Electrical Blocks - this can't be right?

Each TIU channel can support 250-300 feet of track broken into as many as 12-14 blocks on each channel, where each block can be as long as 11-12 sections of track. With 4 TIU channels providing support for 14 blocks each, you could have 56 blocks of 11-33 feet each.

Your layout needs:

• (2) mainlines with 8 blocks each = 16 blocks
• (6) classification yard sidings with 1 block each = 6 blocks
• (3) staging tracks with 1 block each = 3 blocks
• (3) Els with 3 blocks each = 9 blocks
• (3) sidings with (maybe) 2 blocks each = 6 blocks

That's a total of 40 blocks, well within the capacity of one TIU.

Switch tracks and accessories don't need the DCS signal and should not pass through a TIU channel.

I don't know what is the "MTH DCS supplement book instructions" to which you refer, however, a much better guide would be (no false modesty here!) The DCS O Gauge Companion. According to the feedback I get (see some of the recent threads on this forum), your DCS experience would be much improved if you purchase a  copy of the book.

Everything you want to know and a whole lot more is all in "The DCS O Gauge Companion 2nd Edition", now available for purchase as an eBook or a printed book from MTH's web store site! Click on the link below to go to MTH's web page for the book!

One TIU should do it.  I disagree with Russell, in that I think a 50' block is too long for good electrical distribution.  I agree that sidings should be able to be powered off.  I use toggle awitches on a control panel, but all tracks are fed through the control panel since my layout was built long before DCS came out.

Hi Barry,

Yes, I was referring to your book - that's what I've been studying!  So grateful for the book and happy to know you're on this forum.  

Barry - I think I used the wrong terminology.  I think I meant to say "power district" not block.  Between your book and the OGR DVD for DCS, I thought the idea was to spread power demands among different transformers in cases where one transformer would be inadequate.  In the OGR DVD example, running 2 passenger trains on one loop could require more power than available from one transformer.  So they propose breaking that loop into 3 power districts, all powered by separate power sources.  From this, I assumed each of our 2 main lines would need more than one power district, as we like running passenger trains.  If this is correct (??), then based on your book, I would need a different TIU channel for each power district.  And, with each loop requiring at least 2 power districts, I will have used up at least all 4 channels on one TIU with the two main line loops.  Similarly, the EL part of the layout will run passenger trains and, with 3 tracks, can accommodate 3 passenger trains.  Based on the power demands, I assumed this too would require more than one power district powered by more than one power source.  Finally, I thought switches and accessories would run through the DCS if I want to control them remotely via the DCS handheld? 

Also - any insight on how to star wire when dropping a line from each section of track would be very appreciated too.

Again, I'm a novice, so my apologies if I'm asking a lot of stupid questions. 

Can't thank you all enough for your help! 

Peter 

Peter,

Between your book and the OGR DVD for DCS, I thought the idea was to spread power demands among different transformers in cases where one transformer would be inadequate.   In the OGR DVD example, running 2 passenger trains on one loop could require more power than available from one transformer.  So they propose breaking that loop into 3 power districts, all powered by separate power sources.  From this, I assumed each of our 2 main lines would need more than one power district, as we like running passenger trains.

Yes, however, a TIU channel can handle 10 amps of continuous current draw, and a PS2 engine, with lights and smoke only draws about 2 amps. Therefore, one TIU channel can operate at least 4 PS2 engines, with power left over for lighted passenger cars. You can use 1 TIU channel for each of your 2 mainlines.

Similarly, the EL part of the layout will run passenger trains and, with 3 tracks, can accommodate 3 passenger trains.  Based on the power demands, I assumed this too would require more than one power district powered by more than one power source.

The three tracks of the EL can be served by 1 TIU channel. The remaining TIU channels can handle all of your various yards and sidings.

Finally, I thought switches and accessories would run through the DCS if I want to control them remotely via the DCS handheld?

You control switches and accessories via AIUs that are connected to TIUs, not TIUs themselves. Power for the accessories and switch tracks would come from separate transformers, without passing through the TIU at all.

Also - any insight on how to star wire when dropping a line from each section of track would be very appreciated too.

I'm not certain of what you mean, however, how to wire for DCS is covered in Part III of The DCS O Gauge Companion in great detail.

Peter, thank you for posting this, I have been formulating the same question.   From the OGR DCS video and/or Barry's book, I learned that you should have one more power district per loop than you plan to run trains on that loop: two trains per loop would mean three power districts per loop.  Each power district star wired with a power feed to each block of 10 track sections, etc.  This is to provide adequate amperage to the loop for the train load.   But what this means is that you end up having to have a lot more transformers and a lot more TIU's than Barry's formula in this thread would indicate?  

Three power districts per loop means using three of the TIU channels for just one loop, yes? And that means three transformer handles for that one loop also, no?

I don't want to take over Peter's thread, but I think we're talking about the same thing, and I'm also a beginner trying to grok this whole DCS schtick.  Barry, I hear what you're saying, and I guess I'm missing something here, but when I take your diagram on page 64 and apply it to my layout, I still run out of capacity before I can make the diagram work.  

My layout has 4 loops.   Wanting to run 2 trains on each loop, each loop therefore has 3 power districts.  (one more power district than trains).  Each power district has 10 power feed blocks (each power feed block consists of ten sections of track).  The 10 feeds pretty much maxes out the terminal block, and it isn't recommended to make more than about 10 feeds per power district anyway, am I right?  

So that means 3 x 4 is 12 power districts.   I think both Peter and I are both assuming that one power district means one TIU channel.  If this is the case, then I need 12 TIU channels, or 3 TIU's.   Is this not the case?  Can you feed two power districts from one TIU channel?   If so, haven't you just made one bigger power district?  And I can't do that anyway, because I've maxed out my terminal block.

I think we are also sassuming one transformer handle for each TIU channel, and your post above seems to say that is not necessary: that we can power more than one TIU channel with one transformer handle.  But if that's the case, then we'll be breaking the rule of one more power district than trains running on the loop.   We may as well just wire each loop as one power district, except that again you run out of terminal block space.

Like Peter says, this can't be right.

My layout is standard gauge, so i think the amperage requirements may be a little higher?

I guess I'm missing something here, but when I take your diagram on page 64 and apply it to my layout, I still run out of capacity before I can make the diagram work.

 

Like Peter says, this can't be right.

That's correct - it isn't right.

You only need to use power districts when you need more than 10 amps per loop. Since you can easily run 2-3 trains (even passenger trains) using 10 amps, you generally don't need to implement power districts.

The diagram to which you refer is not mandatory. It's there to show how to implement power districts - but only if you have to do so.

Further, power districts are not a DCS thing. They were "invented" long before DCS as a straightforward, generic solution to how to run more trains on a loop than one transformer can handle. They are used less frequently at present because while pre- and postwar engines required a lot of power, modern engines use considerably less.

Ah, the proverbial light bulb goes off.  This explains (almost) everything.  Thank you sir.

"...while pre- and postwar engines required a lot of power,..."  So on my tinplate standard gauge, i just need to take a close look at amperage needs, and tailor accordingly: maybe one transformer handle is fine running two MTH proto 2 standard gauge trains on one loop.  And the pre-war conventional trains, you're only going to run one train per loop anyway, using z4tracks.

It seems to me that we still run into the limit of 10 track sections per feed and 12 feeds per terminal block (page 62).  On my layout, this amounts to something more like 150 ft of track, far shy of the 250 max per TIU channel.(p. 62 again)  If you are not supposed to daisychain terminal blocks (also p. 62) then how do you physically add more track to a TIU channel?  (Each on my loops is about 200+ ft of track - standard gauge tinplate tubular).

I haven't had so much fun learning something new since...

Cheers.

Hi,  

I have a 10 X 20 foot home layout with 4 loops.  I run it with one TIU.  A Z-4000 powers two channels.  The other two channels are powered by a MRC Dual 270.  I run a lot of passenger trains.  I have found no problems with running 6 trains at the same time.  

My biggest limitation is lack of track space.  I think you will run out of track space and train operators before you reach your power limit.  I suggest that you start with one TIU and see how it works out.  You can always add more transformers and TIUs if they are needed.

I am in the process of replacing the lights in my Williams passenger cars with LEDs.  I have two 6 car Williams trains that drew about 3 amps each.  The trains draw less than 1 amp each with LEDs.  My MTH, Weaver and Lionel passenger cars do not draw as much power as my Williams cars.

Joe

Thank you Joe for the reality check.  good points.

But how do I physically connect 20+ power feeds to one TIU channel, given the recommended maximums on p. 62?

standard gauge layout, 40 x 30 - it's in what was the hayloft of a barn.  (standard gauge: think big.)  I think i'm going to need more TIU's and more transformer handles not because of the needed amperage, but just to cover the track within the recommended guidelines for DCS wiring?)

It seems to me that we still run into the limit of 10 track sections per feed and 12 feeds per terminal block (page 62).  On my layout, this amounts to something more like 150 ft of track, far shy of the 250 max per TIU channel.

Note that optimal length track blocks can be up to 11 track sections long. Using Gargraves 33" track sections, that's up to 27.5' per block. Ten blocks would be 275' per channel, or 1100' per TIU.

Also, 10-12 blocks per channel is not necessarily the limit. With strong track-to-track connections, you could possibly get more blocks per channel. 

hojack - I totally appreciate you commenting.  It only enhances the discussion, so thank you.

Barry - Thank you for the advice, but I am now thoroughly confused (likely because I'm missing something).  I hear what you're saying, but it seems "Rich" in the OGR DVD has a different view on power requirements for trains and the difference in opinions has left me scratching my head.  According to him, a passenger car uses 0.2-0.4 amps, so a 10 car passenger train (without the engine) requires 2-4 amps.  And, he says a two-motored diesel uses another 3.5-4 amps.  Added together, this is around 7 amps for one train.  Two similar trains would add up to 14 amps on one track.  

Fundamentally, I'm trying to determine how to wire for the most trouble-free operation so my boys stay interested and have fun.  From what I've read, I should drop a feed wire from every track section and I would like to do this.  With my track length, that's about 50-55 track sections per loop.  My understanding (from your book and the OGR DVD) is that each track block simply gets wired to a centrally located terminal block.  Perhaps I'm mistaken, but it seems your book and the OGR DVD assume that all the track sections in a given track block will be powered by one set of wires (rather than each track section in a block having its own set of wires). 

MTH terminal bocks allow for either 12 or 24 connections (depending on which one you buy).  What I'm not appreciating is how to create track blocks (for star wiring) when wiring each track section?  Again, I'm a novice, but it seems to me that either: (a) all those track section wire pairs are going to go to a terminal block (so I'd need at least 4 terminal blocks to accommodate 100 track section wire pairs (not counting the EL); or (b) all those wire pairs are going to connect into another pair of wires that make up a track block (akin to buss wiring), and that one larger wire pair per track block then connects to the terminal block. 

Again, thank you.  I know this is just basic stuff needed for any layout, but trying to undertake this as a first-timer is quite challenging, so people willing to share their experience and expertise is incredibly valuable.

Peter

 And, he says a two-motored diesel uses another 3.5-4 amps

Maybe so, if it's a Lionel postwar engine. An MTH PS2 diesel (2 motors) uses approx. 2+ amps.

From what I've read, I should drop a feed wire from every track section and I would like to do this.

That would be wrong. You didn't read that in any book elated to DCS, did you?

it seems your book and the OGR DVD assume that all the track sections in a given track block will be powered by one set of wires (rather than each track section in a block having its own set of wires). 

That would be correct.

"That would be wrong" doesn't provide any understanding as to why this is your opinion.  Could you elaborate?

No, I cannot. I have no more to say on the subject.

Wiring every piece of track is what's recommended by the absolute overwhelming majority, hence, the reason why, as a first-timer, I assumed this should be how I proceed

Do as you like. If you make each track a block, it will work, however, you'll most likely triple your TIU expense for no additional gain. I now have no more to say on this subject, either. Your group of non-DCS "experts" is focussed on severely overkilling a voltage drop issue that's most likely going to be nonexistent on your layout. One foot blocks? Really??

"That would be correct."  Again, this offers no undderstanding of why this is your opinion.  Could you please elaborate?

Nope.

What I guess I'm gathering here is that there would be no ill effects from over-designing or over-powering my setup.  The only thing I can think of is that more TIU's running in Super mode adds a complication and room for things to go wrong that could be avoided by using just one TIU.  

The only "ill effects", as I stated above in this post, is no appreciable gain in return for tripling the TIU expense. 

OK, thanks for the reply.  

Wasn’t suggesting each track section making its own block.  I was asking something different – in essence, if there are (say) 10 track sections in a given track block and I decide to do a power drop from each of the 10 sections to form one track block, if I were to use MTH terminal blocks and star wiring, it seems my layout would require multiple terminal blocks and TIUS (based on your book’s one terminal block per TIU channel).  This resulted in my surprise and creating this thread.  Thanks to comments from hojack, Joe Barker (and others) and your earlier comments, it’s now clear that I don't need to wire every track section, so the extra hardware is not necessary.

Keep in-mind that we’re just getting into the hobby so we’re looking at the available data obectively.  Deciding factors in where we give our continued business include ease-of-use and reliability – especially since this is a father-son project with young children.  Deciding to continue with MTH products over Lionel (and its almost universally accepted TMCC/Legacy system that requires simple buss wiring), requires getting to a comfort level with the additional MTH wiring requirements.  While we’d like to continue buying MTH trains, without a comfort level for its DCS system/wiring needs, this would not be an option, given MTH trains cannot be run in command mode using TMCC/Legacy.  Thanks for helping us get closer to attaining that comfort level.

Barry - if my questions have, for whatever reason, annoyed you – my apologies.  Again, I’m very much a novice just trying to make sense of all the data and am probably asking a lot of stupid questions.  In any event, thanks so much for the earlier posts (and your quick replies!) – they were quite helpful. 

Best,

Peter

 PJB, I have always used power drops at every 4th track joint, even that is overkill, no problems ever. For what it is worth you do sound snotty to Barry, and he is the man for DCS information. Just sayin.

Really?  Thanks for the feedback, but that was not at all my intention.  Simply asking for an elaboration on "that's wrong" so I could better understand where I went wrong.  I am actually thrilled that he is available and willing to help a novice like me with valuable advice.  In fact, as soon as he decided to offer help, I was thinking "awesome!"   

Yikes

PJB;

You are trying to mix very old and new technology here.

Very old track was poor in conductivity and for a layout that stayed up was usually in a museum or such place. They needed Max reliability in the power to the rails.

New Track is better in many ways. It no longer needs a power connection at every section.

DCS also has a problem with that kind of wiring redundancy.

Think of DCS as sending runners down the wires, when it hits the terminal block, the runner gets cloned onto each wire to the track.

When it hits the track they clone again and one runs each way.

If you have more than one power connection per block, they eventually pass one another.

Now, this is a digital signal, Think of morse code only VERY fast.

When they cross, they interfere with each other. If an engine is at the point they are crossing it just hears noise.

To make it worse, the TIU will repeat the signal several times until it gets a response or times out. Now you have multiple copies of varying age running past each other all over the track.

Furthermore, the signal can bounce (at lower strength) when it hits the end of the block. This creates even more signals crossing, but since the strength fades as it bounces it's not as bad.

Thus the Rule, One power drop per block.

And you may have heard of the "Magic Light Bulbs" or Engineered filters. These are to kill the bounce in the signal. And thus they help get a stronger signal reported.

I suspect the signal is more of a Number of signals sent compared to number received and responded to. Scaled down to 1-10.

Hope this cleared up some confusion without making too much more.

Also,

Barry has explained things so many times he made a book, believe the book.

He has more DCS experience than the next ten people combined.

  I have a simple question or two on a power district.  My understanding for wiring one loop with 3 blocks, then 3 Bricks are used 1 for each block plus 1 TIU channel ( TIU's have 4 channels).  Are the bricks or transformers, 2 wires + and - connected just to the tracks or are they conected to the TIU?  If connected to the TIU, I assume its to the output side.  With all this power in one district isn"t there a feedback problem?

Dr Jack, a "power district," often called a "loop," is by definition the area controlled by one transformer handle's output.  Many blocks can be on one handle.  I have 6 "power districts" with over 70 blocks.  The bricks are used as the equivalent of transformer handle, and feed into INPUT.  The OUTPUT is used to feed the blocks in that district.

The above posts give DCS a complexity it doesn't have.  My layout preceded DCS by a decade, using a common bus for the outside rails, and with each handle feeding a section of the control panel, where the wire jumped from toggle to toggle.  Every bit of the layouy is blocked & toggle-switched, since I wanted to run several conventional locos in each district.  When DCS came out, I plopped the TIUs betwen transformers & control panel & have been happily running since.  I did have to add a few bulbs to ends of some of the sidings, using a technique I described elsewhere.

Russell - OK, thanks.  This is making a lot more sense to me now.  I forgot that DCS is different than TMCC/Legacy in that it uses a digital communication signal through the center rail.  Thanks for the help. 

I've read about the magic lightbulb in Barry's book.  I also saw something on another forum about the filter in lieu of the lightbulb.  Another novice question - any idea on where would I get a magic lightbulb or filter (assume it also comes with installation instructions)?

Thanks again.   

Look on the bright side Barry, by the time you get done with the likes of us you will have achieved total Zen enlightenment.  

Is there one of those little smiley emoticon/graemlin things for "slap upside the head"?  Barry could just drop one of those in once in a while.

I'm one of the worst, and I do apologize.  I'll get this thing up and running in spite of myself.

Buy the book.  Read the book. Do what the book says.

PJB:  Some posters who have used the Rev L TIU say the use of a bulb degrades the signal from a Rev L.  Not having one, I can't say.

I don't want to belabor this, but someone on the electrical forum just made a comment, about something else, that has meaning here:

"you are correct in that some do seem to forget that not everybody can understand the answer provided but the reverse is also true; sometimes the person asking the question doesn't have the ability to understand the answer. "Any technology suficiently advanced will appear as magic to those unknowledgeable.""

and here on the DCS forum, a little while ago someone mentioned tht they would be taking their TIU to York to have it updated: there was a reply to the effect of, why don't you just do it yourself, it takes 5 minutes, what's the big deal; and the first person answered, "Fear."

To those of you who have the background and training and experience to understand how DCS works, the answers to our questions are obvious.  But to those of us who do not understand how it works, it is a little like magic.  The best we can do is to try to follow the directions as carefully as we can and hope it works.  We are not trying to be difficult.  We are just trying to figure out the directions.

For example, it is obvious to those of you who really know DCS, what is good information and what is unnecessary of even incorrect information - and, for that matter, what is so obvious that it doesn't need to be said.  But there's no way the rest of us can make that distinction.  Barry's book is the best thing out there, but there are still things that the book doesn't say, questions it doesn't answer (it would be impossible to answer all questions).  And, Barry's book and the OGR DCS DVD and the MTH DCS booklet do not always appear to say the same thing.  This information is obvious to those of you who understand DCS backwards and forwards, but the rest of us need to ask the question so that we don't inadvertently wire things up wrong. Fear.

Having said that, I also understand that those like myself that are learning, need to do our homework, and not expect you to do it for us.  There is such a thing as a dumb question. We're trying.  But it's kinda magical, and you guys are the magicians.  Thank you for all your help, and for putting up with what appears at times to be idiocy.  You are good teachers.  For some of us it is a very vertical learning curve, and we're just a little out of our depth.

Well stated, Hojack

To those of you who have the background and training and experience to understand how DCS works, the answers to our questions are obvious.  But to those of us who do not understand how it works, it is a little like magic.  The best we can do is to try to follow the directions as carefully as we can and hope it works.  We are not trying to be difficult.  We are just trying to figure out the directions.

Barry's book is the best thing out there, but there are still things that the book doesn't say, questions it doesn't answer (it would be impossible to answer all questions).  And, Barry's book and the OGR DCS DVD and the MTH DCS booklet do not always appear to say the same thing.  This information is obvious to those of you who understand DCS backwards and forwards, but the rest of us need to ask the question so that we don't inadvertently wire things up wrong. Fear.

Hojack – great post!  I think it crystallizes a very imporant point.  I would add that I think some people may lose sight of the fact that the very reason some of us are on these forums is because we are novices seeking helpful advice from those with experience or expertise.  For me, I came into this forum after spending the last 5 months compulsively researching everything from train manufacturers and command control systems to switch machines.  With regard to DCS, there isn’t a whole lot out there.  There are 3 major data sources – the DCS manual, the OGR DVD and Barry’s book.  None are complete and, as you say, they sometimes differ from each other.   I was, therefore, elated when Barry offered valuable assistance earlier in this thread.  

I’m now a little gun-shy in posting here, however, as some have hinted that asking for elaboration on some points was poorly received.  Personally, I’d prefer to ask the question and understand rather than mindlessly follow.  And, I can’t imagine that anyone who would voluntarily provide "expert" help would then get bent out of shape because there are follow-up questions.   

As a follow-up to my earlier questions about DCS, I have yet another novice question  – I’ve read that, for proper DCS function, every switch should start/end a track block so the digital signal doesn’t get garbled (this seems to make sense).  First, I was hoping for confirmation on whether this is true.  If true, two switches would be needed to connect a double mainline (which is what we are planning) – would the track block separation point be at the meeting point of the two switches?  Also, what would the track block scenario look like for a switching yard?  There would be the main line going into the yard, the classification track, the run-around track and the actual yard tracks.  This would mean several switches all within inches of each other – so does that mean several track blocks??  What if any special track blocking considerations are recommended here for proper DCS functionality?  

Thanks again to everyone for all the help.      

I’ve read that, for proper DCS function, every switch should start/end a track block so the digital signal doesn’t get garbled (this seems to make sense).  First, I was hoping for confirmation on whether this is true.

It's not true at all and if you read that, it's not something I wrote.

My statements, both in the book and on the forum, are that it's convenient to start and end blocks at switch tracks simply because it makes it easier to evasion a block by having a switch at each end and also because many switch tracks (Gargraves and Ross unwired switch tracks come to mind) come with a natural break in the center rail so that each of the three legs are already isolate from each other.

There's absolutely no benefit to the DCS signal by starting or ending a block at a switch.

would you agree with Russell's point that I've quoted above?

I would agree with the general description and, if it makes sense to you, that's fine.

If so, and I have 2 main lines connected by switches (with trains running in opposite directions and receiving signals through two different TIU channels), wouldn't this mean I have DCS commands running into each other from the two main lines at the switch point?

No, why would it?

As long as there's only one wire pair from the TIU to each isolated (center rail) block, the engine only sees each command once. In that case, the only time that the engine can see more than one occurrence of a command is when the engine's (or a car's) pickup rollers bridge two blocks at the exact same instant that a command is issued.

And, with your advice that I need not create a new block at every switch, would this also mean that each main line should have a much shorter track block wherever there is a switch conection in order to avoid a super long track block (for instance, if each main line track block contains 12 track sections, then I would be creating a 24 track section track block if two 12 track section blocks are connected by switch)?

That statement gives me a headache.

Just follow the guidelines for creating blocks as described in the book and stop over-thinking things.

OK, thanks. 

Appreciate the help!

Posted this in my other thread, but realized it's probably better to post here - 

I was watching the OGR DVD for creating a programming track.  According to Jim Barrett, a DPDT switch connected from track to terminal block is how to do it.  There are others, however, that say NOT to do this, as it lengthens wire runs.  Instead, many people say that the DPDT switch should be placed "close to the action," and then to use a close-at-hand relay (or other remote device) to control the DPDT switch.  Even Barry's book suggests to complicate the "perfect set-up" (track-to-terminal block-to-TIU) as little as possible.  Is there a preferred method?  Many thanks.

Just put the switch in the circuit where Jim says it should be placed. Even an extra 30-60 feet of wire won't affect the ability to use the track for DCS programming purposes.

If you're still concerned about even the slightest loss of power (the DCS signal won't be affected by the extra wire length), use 14 gauge wire.

There is no general agreement that relays should be used with toggles.  I don't and have no problems.  To me, relays introduce added complexity and something to fail.

If you go the relay route, be sure to get relays with contact capability of at least 20 amps, since they must handle the load of a short circuit, including breaking that load.

RJR - are you saying you run DCS and use toggle switches?  If so, what is the max distance between the TIU and toggle (distance and then also considering the extra wiring needed for the toggle)?  Thanks. 

PJB, my layout was originally designed in 1992 for conventional operation.  It uses 6 TIU channels fed by 3 transformers (1 Z4000 w/ receiver feeding 2 FIXED & 2 ancient ZWs feeding 4 VARABLES).  The longest "loop" is 0.8 scale miles long. Every bit of track on the layout is fed through a toggle switch; there are some 70 such blocks.  Each of the 40+ track switches marks the end of a block and longer stretches are split into multiple blocks. 

There is a control panel with 6 segments, where all toggle switches are located.  Initially, each transformer output went to a segment of the control panel, where it hopscotched from toggle to toggle.  When DCS first came out, my son gave me a TIU & remote (I bought another TIU).  I simply plopped these into the existing transformer>control panel leads:  transformer feeds into red input, and the lead that ran from transformer to panel now runs from red output to panel.  From each toggle switch, a 14-gauge wire runs to the controlled block, with runs up to 30 ft or so.  The wiring behind the control panel was not altered.

The toggle switches I ordered from Allied; don't recall the brand.

From transformer to TIU longest run is about 3 ft, and from TIU to panel longest run is about 4 ft.

Ground for the layout is 12 gauge uninsulated bus that loops from the U terminals of all transformers around the layout and comes back to the U terminals.  Ground is not fed through the TIUs, but all black outputs of the TIUs are connected to the bus (nothing connected to black inputs).  I did this because I didn't want the power from accessories, which are also grounded into the bus, flowing through the TIU because that might overload the internal circuit traces.

Some sidings had to be tweaked with 18-volt bulbs to get a good signal from my old Rev G TIUs, but signal strength is not a problem.

The use of these toggled blocks has many advantages:  (1) it permitted the tweaking operation to ascertain which sidings needed 18-volt bulbs to avoid diminution of signal on the mainline (yes, you read it correctly, the bulbs affect blocks other than those to which they are connected); (2) it has aided in locating shorts, such as a derailment (which train on the "loop" is the culprit?) or (in one case) an undertable short caused by a filing cabinet; (3) if there is a derailment, I needn't shut down the whole "loop" but can keep running until I'm ready to walk around the layout; (4) there is no need for a separate programming track as any convenient block can serve that purpose with all other blocks turned off; (5) trains can be stopped anywhere without fear of stealth movement.

I strongly believe that the secret to making DCS work well is adequate gauge wiring.  At risk of starting a verbal battle on the forum, I will state that I am not a proponent of the "star system" & do not beleive it is feasible if you also want to run many trains under conventional control ( have many).

The thing is, every layout is different and there are just too many variables to make hard and fast rules.  I'm learning you have to be willing and able to improvise to make your situation work.  I think this is some of Barry's frustration at times, is that some of us amateurs don't know enough about the principles involved to be able to improvise, so we're looking for someone to tell us what works and what doesn't.  The answer is maybe.  Or, the answer is, try it and see.

I have a standard gauge tinplate layout, so I wanted to use lionel's 439 and 440C switch panels, which essentially are just a bank of 6 old fashioned brass knife switches. These are used instead of the more modern toggle switches, and look more in place on the tinplate layout.  I asked on this forum if anyone knew whether these old type switches could be used with DCS, and nobody seemed to know.  So, I did it myself and I'm reporting back.  They work fine, with no apparent DCS signal degradation.  Granted, most of them are only feeding sidings, or fairly short track blocks, so there is no great mileage of track downstream from any knife switch, but that's what these old knife switches were intended for.

I was also worried about the standard gauge trains pulling more amperage, since I was pretty sure Barry's comments on not using power districts were based on O gauge equipment: that was an unknown going into this, and I couldn't get any information on that.  I was running my super 381 state set again today and it drew 3.4 amps. It's worth asking, in case somebody knows, but it's kinda cool that any of us are as much a part of this process as anyone and may have something to contribute at the end of the day.

I agree with rjr on the adequate gauge wiring.  That's one area where I stuck to my original plan and "over built" it a bit.  I'm very happy I did this: I think it makes the whole system more robust and forgiving of other irregularities (like 18 feeds off a 12-terminal block...).

the layout still has a ways to go, but i'll be getting some photos out here eventually.  thanks again

The thing is, every layout is different and there are just too many variables to make hard and fast rules.