Tehachapi Loop II

[ednadolski]

these are going to be HIDDEN on top of the helix just east of Tunnel 10?

More like "offstage" is the word that I would use.   I wouldn't want to put any scenery or another level of the helix above them, otherwise they would be too hard to access for maintenance.   I'm also planning to mount those switch machines on the top of the roadbed like that.

Are you going to gap those for block detection?

Yes, and since these are so close together my thought is to make them all part of one block.  The four staging tracks on the diverging end thus would each be a separate block. Likewise, the single track on the other side (going thru tunnel 10 and up to the East Walong switch) would also be a separate block.

I am not sure any more about doing optical detectors in the helix, since the more I learn the more I am thinking that I will use a current detection scheme.   Since each staging track thru the helix is so long (like about 2 scale miles each), it should be multiple blocks to allow for at least two trains per track.   I am also thinking it would be good to to have one (perhaps even two) additional, short block(s) in the middle of each track.  These would serve both as a buffer between trains, and also as an 'overflow block' in case I want to run an extra-long train.

All that being said, I still think it would be really neat to have optical detectors driving a set of LEDs  on  a panel that would show the progress/occupancy of trains in the helix.  Something like one LED for every quarter-turn of the helix.  I'm just not too sure what the right place would be to put the actual panel.

Have been rolling my own #6's recently using the Fastracks Jig and am yet to try anything free hand.. Looks like time to give it a go.

With curved turnouts like this I found it very handy to pre-bend the rail to the approximate radius with the Fast Tracks roller tool.  Not only does that keep the rail from wandering around when trying to position it, but it also makes the curved frogs & flangeways more accurate and easier to build.


Here is one more part of the puzzle that I've been working on.  This is the other end of the staging tracks:  if the curved turnouts correspond to Mojave, then these would correspond to Bakersfield:




Right now this is just built onto a sheet of 12" x 96"  plywood, but I intend to install it as soon as the partition wall to support it is in place.   I originally designed this with hand-built #12 turnouts, but when I saw that the #10s were actually available I figure I would give them a try.  (Hopefully they will perform satisfactorily but if they turn out to be problematic then I *will* rip them out and put hand-built ones in their place.)  Note that in addition to the four staging tracks (which will run into the bottom of the helix), there is also a siding track for parking a track-cleaning train.  The topmost track (with the two re-railers) is going to be used as a decoder programming track and for speed-matching locos.


Ed

[jagged ben]

...

Yes, and since these are so close together my thought is to make them all part of one block.  The four staging tracks on the diverging end thus would each be a separate block. Likewise, the single track on the other side (going thru tunnel 10 and up to the East Walong switch) would also be a separate block.

A disadvantage to this is if you might have a train fouling the outside turnout, and there's plenty of room for a train to move on the inside track, but your CTC system says no.   Mind you, I don't think prototype does it much differently, but sometimes we modelers want to use every inch to pass the longest trains.  Admittedly, with your double-train-length helix it's probably a non-issue. 

All that being said, I still think it would be really neat to have optical detectors driving a set of LEDs  on  a panel that would show the progress/occupancy of trains in the helix.  Something like one LED for every quarter-turn of the helix.  I'm just not too sure what the right place would be to put the actual panel.

While it might not work out to quite the same effect, it would be probably much easier to do this your regular current sensing blocks and outputs from an RR-Cirkits tower controller to drive the LEDs.  Or just make it a JMRI screen you can switch to.   

[GaryHinshaw]

I agree with jb on both points: a single block has some disadvantages, but its probably fine for your needs; and a jmri screen is way easier than an LED panel.

[ednadolski]

Good points esp. about the JMRI screen.  It would take a lot of wires to drive that kind of LED display.   

This seems like perhaps an opportune time to talk more about blocks and signalling.   Here is what I have so far (sorry for the large image, the size is needed to see the details):



The blocks & signals for the onstage portion of the layout are taken from the prototype so these are pretty much a given.  The rest of it is about setting up the blocks for the staging.  Note that in this drawing, the circular helix tracks are not shown, otherwise the drawing becomes impossible to read.  Just keep in mind that there are actually four full turns of the helix tracks in there between the curved turnouts and where the staging tracks emerge onto the bottom level.

The turnouts at either end of the staging tracks are each their own blocks, so any train must clear that block before another can enter it.  (This is pretty much the same as for the onstage blocks.)  My thoughts at this point are to divide each of the four long staging tracks into five blocks apiece (this is not on the drawing either).  There would be a 'long' block at each end to allow for serially staging two full-length trains per track.   Each track would also have two 'short'  blocks (precise length is TBD) that could be used to soak up an extra-length train that 'overflows' the end block.  Finally there would also be a 'guard block' (perhaps one-quarter turn in length) at the center of each staging track.  The whole idea behind this scheme is to allow serial staging of variable-length trains on each track, while simultaneously ensuring that there will always be at least one unoccupied block between trains.

I don't have much experience with serial staging, but that last point seems pretty important since the helix/staging will be unobservable during normal operations. For a train to enter a staging track, the requirement is to have a sufficient number of clear blocks assigned to hold the full length of the train.  Trains are stopped as soon as they exit/clear the last unassigned block, so there is no chance of a collision as long as the assigned blocks are long enough to hold the entire train.  This is also followed when advancing a train in the staging to the next set of blocks vacated by the previous train.

This scheme relies on the correct directions always being followed.  Southbound trains (which run uphill on the Loop) are always assigned to staging tracks 1 and 3 (see drawing), and northbound trains (going downhill on the Loop) always use staging tracks 2 and 4.   Unlike the hard-wired blocks, these assignments can be changed if desired.

My goal is to allow a dispatcher to send trains in a prototypical order, which typically will be several trains in one direction followed by several trains in the other.  There is also the opportunity for meets at Walong siding, where longer trains can pass or overtake shorter ones.  Straightforward enough, but an ops session still should take a while considering that the trains are limited to a maximum scale speed of 23 mph.

Thoughts?

Ed

[ednadolski]

Uncached image link : https://i.imgur.com/NGGOZdH.jpg

[Ed Kapuscinski]

Well, the prototype answer would be a single block, and having trains proceeding into staging at Restricted Speed.

But that doesn't seem possible. Are you trying to use the detection to figure out where to stop something while being staged? That seems to be the complicating factor.

[ednadolski]

Are you trying to use the detection to figure out where to stop something while being staged? That seems to be the complicating factor.

Yes, it seems to me that stopping a train as soon as it clears should be more reliable than waiting for it to enter the next block.  Otherwise it would be harder to tell how much actual distance is between the train and the previous one.

Ed

[GaryHinshaw]

I'm not sure I'm grasping the block arrangement.  Is it like this (for the right-to-left direction)?

|--------------Long------------------|-----Short-----|--Gap--|--------------Long------------------|-----Short-----|

If you are really planning to park trains blindly, I would put another gap block on the left, as an indication to the engineer that the lead loco has reached the end of the staging track:

|--Gap--|--------------Long------------------|-----Short-----|--Gap--|--------------Long------------------|-----Short-----|

You would basically park trains with their lead units in the gap blocks.   Also, I'm not sure I see great value in the short blocks.  If you are content to have two trains per track, what additional functionality do they really offer?

[ednadolski]

I'm not sure I'm grasping the block arrangement.  Is it like this (for the right-to-left direction)?

|--------------Long------------------|-----Short-----|--Gap--|--------------Long------------------|-----Short-----|

Close, it would be more like this:

|--------------Long------------------|-----Short-----|--Gap--|-----Short-----|-------------Long------------------|

So that way it is symmetric if used in either direction.

Admittedly I have not yet crunched any numbers to see how this works out with the train lengths that I have in mind.  That is something I need to do before I start laying track in the helix, so that I know the size and location of each block.  I suppose another option is to build with multiple, smaller blocks and then bring all the feeder wires out to a terminal block so that I have some wiring flexibility in the future.

If you are really planning to park trains blindly, I would put another gap block on the left, as an indication to the engineer that the lead loco has reached the end of the staging track:

My thought was that the engineer would stop his train at the point when it clears the last (ie. previous) block, since that is something that can be readily observed on a panel.   So in the above drawing the trains are sort of 'right justified', which does eliminate the possibility of hitting another train as long as the train lengths do not exceed the assigned block(s).   Perhaps tho that is a sufficient departure from standard practice that operators would find it too confusing.

Does the JMRI panel give an indicator of when a train is in motion?   Does that require transponding to be set up?

Also, I'm not sure I see great value in the short blocks.  If you are content to have two trains per track, what additional functionality do they really offer?

The main idea of the short blocks is that a train can be longer than half the staging track length and still can fit on one staging track with a shorter train, while always having at least one unoccupied block in between.  As per above the longest train possible train would be 'Long' + 'Short' + 'Gap' (and it may turn out that 'Short' and 'Gap' end up pretty close if not equal).  Of course if all the trains were less than half a staging track length then it would be much simpler to just divide each staging track into two equal blocks.

Ed

[GaryHinshaw]

If I'm reading the tiny font on your plan correctly, you have about 800-900 inches per staging track, or about 66-75 feet.   That is a lot of capacity!  Do you have a target length for your longest routinely run trains?  If it is less than 24 feet, I would recommend dividing the length in thirds with short gap blocks.  (I'm still mezzo-mezzo on the usefulness of the "short" blocks.)  It could be something like this:

|--1'--|------------20-23'------------|--1'--|------------20-23'------------|--1'--|------------20-23'------------|--1'--|

or this:

|--1'--|----------------------30-36'----------------------|--1'--|---------------------30-36'----------------------|--1'--|

Or you could divide the shorter inner tracks into 2 long sections, and the longer outer ones into 3 short blocks.  A few other comments:

* For hidden staging, I would not count on current sensing detection to be 100% reliable about sensing freight cars at the end of a train.  It is very reliable for locos however, which is why I recommend the very short gap blocks to detect when the head-end locos arrive at their destination.

* Current sensing does not detect direction of travel, but jmri does have a useful feature called memory variables which can track a train that traverses contiguous blocks, as long as you provide the initial conditions by hand.  It is then very simple for PanelPro to display the train ID on the panel next to it occupied block(s).

[ednadolski]

If I'm reading the tiny font on your plan correctly, you have about 800-900 inches per staging track, or about 66-75 feet.   That is a lot of capacity!

Indeed it is! [1]  Altho helix staging has no shortage of tradeoffs, that is one of the benefits - it can just soak up a lot of train! And while this is more capacity than I expect I will ever need, the staging helix for my current Loop filled up way quicker than I thought it ever would.  So it does seem preferable to build the extra capacity from the start rather than having to undertake a future expansion project.

I've re-run my numbers,  and according to my plan then there is actually a little more capacity than I thought:

Track 1 (southbound): 831" (69' 3")
Track 2 (northbound): 871" (72' 7")
Track 3 (southbound): 911" (75' 11")
Track 4 (northbound): 952" (79' 4")

And here are the train lengths I came up with (avg. car = 4.2", engine = 6"):  [2]

Short:  60 cars + 5 engines:  282"  (23' 6") (equivalent proto = 4700') (fits on Walong siding)
Typical: 80 cars + 6 engines:  372"  (31' 0") (equivalent proto = 6200')
Mega: 100 cars + 8 engines: 468" (39' 0")  (equivalent proto = 7800')

My thought is that I should have at least one 'short' train for each direction, in order to allow for Walong siding meets.  I expect only to ever have one 'mega' train.... probably a UP southbound with DPUs, such as in this video: https://youtu.be/Ly1nREJ6l8k?t=2m57s


So one of the southbound staging tracks should handle the 'short' and the 'mega' train, with the rest having two trains per track of either 'short' or 'typical' lengths.  For hardwired current detection blocks, I may set up Track #1 something like your first illustration (from your last post)  to handle the short+mega, and the remaining tracks as in your second illustration, each to handle a typical+typical.

For hidden staging, I would not count on current sensing detection to be 100% reliable about sensing freight cars at the end of a train.  It is very reliable for locos however, which is why I recommend the very short gap blocks to detect when the head-end locos arrive at their destination.

Thanks, this is a pretty important point.  This was one of the concerns that initially led me consider optical detection, but I don't have enough experience with either method to really be sure of all the potential issues & tradeoffs.

Do you think that if I go with the 1' detection blocks, I could defer equipping freight cars with resistor wheelsets?  That would help me to get things running, even tho I still would eventually need the wheelsets for the signalling system of course.


Ed

[1] The full-res drawing is more readable, please see https://i.imgur.com/NGGOZdH.jpg

[2] I will of course be starting with much shorter trains than these, and then 'grow' into the target lengths over time.

[GaryHinshaw]

Do you think that if I go with the 1' detection blocks, I could defer equipping freight cars with resistor wheelsets? 

Yes, I think this would work quite well.  You'll need to keep track of train length "by hand" to know if a given train is still occupying any trailing blocks, but that shouldn't be hard to manage.  And since most/all of your loco consists will exceed a foot in length, they will trigger detectors in both the gap block and the long block, whether or not you have resistive wheels, so the only unknown is whether your train exceeds a long block.

And while this is more capacity than I expect I will ever need, the staging helix for my current Loop filled up way quicker than I thought it ever would. 

QFT.

[ednadolski]

You'll need to keep track of train length "by hand" to know if a given train is still occupying any trailing blocks, but that shouldn't be hard to manage.

Agreed.   I don't anticipate re-staging or changing the trains very often.

And since most/all of your loco consists will exceed a foot in length, they will trigger detectors in both the gap block and the long block, whether or not you have resistive wheels, so the only unknown is whether your train exceeds a long block.

This makes me wonder if I can get by without any detection in the long blocks.   Probably best to wire the feeders that way, and then install the actual detection hardware later, as-needed.


I put together some cheesy ACII art of the layout diagram (shows just how lazy I can be sometimes, not wanting to bother with setting up Panel Pro right now to do it  )    This shows the block arrangement on the rest of the layout, in addition to the block lengths for the staging:




So Track 1 can be for staging one short train and one mega train. The latter will occupy up to 1+15+1+25 = 42', and they can be put in either order.   The rest of it is generally intended for two trains per track (either short or typical) and can be used for either direction.   Track 4 is the outer helix track, and thus is so long that it could nearly handle two mega-length trains!   

Ed

[svedblen]

60 cars is "short"! :Dn

[jagged ben]

I agree with the Gary about the 1' sections.   (Heck, maybe 6 or 8 inches is enough).  On our club layout we have set up 1' detection sections in front of each signal for the eventual purpose of automatic train running (which hasn't actually happened yet).    Generally this means a 1' detection section at the end of each block.

FWIW, we've also divided the remaining lengths of the block in two, the idea being that when a train gets half way through a block it will start slowing down to a crawl so that when it reaches the end section at the signal it will creep to a stop realistically.   The 'half-block' sections also come in handy for 'early tumble-down' when going from sidings to single track.    Not that any of this necessarily applies to your operational scheme, unless you also like the idea of automatic running...