, he moved to the Tehachapi area and took the whole thing with him.

[GaryHinshaw]

Great photos Tim (and James)!  Live Earth is very impressive - I think I need to switch over from Google maps because these photos are much better.  (MS kicking Google's butt? Wow.)  Your terminal/staging plan sounds intriguing; I think I understand what you're describing, and the automation sounds very cool.

Ed - yes 3rdPlanIt can import graphics.  When I was starting out I took Google images of Caliente and Walong and superposed them over my room drawing.  Then I tried endless scalings, rotations, etc. to find what worked best, given the room constraints.  When I was satisfied, I drew track following the spirit of the line as best I could.

There are a number of ways of handling terrain; what I used for the crude version I showed was contour-based: you draw contours by hand with specified heights, then it forms the terrain sheet to those contours.  The contours are defined by a sequence of connected discrete points and the points are readily movable if you want to edit the terrain.  So I sat there with a topo map of the area and tried to mimic the terrain, accounting for compression, distortion etc.  It's quite handy.

I should add that I don't have any experience with other planning software, so my judgments are only absolute ones, not relative.

BTW, I looked up the Jan'99 MR and saw the article on Les' layout.  I remember it now - very nice!
-gfh

[GaryHinshaw]

Can't quite tell how the tankers are serviced in the refinery photo, but that siding is the perfect excuse for some of TrainCat's loading platforms:

http://traincat2.com/images/models/1000701B.jpg

[Mark4]

Hi there!

I am based in New Zealand...
Cheers

Tim

Can you be a little more specific, please Tim?

Mark
Glenbrook Beach
NZ

[tom mann]

This is a great planning thread.  Thanks to all for sharing/posting.

[Denver Road Doug]

This is a great thread.  I love these.

Kinda not relevant maybe, but I have an "ATSF Track Chart: Barstow to Oakland" reprinted by the Santa Fe Railway Historical and Modeling Society that would probably be immensely helpful were it not circa 1950's.  It was in a lot of items I was selling for a buddy and he's never taken them back and I took a look as there are several of the track charts from various areas.  Didn't realize they were from the 50's but for a second I thought I hit the jackpot.  Anyway, if anyone is modeling an earlier era or thinks they'd like to have it let me know...maybe we can work out a trade or something.

In the meantime I'd be glad to check it if it can answer any questions.  There's a nice yard diagram of Bakersfield showing a huge steam-era locomotive service area.

[randgust]

This is one of the more "wow.....!" threads I've seen in a while.

I'm not speechless but there's not a whole lot to add, either.

My experience in general with operations of that type is that 3 'old' Kato six-axles can do about 35 cars up a 2.5% grade with 13" min curves.  My visible main runs 13" inside, 15" outside.

Newer six-axles, be they Atlas or Kato, seem to have about 2/3 the tractive effort of the old 'beasts'.  I'm a DC guy though, so I really don't care.

If you want mid-train helpers, or end helpers, you've got to go DCC. 

I know there's guys that will rag on this, but its OK.  With the physics and train sizes you'll be looking at here, test things very carefully on the flange sizes before you build an empire that can't take it.  I'm 'right on the edge' of lo-profile performance with three units dragging a train weight like that around a curve uphill on a grade.  It REALLY wants to stringline, and lift cars to the inside and climb the flanges over, usually on the first 3-4 cars of a train.  The slightest power irregularity, and in she comes.  Because of that, I've decided to stay with original flanges.    And because of that, my track of choice is Peco C55.

Now, with metal wheels, really good track standards, testing, testing, testing,.... you may be OK.   I'm not saying it won't work, I'm just saying that I know guys that have spent three years building track before they ran trains on the 'worst area' and discovered to their horror that they'd developed a fatal error.   

And there's no advantage at all, none whatsoever, on using C55 on areas like your helix.

I think the tactics I'd look at is seeing if I could get that helix built and tested, using the train lengths and operations methods you're planning on, before I went further into what may become a FANTASTIC layout.  You're doing an A++++ on the planning end, both research and design.

The other issue I hit, and it should be OK, is the RDA couplers.   Older MT couplers (maybe until the early 1990's) had a mold separation line in the center that caused the 'finger edges' of the coupler (if you think of a coupler as a fist) to be tapered higher in the center.  That led cars to vertically separate in situations of extreme stress - they would randomly uncouple.  This was just a nightmare on my layout when I first built it.  Jim Fitzgerald of Ntrak came up with a method of trimming the edges of the fingers to a mild "V" instead of "^", which resulted in the couplers becoming self-centering under stress instead of pulling apart.  MT adopted that as the "RDA" coupler.  So new MT's have it, but old ones don't.  It can be easily done with a #11 blade to old couplers, don't panic. 

The combinations I've used - Peco C55/C80, hi-pros, RDA couplers, has led to what I consider 'extreme reliability' in performance.  Just test everything, hard, before you literally buy and build yourself into a corner that's expensive to back out of.