Well, I asked the VP some questions about his plans at the layout design committee meeting on Monday night and it turns out, how shall I say this, that my gut feeling was correct. When I shared what little information I had about the possible pitfalls of his plan, and mentioned that having all the bits mounted in a central location wasn't essential to achieving what he wanted, and mentioned the amount of extra wire needed, he acknowledged that maybe his idea wasn't that great afterall.
As a result of the conversation the committee agreed that we will stick with proven methodology and components. One key point I made was that, whatever we do, others working on the layout in the future have to be able to understand what we have done.
So, thank you gentlemen for answering my questions. Your advice proved to be rather valuable.
Dave
I'm just a dude with a bad back having a lot of fun with model trains, and finally building a layout!
carl425I will provide the first step in that education. Randy was talking about RR-Cirkits and I was just clarifying a misunderstanding about their "off the shelf" status.
Thanks for clarifying that Carl.
Sorry Randy. Credit where credit is due.
ROBERT PETRICKIn that other thread there was some general discussion about space allocation. Has any sort of layout design started to emerge? Even a rough outline? Theme? Era? Locale? That sort of stuff. And how are the club politics holding up? Designing by committee is difficult under the best conditions. Is any consensus forming?
Hi Robert:
We have a couple of design concepts already. I say 'concepts' because they need a lot of fine tuning, but they do meet many of the desires that club members have put forward. They need more spurs for operating purposes, and we are having a bit of a time fitting in townscapes, but those issues are not insurmountable.
The club politics are doing fine. We don't have a consensus on the design yet because the designs are still pretty rough, but people seem to like what they have seen so far. I was a bit disappointed that we didn't get feedback from more members, but about half the guys put their ideas forward so we have a fair amount to work with.
We have a design committee meeting Monday evening and this thread has already given me some material to work with. I don't think the VP will be offended if I ask him some detailed questions about his plans for the bus and Loconet.
hon30critterall the components in a central location I immediately asked myself how much wiring that would involve, and whether or not it was a good idea. Perhaps I'm not hopeless after all.
You are not hopeless. The choice between Digitrax and NCE is similar to the choice between chocolate and vanilla ice cream. BUT... one of the strongest arguments in favor of Digitrax is Loconet and its ability to effectively support a distributed implementation. In fact there users of competitive DCC systems that still use Loconet for control of other than locomotives. Take a look at the Canadian Canyons layout that MR is doing on Video Plus. This layout is barely larger than a 4x8 and they are doing a distributed Digitrax instalation.
hon30critterAlso, I will try to educate myself more about other detection options like Randy's transformer style detectors and the RR Cirkits that Carl suggested.
I will provide the first step in that education. Randy was talking about RR-Cirkits and I was just clarifying a misunderstanding about their "off the shelf" status.
I have the right to remain silent. By posting here I have given up that right and accept that anything I say can and will be used as evidence to critique me.
Hey Dave
In that other thread there was some general discussion about space allocation. Has any sort of layout design started to emerge? Even a rough outline? Theme? Era? Locale? That sort of stuff.
And how are the club politics holding up? Designing by committee is difficult under the best conditions. Is any consensus forming?
Robert
LINK to SNSR Blog
Hi guys:
Let me try to answer some of your questions.
First, the layout won't be huge. We have 500 sq. ft. to work with. Our initial plans suggest that we will have somewhere around 300' of main line if it is double tracked, and about 160' of secondary track.
To answer your specific questions:
Robert - "Does the VP know what he is talking about?" I probably didn't quote him (Mike) properly so your question is quite likely a result of my inability to quote him correctly as opposed to his lack of knowledge. I believe that Mike has a pretty solid grasp of the Digitrax system, but I don't want to just rubber stamp his suggestions blindly. Hence the reason for the thread. I will talk to him in greater detail this coming Monday. Interestingly, despite my general ignorance of Digitrax systems, when he said he wanted to have all the components in a central location I immediately asked myself how much wiring that would involve, and whether or not it was a good idea. Perhaps I'm not hopeless after all.
I will make notes about the more technical suggestions that all of you have made like not using the command station to power track, and making sure we use data type cables as opposed to phone type. Also, I will try to educate myself more about other detection options like Randy's transformer style detectors and the RR Cirkits that Carl suggested.
Thanks for now. I'm sure I will be back with a ton more questions.
Cheers all!!
carl425 ROBERT PETRICK I went with an off-the-shelf system because I don't know enough to design a customized system and select the components and/or firmware to make it work. RR-cirkits is "off the shelf" and the help and advice you get from the owner Dick Bronson alone is worth the price of the components.
ROBERT PETRICK I went with an off-the-shelf system because I don't know enough to design a customized system and select the components and/or firmware to make it work.
RR-cirkits is "off the shelf" and the help and advice you get from the owner Dick Bronson alone is worth the price of the components.
Hey Carl-
I believe you. RR-CirKits might be a better system, maybe even a much better system, but I didn't have that info at hand when I decided to go with Digitrax.
I will continue what I'm doing for now. Things may come to the point when I am not satisfied how things have turned out, and I might scrap everything and start again. By then I hope I will have learned something.
I would certainly not try to dissuade Dave from checking out as many possibilities as he can.
hon30critter,The first question is how big is the layout to be? If it's a very large (or very linear) layout, then centralizing everything won't work. The wire runs just become too long.
I'm a member of a very large club (6300 sq. ft.), and we've run into this with the BDL's. The way they work is through resistance, and believe it or not, if you connect too much wire into a BDL circuit, the detector will light all by itself even when nothing is there. Seriously. I talked with Digitrax at a show once, and they recommended not exceeding 60' of conductor on any one BDL circuit...and that includes the track. So if you're block is 20 feet long, you only have 40 feet of wire that can go there. It sounds like a lot, but when you have 16 blocks in a BDL, some of the wire runs can become quite long, especially as it winds its way through and around the benchwork.
The next issue is track voltage drop, as Martin says. I can confirm that is an issue. It's not really a DCC problem as the power is the signal (so as long as you have power, you have signal), the issue you run into is a slower trains. And then it becomes an issue when the trains run from a one voltage block to a different one and the train goes faster or slower as it transitions. This can be quite noticable.
Like Randy says, we also have found the BDL's to be trouble. We're slowly replacing ours with RR-CirKits. The RR-CirKits use a toroid, basically a wire-wrapped donut, to feed the track power bus through. This detects by amperage sensed by the toroid, rather than resistance like the BDL. This means the wire run can be basically infinite, and as long as amperage is detected, it will work. When there's no amps, there's no detection.
One thing I would highly recommend are the LocoNet Repeaters, the LNRP. Buy them and buy them now even if you only have a medium sized layout and plan on detecton and signalling. Every single thing you plug into the LocoNet draws power from it: throttles, boosters, detectors, clocks, signal drivers, radio receivers, computers, etc. The LNRP's not only boost the LocoNet's power, they also can trouble shoot the network and isolate any shorts.
My recommendation is to place the "brain" in the center of the layout and put all the LNRP's around it (one LNRP per section of the layout). Then run two LocoNet cables down each leg. One would be just for throttle jacks, clocks, and radio receivers, the other for all other devices like boosters, signals, and detection. Mark one cable with tape or buy black 6-cond. flat phone cable to make sure they're kept isolated.
As you add things to each LocoNet, check the appropriate LNRP to see if you're getting close to overloading that branch of the LocoNet. There will come a point where it will happen. At that point, either add another LNRP or remove some items from that LocoNet branch.
I would definitely install all boosters locally with the track circuit breakers and block detectors near them. I would not centrally locate them unless the layout is small-ish (~1000 sq. ft.) and the wire runs direct. Further, as Martin said, do not use the "brain" as a track booster. There can be problems with auto-reversing if using the brain's track booster next to a reversing section.
And for your sanity, follow Randy's advice and only use "DATA"-type 6-conductor phone cable. Be draconian about it. It simply must be followed or bad things will happen. And when I say that, I mean things like losing control of your train and things like that. Our layout was a mess until we went around and double checked every LocoNet connection and flipped any that were set up as "phone" vs. "data". It took a while, but reliability increased.
rrinker Much as I love my Digitrax equipment and won;t be changing any time soon - there are some things I just won;t use - one of them being BDL168's. I absolutely hate the concept of centralized wiring, ir destroys one of the benefits of Loconet in that the high current wiring can be limited in length. With everything having to flow back to centralized BDL168's, now you need 16 heavy runs of wire for each BDL-168 snaking out under the layout. Plus they cause a voltage drop themselves.
Much as I love my Digitrax equipment and won;t be changing any time soon - there are some things I just won;t use - one of them being BDL168's. I absolutely hate the concept of centralized wiring, ir destroys one of the benefits of Loconet in that the high current wiring can be limited in length. With everything having to flow back to centralized BDL168's, now you need 16 heavy runs of wire for each BDL-168 snaking out under the layout. Plus they cause a voltage drop themselves.
Hey Randy-
I kinda agree with your assessment, and I knew your thoughts on this topic because I read many of your posts over the years. I went with an off-the-shelf system because I don't know enough to design a customized system and select the components and/or firmware to make it work. Kinda like stereo systems in the seventies: some guys would have killer systems made up of components I never heard of. I had a pretty decent Pioneer setup; plain vanilla, but suitable for my needs.
rrinker . . .you need 16 heavy runs of wire for each BDL-168 snaking out under the layout . . . you have hundreds of feet of cabling connecting where you just have detection blocks across a stretch of single track line
. . .you need 16 heavy runs of wire for each BDL-168 snaking out under the layout
. . . you have hundreds of feet of cabling connecting where you just have detection blocks across a stretch of single track line
Yes, these two sentences are very true and can be a real pain in the buttocks, especially the snaking out business. I also have some busses that are 15 or 20 feet long that serve a detection block that is maybe 3 feet long with a single pair of 22 awg feeders. But, I just happen to have several 500-foot spools of 12 awg and 14 awg THHN single conductor wire on hand. Adding dozens of sticky tape labels to each and every buss was more of a pain than having all that copper wire all over the place. And even though it was 'free', it is still kinda hard to justify that it wasn't a waste.
My layout is still under construction and I am still feeling my way through the dark. I can't give Dave specific recommendations, I can only relate to him how I am proceeding with my meager talents and resources.
Instead, I prefer the transformer type detection, Team Digital and RR-CirKits have Loconet compatible ones. In the case of the RR-CirKits units, the transformers are mounted close to the sub bus they are detecting and feed back to the controller via twisted pair network cable - common Ethernet cabling. Plus they are in groups of 8 - ganged enough for more complex interlockings but not so many that you have hundreds of feet of cabling connecting where you just have detection blocks across a stretch of single track line. And no voltage drop, so if you have undetected areas, it won't matter.
As for the Loconet wiring, it can be bus or star, or a combinatioon. Those ARE 'technical' terms but they mean eactly what the words mean otherwise, as well. The only thing you CAN'T do it loop the Loconet back on itself. You CAN connect the command station Loconet to a 5 way splitter and then run 5 lines out around the layout. This can reduce some of the wiring. The key is a proper splitter - Loconet uses straight through cabling, phones use twisted, which reverses the order of the wires. Loconet actually works, with the exception of devices that use the Railsync signal - this is a low power copy of the DCC track bus. This signal is what is fed into boosters, and some devices like the BDL-168 also use the signal. If the wires get reversed, the signal becomes inverted. What happens this is that what is the Rail A output on one booster becomes the Rail B output of the booster on the other side of the twisted cable. AKA dead short when crossing the rail gaps between the two sections. So it is important to always use DATA cabling, not PHONE. RR-CirKits has proper DATA wired splitters to do the star wiring.
--Randy
Modeling the Reading Railroad in the 1950's
Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.
hon30critter This is related to my club's soon to be built new permanent layout. Our club's VP and current acting President wants to have all of our Digitrax layout components in one area of the layout. The primary reason is that he wants to make use of Digitrax's ability to control signals so we can have more or less prototypical operations on the layout. That sounds like a good idea to me but I have a real lack of understanding of how the Digitrax Loconet system works. The VP is suggesting that, instead of a continuous bus, we wire the layout with five (or so) separate bus districts with all of the feeds (buses) originating from one central location. The VP is very tech savy so I'm not questioning his decisions but I would like your help in understanding them better. I'll openly admit that I have a hard time understanding the more advanced concepts in DCC. I'm a computer dinosaur (and proud of it!). Please tell me what the things are that we need to be aware of so I can contribute to the discussion, or at least ask intelligent questions. Is this a good idea? Are there potential complications to having several separate buses? Thanks as always, Dave
This is related to my club's soon to be built new permanent layout.
Our club's VP and current acting President wants to have all of our Digitrax layout components in one area of the layout. The primary reason is that he wants to make use of Digitrax's ability to control signals so we can have more or less prototypical operations on the layout. That sounds like a good idea to me but I have a real lack of understanding of how the Digitrax Loconet system works. The VP is suggesting that, instead of a continuous bus, we wire the layout with five (or so) separate bus districts with all of the feeds (buses) originating from one central location. The VP is very tech savy so I'm not questioning his decisions but I would like your help in understanding them better.
I'll openly admit that I have a hard time understanding the more advanced concepts in DCC. I'm a computer dinosaur (and proud of it!). Please tell me what the things are that we need to be aware of so I can contribute to the discussion, or at least ask intelligent questions. Is this a good idea? Are there potential complications to having several separate buses?
Thanks as always,
What does the VP mean when he says he wants to make use of Digitrax's ability to control signals for prototypical operation? Does he know what he's talking about?
I use three SE8Cs to control signals (and Tortoises) and one BDL168 and one BD4 for block detection (16 + 4 = 20 blocks). The SE8Cs are disbursed around the layout to be near the actual signals they control (connected by pretty expensive 10 wire flat ribbon), and the block detectors are adjacent to the centrally located command station. Twenty detected blocks means 20 busses (long explanation about that). Everything connected with 6 wire telephone cable (The Loconet).
Research Digitrax BDL168 and SE8C. That'll get you started. You have a lot of homework ahead. Don't get discouraged. Good luck.
You need to be aware of voltage drop. I actually do not recommend "central" booster set ups. That's because there will be excessive voltage drop in the districts ffarthest away from their boosters. This causes unequal track voltages and possible signal degradation. You will need larger wire to overcome this. Much larger.
Loconet can run a couple of hundred feet. The one and only one command station will be handling all signal packets. Everything else is just slaved to it. So place the Command station in a central convenient location. Place the other boosters as close to the center of the power districts they will be running.
BTW, I also don't suggest actually using the command station to power track on large layouts. Martin Myers