Tehachapi, BC

[Steveruger45]

I do like the idea of a narrow fascia on the upper level, and I think it would work really well if it gently undulates horizontally with the scenery. One bit of awkwardness is that transition in height at the vertical brace I'd highlighted earlier (which will hopefully go away): perhaps do a gentle vertical transition there, instead of an abrupt one?

Edit: A picture is worth...



In the image above, I attempted to simulate the fascia on the second level; a thin yellow line helps indicate the undulations. Also note that the right end of the upper level fascia curves into and blends with the existing fascia to the right at Walong.

In addition, as a suggestion, I modified the lower level landform so that the fascia rises to meet the upper level fascia at the corner, matching its final curve, hopefully creating a less-awkward transition to the split levels. The lighter-colored portion to the right is an optional removable access panel that simply serves to extend and blend the lower level fascia into the existing fascia at Walong.

FWIW, etc.

Yes, IMHO as per DKS above, and your  “ Curved vertical fascia the whole way?  Follow the access road (where feasible)” this is Pretty much what I had in mind in earlier posts.   Separation but maintain the overall aesthetics. Color of the facia needs to be something definitive to end the scene but subtle enough so as not to jar the eye too much even though, like others have said when operating your focus is elsewhere.   

[GaryHinshaw]

Good input.  I agree that an undulating vertical fascia is a good way to gently partition the two scenes while still looking natural.  I find these full-scale mock-ups (real and virtual) to be very helpful, so thanks.  RE the lower edge of the fascia - I will taper it for sure.  There are many places around the layout where I do that already, the most obvious one being on the desert shelf along the opposite wall:



I can make the transition under the tunnel shelf even more gradual because there is less (read: no) action under that portion of the scene.

[You can also tell that Edison on the lower deck is one of the few flat surfaces left in the layout room, hence a natural place for stuff to accumulate. ]

RE the transition at Walong - the upper deck transition is pretty flexible, subject to the constraint that I don't block off significant aisle space, so I'll play around with that.  However, the middle deck heading under Walong needs to remain visible so operators can see their train as much as possible.  This transition between lower & upper deck is the diciest stretch of the run for most crew because it is quasi-hidden, and I don't want to exacerbate that.  I also need to keep that cubby space easily accessible for all the other tracks that run around back there.  DKS's proposal does have a nice aesthetic though.

[DKS]

... the middle deck heading under Walong needs to remain visible so operators can see their train as much as possible.  This transition between lower & upper deck is the diciest stretch of the run for most crew because it is quasi-hidden, and I don't want to exacerbate that.  I also need to keep that cubby space easily accessible for all the other tracks that run around back there.

Cheap camera/monitor?

[GaryHinshaw]

Time flies... I had hopes that this summer of quarantine would be productive for TBC, but alas almost nothing has happened on the layout since the last post, as a number of work, life, and world events ended up taking priority.  Nonetheless, the fires still burn (bad metaphor, sorry) and things are slowly starting to happen again.

One of the excuses I allow myself for not making progress on scenery is that I want to have the operational infrastructure more or less complete before I make things less accessible.  Aside from a bit of track-work still to finish at Kern Jct and in the industrial areas, the main infrastructure task left is the signal build.   I'm using the wonderful Showcase searchlight kits and more or less following the configurations documented in this invaluable web site, which records the tail end of the searchlight era on Tehachapi.  We last left this story over a year ago at CP Summit with this shot:



While they are very nice kits, the head castings leave something to be desired for two reasons: 1) the visor is shorter than the prototype, and 2) the inside width of the casting is slightly too narrow to accept the LEDs I am using, thus requiring that some material be milled out (a big pain) or the LEDs be inserted at an angle, which causes illumination issues.  (Some of the walls also have holes in them.)  So @C855B to the rescue!  Mike designed and printed some beautiful H2 heads for me on his Photon printer and sent me a very generous care package.  That shipments solved all of these problems, and they look better in the process.  So I have gone back and rebuilt the signals at Summit now with Mike's new heads and have gone ahead and painted them:



I am super pleased with how these turned out and am really glad to have some I consider to be 100% complete, finally. 

One of the cool things about the signals of this era is that they're all a little bit different.  Case in point is this dwarf mast signal that protects the north end of Allard siding:





It's good to be back at it. 

[GaryHinshaw]

But what's a story without a conflict... The astute reader will notice that I have not posted any 2-head masts yet.  Here's an in-progress shot of such a signal:



The problem is that I have yet to find a way to thread all 8 LED leads down the mast tube.  (Each LED comes pre-wired with R, Y, G and Common leads).  I can reliably get 6 leads threaded, and sometimes 7 (with some swearing and hunched shoulders) but never all 8.  Swapping out LEDs is not really an option, nor is swapping out mast tubing for thin-walled SS, because the brass mast is cast into the relay cabinet.  (It is an option for the pedestal-mount masts though.)  So I am looking to using the mast itself as a Common lead for both heads.  The challenge will be to solder the Common leads (shown dangling, above) to the mast without making the joint too conspicuous.  Hopefully practice makes perfect.

[wm3798]

Overhead pole lines...

[Sokramiketes]

Yes, two things I found easier... 1) making the mast a common + and 2) feeding straight wires up the tube instead of bent wires down the tube. 

I use a small wire pick to pull the wires out at each junction: https://www.therailwire.net/forum/index.php?topic=31256.msg396837#msg396837

Your two head problems compound further with three heads, if you need any. 

If you're soldering those ladder connections to the mast anyway, that might be a convenient spot to tie in the common wire.  In the image linked above you can see I go from a solder tab on the LED straight to the post. 

For clean soldering, remember to shave some solder and pick one up with your iron as needed: https://www.therailwire.net/forum/index.php?topic=31256.msg345284#msg345284

[C855B]

... 2) feeding straight wires up the tube instead of bent wires down the tube.  ...

This isn't going to work with pre-wired LEDs, which is what Gary and I are using. As to three heads:

Not a valid vimeo URL
Not using the mast as a common, there are 12 #36 wires fed top-down. The trick is an ngineering.com SS mast instead of the as-supplied Showcase brass tubing: much thinner wall.

[parkrrrr]

I don't really have a sense for the scale involved, so they're probably too large at about 5-6mm square, but I wonder if there's space somewhere to add a pair of WS2811 chips at the top of the mast? You'd need some custom hardware to translate the signals from your existing controller under the layout, but it'd mean you'd only have to run three wires up the mast (or two if you ground the mast.)

[C855B]

There is no room at the inn. The heads as they are, are already about 10% over-scale to barely accommodate the RR-CirKits COB LED at 1.2 x 1.6mm. Like I said, the "wire problem" is addressed by thin-wall SS tubing. A custom circuit solution creates a whole new level of design and implementation overhead, although I would find such a three-wire bus idea interesting for taming the wiring between the control bus - DCC or LCC - and signal bases. I have one junction (of several) that will have over 80 wires between the signal controllers and the masts. 

Everything in the three-head demo is done with available parts and RR-CirKits SHD-2 controller boards. Gary's solution for driving the LEDs is different, but still off-the-shelf.

[DKS]

It may be a technical challenge, but I'll bet you could easily thread all the leads you need down that mast if they were #40 magnet wires. I had to cram 13 of them down similar thin-wall poles to make these stoplights, which had signals in two directions including a left-turn arrow and walk/don't walk signs:

[CRL]

^^^^There’s your answer... magnet wire. Soldering to the mast invites cold solder joints which I suspect you have from the photos. Not a big deal structurally for building these signals, but is a big deal for electrical continuity.

[Wutter]

The problem is that I have yet to find a way to thread all 8 LED leads down the mast tube.  (Each LED comes pre-wired with R, Y, G and Common leads).  I can reliably get 6 leads threaded, and sometimes 7 (with some swearing and hunched shoulders) but never all 8.  Swapping out LEDs is not really an option, nor is swapping out mast tubing for thin-walled SS, because the brass mast is cast into the relay cabinet.  (It is an option for the pedestal-mount masts though.)  So I am looking to using the mast itself as a Common lead for both heads.  The challenge will be to solder the Common leads (shown dangling, above) to the mast without making the joint too conspicuous.  Hopefully practice makes perfect.

Gary if you're interested, I've been able to source a very similar LED to the RRCirKits/Showcase Miniatures RYG with the proper 505nm green, the LED lens itself is 1.5x1.6mm. I still prefer soldering the connections to the back of the LED first, but this also allows you to do the feed-first, solder-after if you prefer.  I've found that for the 0.8mm OD, 0.6mm ID brass mast, (like DKS said) I had to step down the wire size from 38 gauge magnet wire down to 40 gauge to be able to fit 12 wires down the mast no problem.

[GaryHinshaw]

Thanks for the ideas guys.  At this point, the path of least resistance (slight pun intended) is the thin-walled SS tubing.  I just need to source some and try it out.  I'll pursue the other suggestions if that doesn't work out. 

The only real impediment I see is dealing with the cast-in brass masts on the cabinet-mount kits.  Those will have to be extricated somehow.  Mike, how did you manage the relay cabinet on you 3-head signal?

-gfh

[C855B]

A pair of pliers for a good grip on the brass mast and it pulls out fairly easily.