DCC Bus reversing question


Blair
 

I have two large staging yards.  They are accessed via two parallel tracks that enter an optional crossover - trains can go straight through either side of the crossover, or can cross over.  This allows trains to enter staging and face in either direction;  Or, you could think of it as, any train in staging can enter the layout going in either direction.  However, the layout DCC polarity is fixed.  This means that the entire staging yard needs to be a large reversing section.  My intent was to either reverse, or not reverse, the output of the booster feeding the yard with a relay, dependent on whether the crossover is set to X or ||.

This will work, but the relay approach leaves every loco in the yard seeing a momentary power loss, causing the noisy ones to recycle.  Not fun.

I don't really want to reverse using the Booster Autoreverse, as it then needs to be configured to flip faster than the circuit breakers feeding sections of the yard(can that be done reliably?).  For example, if I use either a PM42 or 4 PSX units to subdivide the yard, the booster has to reverse polarity before those elements react to the short when a train enters the crossover if it's X, if the crossover was previously set to ||.

If I could find a 10A DPDT solid state relay equivalent, I think I'd be away to the races, as this could then simply throw when the crossover changes state.

Suggestions?  I haven't explained it very well, so ask for clarifications if you need to.

Blair


wirefordcc
 

Hi Blair,

For people that have some form of electronic circuit breakers and need reversing, think about products like the PSX-AR that has both in one unit.  The obvious advantage is that they are designed to work together.  They are all solid state, too, so you don't have the relay switching problem.

Allan Gartner
Wiring for DCC


Blair
 

Allan

My problem is a bit more complicated.   I think I'll generate a picture and post it later, it will explain things better.  Essentially, if an area served by 3 or 4 PSX breakers must all be either normal or reversed, why would you do it incrementally with several PSX-AR units, particularly as they are electrically adjacent and therefore violate the "no AR sections adjacent" rule previously discussed many times?  I realize that rule may or may not apply to PSX-AR units, but it's still a good practice.

Blair


On 2021-08-08 8:54, wirefordcc wrote:
Hi Blair,

For people that have some form of electronic circuit breakers and need reversing, think about products like the PSX-AR that has both in one unit.  The obvious advantage is that they are designed to work together.  They are all solid state, too, so you don't have the relay switching problem.

Allan Gartner
Wiring for DCC


wirefordcc
 

Hi Blair,

I'll wait until you post your diagram.  You are right, the rule about avoiding adjacent reversing units is a good one.

Allan


Don Weigt
 

I've only just begun to use auto reversing, and had no problems so far on my modest applications. But, the thought has occurred to be that boosters should have an external phase control. In cases like Blair's, the reversing could be controlled by some external event and not need to use auto reverse or an external switch or relay of any kind.

I'm surprised that it isn't a standard booster feature. It's simple enough to invert the driving signal from the command station to reverse the output phase. One XOR logic gate will do it... I may investigate adding that to my EasyDCC dual boosters.

Don Weigt
Connecticut

--
Don Weigt
Connecticut


Blair
 

That's a good point, Don.  Instead of a >10A
DPDT to reverse the booster outputs, I could swap the Railsync signal wires.  If I dedicate a booster to each yard that would work, if I find a small solid state DPDT.
Hmm.  Makes me want to resurrect my DCS100 and DB150 for staging and put my 8A booster elsewhere.

Sent from his Galaxy



Jim Betz
 

Blair/Don ... and all,

  You don't need to re-invent the wheel.  DCC works just fine as designed.  DCC wiring - if
you follow Gartner's recommendations - is also easy and fool proof (if you follow them
correctly).  Specific responses:

  1) No you don't need to swap railsync wires - and you shouldn't do so.
  2) Yes, you can use some kind of physical switch to "rewire" reversing
       sections.  Most guys like the automated aspect of a Reversing DCC
       circuit breaker.
  3) The best (in my opinion) DCC circuit breakers are the PSX series.  They
       work and work well.  They were designed with sound in mind.  They are
       electronic (which makes them fast).  Yes, they cost a bit more than 
       other brands of DCC circuit breaker - yes, they are worth it.
  4) A reversing section should not be up against another reversing section if
      you are using DCC circuit breakers.  They end up trying to out guess each
      other - and that competition usually goes on forever.  You can, with some
      breakers, set a different trip speed to prevent this ... but it is better to not
      have two reversing sections next to each other.  This is almost always
      possible thru careful selection of where your breaker boundaries are.
  5) A reversing section should be "longer than the longest train you will
      ever run thru it".  It does not have to be "the entire loop" - but most guys
      will set it up that way and it's usually easier.  Another way to say this one
      is "you don't want both ends of the train - or two trains at once - crossing
      the reversing section".  Again, the breaker gets confused and, in this case,
      ends up competing with itself.

  Reversing breakers work like this:

  a) Situation normal.
  b) A short occurs.
  c) The breaker -temporarily- tries reversing the polarity of the reversing
      section.
  d) If that works it leaves it alone and the train continues to run.
  e) If it didn't work then you have an "actual short" and the breaker goes
      into "there's a short" status.
  f) Most modern DCC circuit breakers also have an automated Retry.
     They will wait a second or two and then retest to see if the short is
     still there - this is what allows you to throw the switch and "clear the
     short".

  On my layout I have just one reversing section - it is the huge loop in
staging - the staging is 5 tracks wide and 3 trains long.  All trains enter
(and exit) staging thru just one track that goes to a switch.  All of the
track past that switch is both Staging and the reversing loop.  I have it
wired with just one circuit breaker.  If it ends up that there are too many
trains in staging (amp draw too high) I will cut the power to some of the
tracks (the ones that the turnouts make "unavailable".
                                                                                                - Jim

  


Don Vollrath
 

However to answer the original question…
You SHOULD be able to wire up an auto-reverser from DCC mains (or use an auto-reversing booster) to feed several PSX (or other brand) electronic circuit breakers that feed several CB protected districts inside a common reversing loop/section. Just make sure each of the circuit breakers react faster than the auto-reversing controller and that there is an initial section of track at both sides of the single AR loops district entrance wired before any CB protected track.
However I’m not sure you would want to do that as the troublesome areas will be at the turnouts and it sounds like costly non-effective overkill.

DonV


Blair
 

Thanks for the generic DCC reversing how to, Jim; I'm sure it is well intentioned.  Doesn't address my question exactly, but others will benefit.

I have to reverse a large area of my layout(up to 16 trains); that area is driven by it's own booster, BUT I want multiple sub-sections protected by SS breakers.  Though each track of my staging can be shut off, the upshot is I could have up to 4 multi-unit trains running simultaneously in the yard and approaches.  I want those trains to proceed without a glitch when I change the alignment of the |X|, so the reversing needs to be solid state.  I also want the yard and the approach tracks on separate circuit breakers.  So I see my booster either reversing and feeding breakers, or feeding an autoreverser* which then feeds two or more breakers. 

Your description of your layout is similar, though smaller I expect, than my situation.

The autoreverser must be FASTER than the circuit breakers in order to reverse before they act, else it won't act.  If the short is sustained (i.e. not resolved by the AR), then I want the circuit breakers to act BEFORE the AR swaps the polarity again; I strongly suspect the AR actions will suppress the sensing of a short by the breakers, resetting their timing.  So I don't think this will work well - I expect it will flip-flip-flip ad nauseum. 

Alternatives:

- Reverse the Loconet Railsync signals, which you don't seem to like, but which I see no fundamental issue with, or

- use an autoreverser that can be triggered externally by the same signal driving the tortoises on the |X|.  I can't quite see how to do this using the PSX-AR, but I may just have misread the manual.  Comments appreciated if someone has done this.

I'm presently reviewing my staging plan to see if I can identify a single stretch of track in each approach that could be considered a "reversing section" of it's own, thus allowing the staging yard itself to be fixed polarity.  Problem is, if I do that, I think I can still have multiple trains entering/leaving it simultaneously, due to the layout, which wouldn't be acceptable.  These concerns are what resulted in my "flip-it-all based on one crossover state" approach.

Blair



On 2021-08-08 20:39, Jim Betz wrote:
Blair/Don ... and all,

  You don't need to re-invent the wheel.  DCC works just fine as designed.  DCC wiring - if
you follow Gartner's recommendations - is also easy and fool proof (if you follow them
correctly).  Specific responses:

  1) No you don't need to swap railsync wires - and you shouldn't do so.
  2) Yes, you can use some kind of physical switch to "rewire" reversing
       sections.  Most guys like the automated aspect of a Reversing DCC
       circuit breaker.
  3) The best (in my opinion) DCC circuit breakers are the PSX series.  They
       work and work well.  They were designed with sound in mind.  They are
       electronic (which makes them fast).  Yes, they cost a bit more than 
       other brands of DCC circuit breaker - yes, they are worth it.
  4) A reversing section should not be up against another reversing section if
      you are using DCC circuit breakers.  They end up trying to out guess each
      other - and that competition usually goes on forever.  You can, with some
      breakers, set a different trip speed to prevent this ... but it is better to not
      have two reversing sections next to each other.  This is almost always
      possible thru careful selection of where your breaker boundaries are.
  5) A reversing section should be "longer than the longest train you will
      ever run thru it".  It does not have to be "the entire loop" - but most guys
      will set it up that way and it's usually easier.  Another way to say this one
      is "you don't want both ends of the train - or two trains at once - crossing
      the reversing section".  Again, the breaker gets confused and, in this case,
      ends up competing with itself.

  Reversing breakers work like this:

  a) Situation normal.
  b) A short occurs.
  c) The breaker -temporarily- tries reversing the polarity of the reversing
      section.
  d) If that works it leaves it alone and the train continues to run.
  e) If it didn't work then you have an "actual short" and the breaker goes
      into "there's a short" status.
  f) Most modern DCC circuit breakers also have an automated Retry.
     They will wait a second or two and then retest to see if the short is
     still there - this is what allows you to throw the switch and "clear the
     short".

  On my layout I have just one reversing section - it is the huge loop in
staging - the staging is 5 tracks wide and 3 trains long.  All trains enter
(and exit) staging thru just one track that goes to a switch.  All of the
track past that switch is both Staging and the reversing loop.  I have it
wired with just one circuit breaker.  If it ends up that there are too many
trains in staging (amp draw too high) I will cut the power to some of the
tracks (the ones that the turnouts make "unavailable".
                                                                                                - Jim

  


Blair
 

Attached is a rudimentary sketch to illustrate my staging yard. (remember, there are two such, very similar).

You can see, I hope , how it makes sense to me to simply reverse the entire B/C loop section, and thereby avoid other complications.  I looked at using an AR for each approach, e.g. L to N, or even just M to N, and similarly Q or R to S; in each case, there are use cases where the reversing block thus created could have a train entering and leaving.  Only the crossover bottleneck presents a single, unambiguous reverse vs don't-reverse use case.

I think, on the whole, reversing the booster would be best, if only the booster fed only sections B and C.  However, given that I still have hopes of initially running both staging areas from one booster, as shown, that option is off the table, leaving me with only one - reverse using a DPDT relay dedicated to each staging yard, as shown.  I don't want to use an AR here, as an AR will be used for track section D, and it will be connected to B, forming the dreaded "adjacent AR" conundrum.

Blair


Blair
 

oops.  I forgot, this list strips attachments.

See Sketch "loop schematic e.jpg" in the files.  It'll be there momentarily.

Blair

On 2021-08-18 18:00, Blair wrote:
Attached is a rudimentary sketch to illustrate my staging yard. (remember, there are two such, very similar).

You can see, I hope , how it makes sense to me to simply reverse the entire B/C loop section, and thereby avoid other complications.  I looked at using an AR for each approach, e.g. L to N, or even just M to N, and similarly Q or R to S; in each case, there are use cases where the reversing block thus created could have a train entering and leaving.  Only the crossover bottleneck presents a single, unambiguous reverse vs don't-reverse use case.

I think, on the whole, reversing the booster would be best, if only the booster fed only sections B and C.  However, given that I still have hopes of initially running both staging areas from one booster, as shown, that option is off the table, leaving me with only one - reverse using a DPDT relay dedicated to each staging yard, as shown.  I don't want to use an AR here, as an AR will be used for track section D, and it will be connected to B, forming the dreaded "adjacent AR" conundrum.

Blair







Blair
 

In the photos area, under my name, as we now can't load JPG files to the files area.

Blair

On 2021-08-18 18:09, Blair wrote:
oops. I forgot, this list strips attachments.

See Sketch "loop schematic e.jpg" in the files.  It'll be there momentarily.

Blair


On 2021-08-18 18:00, Blair wrote:
Attached is a rudimentary sketch to illustrate my staging yard. (remember, there are two such, very similar).

You can see, I hope , how it makes sense to me to simply reverse the entire B/C loop section, and thereby avoid other complications.  I looked at using an AR for each approach, e.g. L to N, or even just M to N, and similarly Q or R to S; in each case, there are use cases where the reversing block thus created could have a train entering and leaving.  Only the crossover bottleneck presents a single, unambiguous reverse vs don't-reverse use case.

I think, on the whole, reversing the booster would be best, if only the booster fed only sections B and C.  However, given that I still have hopes of initially running both staging areas from one booster, as shown, that option is off the table, leaving me with only one - reverse using a DPDT relay dedicated to each staging yard, as shown.  I don't want to use an AR here, as an AR will be used for track section D, and it will be connected to B, forming the dreaded "adjacent AR" conundrum.

Blair











whmvd
 

Blair,

In all your deliberations, do not forget that all autoreversers also have their own maximum sustainable track current.

Wouter


On Wed, 18 Aug 2021 at 19:47, Blair <smithbr@...> wrote:

Thanks for the generic DCC reversing how to, Jim; I'm sure it is well intentioned.  Doesn't address my question exactly, but others will benefit.

I have to reverse a large area of my layout(up to 16 trains); that area is driven by it's own booster, BUT I want multiple sub-sections protected by SS breakers.  Though each track of my staging can be shut off, the upshot is I could have up to 4 multi-unit trains running simultaneously in the yard and approaches.  I want those trains to proceed without a glitch when I change the alignment of the |X|, so the reversing needs to be solid state.  I also want the yard and the approach tracks on separate circuit breakers.  So I see my booster either reversing and feeding breakers, or feeding an autoreverser* which then feeds two or more breakers. 

Your description of your layout is similar, though smaller I expect, than my situation.

The autoreverser must be FASTER than the circuit breakers in order to reverse before they act, else it won't act.  If the short is sustained (i.e. not resolved by the AR), then I want the circuit breakers to act BEFORE the AR swaps the polarity again; I strongly suspect the AR actions will suppress the sensing of a short by the breakers, resetting their timing.  So I don't think this will work well - I expect it will flip-flip-flip ad nauseum. 

Alternatives:

- Reverse the Loconet Railsync signals, which you don't seem to like, but which I see no fundamental issue with, or

- use an autoreverser that can be triggered externally by the same signal driving the tortoises on the |X|.  I can't quite see how to do this using the PSX-AR, but I may just have misread the manual.  Comments appreciated if someone has done this.

I'm presently reviewing my staging plan to see if I can identify a single stretch of track in each approach that could be considered a "reversing section" of it's own, thus allowing the staging yard itself to be fixed polarity.  Problem is, if I do that, I think I can still have multiple trains entering/leaving it simultaneously, due to the layout, which wouldn't be acceptable.  These concerns are what resulted in my "flip-it-all based on one crossover state" approach.

Blair



On 2021-08-08 20:39, Jim Betz wrote:
Blair/Don ... and all,

  You don't need to re-invent the wheel.  DCC works just fine as designed.  DCC wiring - if
you follow Gartner's recommendations - is also easy and fool proof (if you follow them
correctly).  Specific responses:

  1) No you don't need to swap railsync wires - and you shouldn't do so.
  2) Yes, you can use some kind of physical switch to "rewire" reversing
       sections.  Most guys like the automated aspect of a Reversing DCC
       circuit breaker.
  3) The best (in my opinion) DCC circuit breakers are the PSX series.  They
       work and work well.  They were designed with sound in mind.  They are
       electronic (which makes them fast).  Yes, they cost a bit more than 
       other brands of DCC circuit breaker - yes, they are worth it.
  4) A reversing section should not be up against another reversing section if
      you are using DCC circuit breakers.  They end up trying to out guess each
      other - and that competition usually goes on forever.  You can, with some
      breakers, set a different trip speed to prevent this ... but it is better to not
      have two reversing sections next to each other.  This is almost always
      possible thru careful selection of where your breaker boundaries are.
  5) A reversing section should be "longer than the longest train you will
      ever run thru it".  It does not have to be "the entire loop" - but most guys
      will set it up that way and it's usually easier.  Another way to say this one
      is "you don't want both ends of the train - or two trains at once - crossing
      the reversing section".  Again, the breaker gets confused and, in this case,
      ends up competing with itself.

  Reversing breakers work like this:

  a) Situation normal.
  b) A short occurs.
  c) The breaker -temporarily- tries reversing the polarity of the reversing
      section.
  d) If that works it leaves it alone and the train continues to run.
  e) If it didn't work then you have an "actual short" and the breaker goes
      into "there's a short" status.
  f) Most modern DCC circuit breakers also have an automated Retry.
     They will wait a second or two and then retest to see if the short is
     still there - this is what allows you to throw the switch and "clear the
     short".

  On my layout I have just one reversing section - it is the huge loop in
staging - the staging is 5 tracks wide and 3 trains long.  All trains enter
(and exit) staging thru just one track that goes to a switch.  All of the
track past that switch is both Staging and the reversing loop.  I have it
wired with just one circuit breaker.  If it ends up that there are too many
trains in staging (amp draw too high) I will cut the power to some of the
tracks (the ones that the turnouts make "unavailable".
                                                                                                - Jim

  


Blair
 

Wouter

Thanks.  A PSX-AR is capable of reversing all of an 8A booster, so I think I'm safe even if I went that route.  Agreed, other AR units don't have the same specs.  I think reversing the feed upstream of a pair of PSX units is going to be my chosen action. Whether I achieve that with a solid state reverser, or with a DPDT 10A relay, is still open.

Blair

On 2021-08-18 16:31, whmvd wrote:
Blair,

In all your deliberations, do not forget that all autoreversers also have their own maximum sustainable track current.

Wouter
_._,_._,_


Steve Haas
 

"In the photos area, under my name, as we now can't load JPG files to the files area."

Blair,

I've downloaded and reviewed your JPG.

In my opinion you're making this more complex than it needs to be.

I'll get to that, but first answer a couple of questions:

1) These are the only paths that a train uses to reverse - Yes/No?
2) How much of this trackage has already been built?
3) How much of this trackage that has been built has already been wired?

Back to the track plan:

1) If you remove (for discussion purposes) the double crossover at "Y", all you have are three loops around the layout, never needing to reverse electrical phasing:
A) Loop "C',
B) Loop "C-B-A-B-C", and
C) Loop "C-B-A-D-C".

Wire "Rail A" to the outer rail & "Rail "B" to the inner rail (or vice versa) and you are done with this stage of electrical design - no reversing sections, just straight DCS> Booster>PSX and PSX-AR in parallel (not series) as you have currently noted in the Jpg.

2) Now let's reintroduce that double cross-over - this creates the possibility to reverse trains:

A) Is this track already laid? If not, there's no physical rework needed - just rethinking and re-drawing the wiring specs . . .
B) What is the wiring architecture of the crossover? Are the two paths totally isolated electrically from each other? If they are it eliminates a potentially thorny problem. If not, might need a Tam Valley Frog Juicer to manage the four points in the diamond.
C) Now let's address where to place the reversing sections themselves: Knowing only what you've shared with us and the fact that this a logical and not necessarily a scale drawing of the layout, I'd suggest the following:
i) Two reversing sections:
` a) the first is in the upper right quadrant of loop "A", starting one train length before the upper left turnout in the crossover, and extending to the gaps in all four rails to the right of that turnout,
b) the second is in the lower right quadrant of loop "A", starting one train length before the lower left turnout in the crossover, and extending to the gaps in all four rails to the right of _that_ turnout.
ii) One reversing section:
a) If you are not wedded to the double crossover, and the straight section in the middle of the diagram is physically long enough, consider replacing the double crossover with two single crossovers at least a train length apart. There are folks involved in layout design that suggest this increases operating flexibility - I'm not in a position to comment on that, perhaps others will chime in. If you're interested in this approach, let me know and I can help with location of the single reversing section needed to do things that way.

Note that all variations of what I've suggested eliminates the need to have reversers of any type be in series with PSX units. This is important, because as others have said, A PSX-AR is just another PSX with an AR unit added in - they are not designed to be used in sequence.

Give a holler if you would like to discuss further!

Best regards,

Steve

Steve Haas
Snoqualmie, WA


Blair
 

Steve

Interspersed comments below.

Regards

Blair

On 2021-08-18 23:14, Steve Haas wrote:

"In the photos area, under my name, as we now can't load JPG files to the files area."

Blair,

I've downloaded and reviewed your JPG.

In my opinion you're making this more complex than it needs to be.
Believe me, I wish it were so.  Read on.
I'll get to that, but first answer a couple of questions:

1) These are the only paths that a train uses to reverse - Yes/No?
Yes, as far as trains on this portion of the layout.
2) How much of this trackage has already been built?
Almost all, the area around the crossover will go in last, probably at the end of the month.
3) How much of this trackage that has been built has already been wired?
In progress; yard wired, sidings mostly so
Back to the track plan:

1) If you remove (for discussion purposes) the double crossover at "Y", all you have are three loops around the layout, never needing to reverse electrical phasing:
A) Loop "C',
B) Loop "C-B-A-B-C", and
C) Loop "C-B-A-D-C".

Wire "Rail A" to the outer rail & "Rail "B" to the inner rail (or vice versa) and you are done with this stage of electrical design - no reversing sections, just straight DCS> Booster>PSX and PSX-AR in parallel (not series) as you have currently noted in the Jpg.
Sure, I get all that, but not where I'm going.
2) Now let's reintroduce that double cross-over - this creates the possibility to reverse trains:

A) Is this track already laid? If not, there's no physical rework needed - just rethinking and re-drawing the wiring specs . . .
Already committed, not laid, see notes below near end.
B) What is the wiring architecture of the crossover? Are the two paths totally isolated electrically from each other? If they are it eliminates a potentially thorny problem. If not, might need a Tam Valley Frog Juicer to manage the four points in the diamond.
Aware of the issues I may face here, but thanks.  The crossover is 4 turnouts and a diamond, all in PECO Insulfrog C100.  As such, I have near-total flexibility for the wiring presuming I put in enough insulated joiners.  As Insulfrog, I don't have any frogs to worry about, per se.
C) Now let's address where to place the reversing sections themselves: Knowing only what you've shared with us and the fact that this a logical and not necessarily a scale drawing of the layout, I'd suggest the following:
i) Two reversing sections:
` a) the first is in the upper right quadrant of loop "A", starting one train length before the upper left turnout in the crossover, and extending to the gaps in all four rails to the right of that turnout,
b) the second is in the lower right quadrant of loop "A", starting one train length before the lower left turnout in the crossover, and extending to the gaps in all four rails to the right of _that_ turnout.
No, that won't work.  Loop A does more than just a pass around the layout.  It interchanges with another RR (the ACR, the real focus of this monster).  If I made your (a) section reversing, I would be making a mainline-and-passing track-and division point yard(location called White River) all reversing; trains could be entering the passing track at one end while another enters the mainline track at the other, which violates Rule 1.  Similarly with (b), the main and siding at Franz may be entered and exited simultaneously.

ii) One reversing section:
a) If you are not wedded to the double crossover, and the straight section in the middle of the diagram is physically long enough, consider replacing the double crossover with two single crossovers at least a train length apart. There are folks involved in layout design that suggest this increases operating flexibility - I'm not in a position to comment on that, perhaps others will chime in. If you're interested in this approach, let me know and I can help with location of the single reversing section needed to do things that way.
Yes, that would work, but is not physically possible in this design.  I have about 4' where those two tracks are parallel and together; it's either the crossover depicted, or nothing.
Note that all variations of what I've suggested eliminates the need to have reversers of any type be in series with PSX units. This is important, because as others have said, A PSX-AR is just another PSX with an AR unit added in - they are not designed to be used in sequence.
N.B. This design originated with the following assumptions.  1) the reversing of the whole yard would be done with a 10A DPDT relay driven synchronously with the Tortoises of the crossover. 2) the yard, approaches, and X trackage would be handled by four sections of a PM42 downstream of the reversing relay.  As I saw it, then and now, the design will work.  At this point, I'm only trying to accommodate the needs of an increasing number of sound units, both in terms of increased current draw, and in terms of 'loss of fidelity' when something happens(momentary shorts, and of course the act of reversing) and the locos all reset.  The latter is partly addressed by turning off tracks that are not routed to - at a maximum, only two tracks in the yard need to be powered at any one time.  Also, if a passing track is isolated by the turnouts at both ends being set against the train, there is no need for that track to be powered.  But I would prefer electronic protection and reversing, if I can accomplish it.  Track D, by the way, allows me to also use this staging yard as a temporary end point for the ACR, as we slowly build out the double deck 600' mainline.  Without it, I'd need another portable yard to allow me to run trains while the lower half of the layout gets built. Similarly, track D on the second staging level allows temporary termination of the north end of the ACR.  Both of these options are very important at this time, but will likely be removed in the final version of the layout.

Discussion of physical reality.  Loop D runs around the walls of a 10x28 room.  The tracks from N to L runs CCW around the room, rising at 2%; similarly, the track from S to Q runs CW around the room, rising at 2%; point Y actually exists 8" 'above' the Letter D, if we were to try to relate this diagram to reality.  I guess I'll upload an actual track diagram sometime this AM, though it's likely to drag the conversation away from the focus - reversing.

Blair

Give a holler if you would like to discuss further!

Best regards,

Steve

Steve Haas
Snoqualmie, WA







Jim Betz
 

Blair,
  I understand more now (fully?).  Since you are committed to this track plan
here is what I suggest as a "solution that will make things possible/easy
enough that it works".
  Figure out a way that you isolate the 4 foot section with the double
cross-over so that - it is strictly "one train at a time" whenever any
train is crossing over.  If you have a second train entering that 4' section
while another train is crossing over you are going to have a short that
will be unresolvable ... without dragging one or both trains forward or
backward by hand.  
  My solution would be to make the approach tracks on the "other routes"
(other than the acitve cross-over route) completely DEAD for at least as
long a distance as an ABBA set of your longest locomotive - even if you
don't ever intend (now) to run such an ABBA set.  Don't forget that an
ABBA set, with keep alives, can run all the way until the last locomotive
is in the dead track before it stops.  I would wire it up/control it so that
the act of throwing either cross-over route kills the approaches on the
other route.
                                                                                            - Jim


Blair
 

See below

On 2021-08-19 10:06, Jim Betz wrote:
Blair,
  I understand more now (fully?).  Since you are committed to this track plan
here is what I suggest as a "solution that will make things possible/easy
enough that it works".
  Figure out a way that you isolate the 4 foot section with the double
cross-over so that - it is strictly "one train at a time" whenever any
train is crossing over.

How so?  When the four turnouts are aligned such that there are two
|| running tracks, there is no electrical problem whatsoever.  Agreed, when
the turnouts are set "X", two trains can't progress at the same time.  That's
just a physics thing; two objects cannot occupy the same space.

If you have a second train entering that 4' section
while another train is crossing over you are going to have a short that
will be unresolvable ... without dragging one or both trains forward or
backward by hand.

Don't see your point; however, I did not indicate each and every electrical
gap on that drawing - I can do so, if you think it will change your concerns.
The passing siding and main are still powered, and
nothing prevents trains from running on them.  As for the turnouts and
crossing, I will assemble it again and power it, but I've checked this all
out electrically previously.

  My solution would be to make the approach tracks on the "other routes"
(other than the acitve cross-over route) completely DEAD for at least as
long a distance as an ABBA set of your longest locomotive - even if you
don't ever intend (now) to run such an ABBA set.  Don't forget that an

ABBA set, with keep alives, can run all the way until the last locomotive
is in the dead track before it stops.  I would wire it up/control it so that
the act of throwing either cross-over route kills the approaches on the
other route.
There are only two settings, as I've repeatedly set.  All turnouts set for X,
or all turnouts set for ||.  As such, in one setting, the depicted section
resembles a big 'O', and in the other setting it resembles a big '8'.

Perhaps I'm just thick, but I'm not seeing your concerns.  Please explain
the added complexity of the crossover that I'm not seeing.

                                                                                            - Jim


Blair
 

I am about to leave for 2 days; I have become aware that postings are appearing here on groups.io that I am not seeing on email; not sure why, as since some are coming through, it shouldn't be a spam filter.  Since I cannot log onto groups.io for the next two days, I will cease responding until I return to this forum, in order to not add a layer of confusion by not responding to posts I'm actually not seeing.
Thanks, everyone, for hanging in with me as I work through this.
Blair


Jim Betz
 

Blair,

  You missed my point "whenever any train is crossing over".  Said
another way - set it up so that whenever the turnouts are set to
cross-over then another train "can't enter the cross-over section".
Obviously the intended cross over route has to be powered - I'm
saying that the 'other' route should have enough track that is dead
(when the cross over is set) that another train can't attempt to
enter the cross over section from either of the other two approaches.
And that it is going to need to be a considerable length of track in
order to cover the situation of an ABBA set of your longest locos.
I would use "3 or even 4 feet" of dead track to protect that short.
                                                                 - Jim