Skip to main content

I have finally finished building my layout table, and have laid my outer loop of track, and am fleshing out lengths, edge spacings, etc...

This loop is approx. 34 ft long, and is temporarily being powered by one by one, single power drop.
However, it will eventfully be divided up into three power blocks, two approx. 14 ft long, and one much shorter, approx. 6 ft block which will contain the two Regular 11 crossover switches, leading into and out of the inner loop.

I had originally planned for two power drops for those two longer blocks, one at each end. However, seeing that the entire 34 ft loop is currently running fairly well with one power drop, and that all my locos run pretty well with that:
I am wondering if these blocks might be suitably powered by one, single, centered, power drop, rather that the two, one at each end of those sections, that I had originally planed for.
Note: These blocks are, again, approx 14 ft long and have approx 10 track joints each.

Does this make sense, or should these blocks still have two power drops?
I want to keep things as simple as I can, but I don't want to risk functionality for simplicity.

Attached, is a pic of my full table and outer loop, as well as a pic of the Transformer and Controls center.
You will notice to groups of four toggle switches, and one group of 9 toggle switches, that will control all power blocks.
1. The left group of four will control the outer loop, shown in the other pic.
2. The group of four to the right will control the inner loop, not yet installed.
3. The right most group of none, will control all of the sidings

The left most toggle in each group is a DPDT switch for selecting transformer control or full direct power from the brick.
Trans former control for all Lionel, conventional and Lion Chief locomotives.
Direct power from the brick is for powering all MTH DCS locomotives with the cleanest, available, sine wave of current.

Attachments

Images (2)
  • IMG_1221
  • IMG_1211
Last edited by RWL
Original Post

Replies sorted oldest to newest

For DCS, Barry's book recommended no more than 10-12 track joints per block. You should be fine with 10. I think my track loops are around 40-42 feet with 4 blocks per loop, blocks are fairly evenly divided and also around 10 track joints per block. Works just fine.

I also have a similar setup to your toggle switches, but only one for each loop, then one for each siding. That all works just fine here too.

@rtr12 posted:

For DCS, Barry's book recommended no more than 10-12 track joints per block. You should be fine with 10. I think my track loops are around 40-42 feet with 4 blocks per loop, blocks are fairly evenly divided and also around 10 track joints per block. Works just fine.

I also have a similar setup to your toggle switches, but only one for each loop, then one for each siding. That all works just fine here too.

Just to be absolute clear:
Are you saying I should be fine with one power drop over10 joints?

For DCS to work properly, there should only be one power drop per block, regardless of the block length.

If there are two power blocks on the same drop, you will get reflected waves messing up the signal.

If you don't ever plan to use DCS, you can wire as many drops per block as you like.

If you use fastrack, you will need a lot of drops, because fastrack likes to develop high resistance joints.

Yes, your 10 track joints per block will be fine.  According to Barry's book, track blocks should not exceed 5 to 6 sections of track on each side of the DCS/power connection to the track. That's not word for word from the book but it's very close.

It goes on to say, Each track section counts as one section regardless of it's length. In any case track blocks should not exceed 100 feet in length.

Also, it sounds like the center of the block is what Barry is saying for your block connections.

Although Barry is no longer with us, the book (DCS Companion for TIU & Remote) is still available and is well worth the cost if you are running DCS. Lots of good information in there that is not in the DCS manual. There is also a Wifi version (maybe Wifi Companion, but I forget the exact title?) that covers DCS using the Wifi App.

Edit: What Royboy & Tom Tee say about having only one feed per block is also a very important rule to follow with DCS.

Last edited by rtr12
@rtr12 posted:

Yes, your 10 track joints per block will be fine.  According to Barry's book, track blocks should not exceed 5 to 6 sections of track on each side of the DCS/power connection to the track. That's not word for word from the book but it's very close.

It goes on to say, Each track section counts as one section regardless of it's length. In any case track blocks should not exceed 100 feet in length.

Also, it sounds like the center of the block is what Barry is saying for your block connections.

Although Barry is no longer with us, the book (DCS Companion for TIU & Remote) is still available and is well worth the cost if you are running DCS. Lots of good information in there that is not in the DCS manual. There is also a Wifi version (maybe Wifi Companion, but I forget the exact title?) that covers DCS using the Wifi App.

Edit: What Royboy & Tom Tee say about having only one feed per block is also a very important rule to follow with DCS.

Thanks for the absolute confirmation.
This does simplify my wiring project a fair amount.
I really do appreciate it.

I wonder if the gauge of the power wires running to the track have any effect on the general rule.  I just use 18 gauge wire, so I have a connection about every 6 sections.    I seem to have no problems at all with power in the track and larger engines will run at full speed around the layout.

Where I have seen power issues is if I have an turnout, following by a short stretch of maybe 3 sections, followed by another turnout.  I almost always get sluggish running on that short section between the turnouts.  So, it is a bit of a pain, but I always run a set of power lines to those short sections, which solves the problem.

It is tedious to run power wires to different parts of the track, but my general policy is that you just can't have enough of them.  And, once you have installed all of them, you are done.   Hopefully, you never have to do that again.

Always build one step above whatever the Code" or General Rule is.  It costs about 5% more, but you will never regret it.  :-)



Mannyrock

@Mannyrock posted:

Where I have seen power issues is if I have an turnout, following by a short stretch of maybe 3 sections, followed by another turnout.  I almost always get sluggish running on that short section between the turnouts.  So, it is a bit of a pain, but I always run a set of power lines to those short sections, which solves the problem.

Mannyrock

Fastrack switches are notorious for this. The "T-bone" connector inside that connects the center rails develops a poor connection. I took all the switches at the club apart and soldered the internal center rail connections.

@Mannyrock posted:

I wonder if the gauge of the power wires running to the track have any effect on the general rule.  I just use 18 gauge wire, so I have a connection about every 6 sections.    I seem to have no problems at all with power in the track and larger engines will run at full speed around the layout.

Where I have seen power issues is if I have an turnout, following by a short stretch of maybe 3 sections, followed by another turnout.  I almost always get sluggish running on that short section between the turnouts.  So, it is a bit of a pain, but I always run a set of power lines to those short sections, which solves the problem.

It is tedious to run power wires to different parts of the track, but my general policy is that you just can't have enough of them.  And, once you have installed all of them, you are done.   Hopefully, you never have to do that again.

Always build one step above whatever the Code" or General Rule is.  It costs about 5% more, but you will never regret it.  :-)



Mannyrock

Two your point about short sections of track with switches at each end:

In my outer loop, there are two sets of crossover switches, between which is a relatively shore portion of track, and those switches and track are one single power block with a connection right in the middle.

The rest of that loop is divided into two power blocks of approx. 14 ft, each with approx. 10 joints.

Greetings all:

For your consideration, here is another interesting DCS environment, only as a bit of a foil in light of the discussion above.

I run trains off of overhead catenary, and always have.  No motors on my roster have rollers.  My layout is not huge by any stretch, maybe ~45ft or so of track on the loop, and another 30ft or so for the siding, spurs and yard, (roughly).

When I had only 1 or two motors going at once, I fed the entire system off of one TIU fixed channel at the northern-most switch on the main loop, (I know, not very good practice).  Even so, the loop's catenary is electrically isolated from that of the siding, spurs and yard.  Predictably though, when I hit 6 or 7 motors online at once, the system started misbehaving; signal strength was bad (3-7, with a few 10s), and there were an increasing number of "out of range" errors on selecting motors from the remote.

I finally found some time recently to do some re-wiring.  Now the loop catenary is on TIU Variable 1 (set as fixed, to accommodate the occasional conventional run) and the siding, spurs and yard are on TIU Variable 2, (also set as fixed).  Both feeds go through a PSX circuit breaker before connecting to their respective TIU inputs.  There is still only one electrical / DCS feed point per catenary circuit, and at the same place as before, (the northern-most switch on the loop).  Trains flawlessly transfer from one circuit to another when coming off of the loop.

Now, there are 12 motors that power-up at once; roughly 1/3 on the loop and 2/3 in the spurs and yard.  Signal strength and performance has been perfect.  Please do not misunderstand me; I do not say this to be boastful or denigrating to Barry in any way, but just for the sake of discussion.  It seems that if one had an uninterrupted run of conductor, as a good stretch of catenary wire would be, then all the resistance points caused by track connections would be removed (even the reduced resistance of a track connection that was soldered), and thus get much more trackage from a DCS signal / TIU channel, right?  (Note this discussion is only about DCS signal; channel current is not a consideration as I rarely draw more than an amp or 3 with most of the motors powered up).

I realize few will scratch-build a catenary system to run trains from, and so this post will be largely irrelevant to most of the audience.  However, I throw it out there for the sake of discussion.  Thanks.

Attachments

Images (1)
  • mceclip0
Last edited by Pantenary
@CAPPilot posted:

I always thought switches caused more DCS issues than just a single track joint, so I count switches as two track joints.  Maybe it should be three.

Might not be a bad idea to count switches as 3. In most places I used switches as block separations. Just kind of worked out that way. The diverging route is either a siding (separate block) or separation between loops of track (also separate blocks). I don't have a large layout so most of my blocks are less than what is recommended. So far I haven't yet run into anything that has caused a problem.

@rtr12 posted:

Yes, your 10 track joints per block will be fine.  According to Barry's book, track blocks should not exceed 5 to 6 sections of track on each side of the DCS/power connection to the track. That's not word for word from the book but it's very close.

It goes on to say, Each track section counts as one section regardless of it's length. In any case track blocks should not exceed 100 feet in length.

Also, it sounds like the center of the block is what Barry is saying for your block connections.

Although Barry is no longer with us, the book (DCS Companion for TIU & Remote) is still available and is well worth the cost if you are running DCS. Lots of good information in there that is not in the DCS manual. There is also a Wifi version (maybe Wifi Companion, but I forget the exact title?) that covers DCS using the Wifi App.

Edit: What Royboy & Tom Tee say about having only one feed per block is also a very important rule to follow with DCS.

Re: "one feed per block": When a jumper is needed to assure center rail power past a switch or other break or unreliable connection, this rule has to be modified.  Understanding that some quantity of added bus wires, power jumpers or drops adds measurable capacitance to the circuit, it doesn't seem to cause signal problems done only where necessary to overcome the resistance of long sections of steel rail, or iffy connections.  Thinking about DCS wavelength and avoidance of partial signal cancellation, I make sure that jumpers are close to the same length as the parallel section of center rail, and bus wires that follow the block likewise are not much different in length, with short-as-practical power drops.

My longest blocks are in the range of 60 feet, with main feeds to the center of the bus wires, and many turnouts. I use Susan Deats' capacitor-resistor "filters" at each end of each block.

Add Reply

Post
OGR Publishing, Inc., 1310 Eastside Centre Ct, Suite 6, Mountain Home, AR 72653
330-757-3020

www.ogaugerr.com
×
×
×
×
Link copied to your clipboard.
×
×