FRUSTRATE, FRUSTRATE, FRUSTRATE! Hey Mike, Robert and Ben, and of course all tuned in. You guys are beginning to frustrate me...grrrrr.laugh: No, not at all. I wired one of my early layouts to the drawing Robert, provided only using the Atlas sliders and the one control that has the common marked on it. Honestly, it doesn't work. After two attempts, an electricians help and advisement (Get rid of it) a model railroad mentor who attempted the same thing...more times then I...we pulled it out and installed DPDT...NO PROBLEMS! PERIOD! Going into each transformer and finding the common for both and connecting them comes closer to accomplishing the task. Installing filters will help. The problem as I see it with Atlas is if you use the transformer's reversing switch you change out the common. And, the results I have a here fore identified...that symptomatic behavior, already described and identified. I've made my point here and I wasn't theorizing I was reporting facts. Old Ben seems to think I'm a screw up when it comes to wiring. One thing I know is wiring. Where I goofed up was using Atlas's sliders and devices in an attempt to make it work. It's DCC I don't know any thing about...YET! Look out, I will master it and then who needs the frickin common wire/rail. On the home layouts I have visited or was called in to trouble shoot, I found they too used the Atlas slider system. Here I found the same symptomatic operating problems. I've already described this action here in post's 2 or 3 times. You can go back and give it a read. Right about now I'm wondering if anyone gave it a read...grin! As recent as last year we tore out the Atlas system on a layout and put in DPDT...for reasons already stated. Why? Because it was working and working well? No, because it...and you should be able to finish this. No problems now! Theorize away...I really don't mind...but what I keep reading is "It should work". Yep, I kept hearing that from the Atlas rep's...as well. I've yet to see it work well. Just my two cents and me thinks I've labored this point sufficiently. Now go build a model, using the Atlas sliders and send me a video of it working. And, I don't want to see trains running in the same direction. I want to see them running in opposite directions....@#$%&*!... without the symptomatic problems already identified here. So, what's the secret? I don't want to hear "It should work". There's your challenge so go get'r done. err:mad:eek:
Yup, it "should" work - provided you don't overrun a block! If trains in adjacent blocks are travelling in the same direction, the polarity of the two blocks (if not the actual voltage) will be the same. You will see an effect on the speed of both trains when the two power sources combine as a loco bridges the gap between the blocks (as would be expected, no new news here). The problem comes when trains are travelling in opposite directions (presumably toward each other). The polarity of the adjacent blocks are in opposition, and as the gap is bridged, the voltages of the two blocks will buck each other giving you some very wierd results. I believe this was illustrated in an earlier diagram in the thread. Bottom line, if you NEVER overrun a block, you will have no trouble (and we never do, right? <VBG>), but given that "descretion is the better part of valor", direct home/bi-wire provides more flexibility and just makes more sense. It also makes the eventual DCC conversion easier.
Well, you'll get weird bucking if you run into a block with opposite polarity, whether you've used common rail wiring or direct home wiring. Maybe if you want to run both DC and DCC on the same layout. If you're converting to 100% DCC, then not really. Rick, it's not that I think you're a screw up, it's just that ... well, you haven't given any specific details that serve to show that another explanation is more likely! You haven't told us what type of throttles or what gauge wire was used on any of those layouts. People tuning in to this conversation see you saying "it never works" and me saying "always worked for me" and they've got no way to tell who's more credible. All I can really say is that if it works for me, then there must be a reason that it doesn't work for other people, a reason that we haven't really identified yet. And the reason is not that the concept is fundamentally flawed, because it's not. Not if it works %100 reliably on our humongous club layout. To repeat what I said earlier (although I didn't do this on my first layout), I would recommend for common rail wiring that you absolutely gap both rails at block boundaries and wire the common rails to a bus. This will ensure that electricity searching for the shortest route finds its way back to the common bus before it finds its way to some other block. Combined with an adequate size bus (I believe the club uses 12awg), this should ensure you don't have the problem Robert described. (The other thing about the club wiring is that most of the blocks have two long connections to a more central wiring location where they are connected to the common, and that probably helps to discourage electricity from taking a very long high resistance route through two blocks.) Failure to gap the common rail at block boundaries is the most likely explanation I can think of for the problems you describe Rick. And I'd agree that some instructions probably say that you don't have to do that, so people aren't necessarily idiots if they don't. (They may still be idiots for other reasons. )
Well it certainly sounds like something in the way Atlas implement common-rail. If it's not the equipment it must be the instructions. Or maybe all Atlas customers are numpties
did the test of passing trains just as barstow rick said. yup, they didn't do it magerly but the stopped train struggled to get up to speed and the other one that was already moving by watching the coupler slack we saw the engines where surging. That and a few tests done on some motors have basically shot down common rail for our club. theres like all of 1 or 2 people who still support it.
There you go. What more do you need? Or asked another way, how much more in details do you need? Look it's not the fault of the directions nor does it have anything do to with wire size. It has nothing to do with any particular transformer. Gap your blocks as per Atlas's directions, using the Atlas Controller and the green sliders to set-up cab A. and cab B. Start up one train westbound and get it running using cab A. Start up another train in the opposite direction using cab B on a separate block and watch the action for yourself. You do your own research and see what happens to the operation of both your trains. Oh you can run one all day long and have absolutely no problems. You can run two trains in the same direction...no problem. BUT, run two trains in the opposite direction. There's the clincher. Again, what more do you need...to prove it to yourself. I know, I'm credible and that's all that matters to me. I know what I'm talking about and I've seen this tooooooooo many times. smile: I also know Ben, is credible. And, I'd recommend a hard look at what he's done. Unless he's pulling my leg.laugh:
Well, I just got out my old Atlas stuff and did a test... I used a short bit of Unitrack divided into two blocks. Everything worked fine of course. Two engines, traveling in opposite directions: changing the speed of one does not affect the other at all. (Well, actually, not everything worked fine. An old trainset power pack I used turns out to be broken, and puts out a constant 13V DC. I was thinking about making a video for Rick's benefit, but that'll have to wait until I find a different powerpack. With this one I'll crash the engines while trying to record. )
That would be expected as long as both locos remained in their respective blocks. What happened as one loco ran into the other loco's block? (Hey, it could happen! :tb-tongue
C'mon Doug, this is irrelevant to the question being debated. Rick is claiming that Atlas components have inherent problems even when trains stay in their blocks. In any case, what would happen is the same thing that would happen if the two blocks were wired with direct home wiring: a loco crossing the block boundary will shuttle back and forth or stall.
Scanning Rick's earlier posts, it was never stated explicitly whether the locos were in the same or separate blocks. Based on the discussion, the assumption is that they are in separate blocks and somehow interact. The only way that would happen is if a gap had accidently closed (e.g., via thermal expansion of the rail), a loco was bridging the gap or there was a mis-wire somewhere. There is one other way which has not been discussed: Ground Loops. If there is any significant resistance on the common (and rail is not known for it's low resistance), then there can be induced voltages between the two on the shared common (refer to Robert's diagram in post #39 of this thread). I would not exclude the possibility that the common that runs through the Atlas components is not as robust as, say, #14 stranded either. Given proper implementation of common rail wiring with quality materials and techniques, all should be fine...unless you run a block, in which case the results are, as they say, "indeterminate". With DC, no harm is generally done, as voltages are usually low and you are not dealing with low impedance voltage sources, but with 12 -15 VDCC with solid regulation, it is more than easy to damage a DC pack, your DCC booster, or both if they somehow try a chest-bump.
In the Atlas system, the two cabs and the common rail all connect through the Atlas controller... In order for the ground loop situation Robert describes to happen, the path would have to come back to the Atlas controller through one of the connections on the right and pass through the controller to the wrong power pack, before coming back to the controller again and connecting through the common to go to the other pack. However robust the connections inside the Atlas controller are, I highly doubt that the common rail connection is less robust than the rest of them. IOW, if the common connection in the Atlas Controller is not robust, then the ground loop connection is even less robust, since it has to go through such wires in two places instead of one. And therefore the way the Atlas components are designed really makes a bad ground loop less likely, not more likely, compared to wiring separate components yourself. My guess for the reason behind the symptoms Rick has been describing is that people use the actual common rail as a bus instead of wiring it to something adequate. In other words, they only wire one feeder to the common rail. Old Atlas directions may say or imply you can do this. It's what I did on my first layout, and I may have just gotten lucky. (I certainly had voltage drop issues when my trains got around to the far side of my loops.) Furthermore, people may fail to solder rail joiners on the common rail (I bet the Atlas instructions don't emphasize that either). Then the common could be actually disconnected for certain blocks and the symptoms Rick describes would be almost guaranteed.
So long as the common connection is a single point, there would be no ground loop, but if there is a common path (not just a point) that has any significant resistance (even a few tenths of Ohms), then the current drawn by a loco in Block A will generate a voltage over the length of the common path that will couple into the voltage applied to Block B - a classic ground loop. The point about using the common rail as the the common connection is dead on, as rail is not particularly low resistance. As a general parctice, I would wire everything using direct home (bi-wire) techniques, and where common rail was called for (such as detection sections within a block, or even the whole railroad), then I would simply wire the feeds for the "common" rails from each block together at a single point under the benchwork to which the common of each power pack is also connected, and WAH-LA! No ground loops! (This can be done with Atlas components as well).
Me thinks we should of started with a definition. A common or ground is used to tie any and all ground wire leads to one wire or a designated rail. Thus you have the term common wire or common rail. Where it gets confusing with Atlas and the controller Jagged Ben, shared a picture of, is your common or ground will also act as a hot or positive when operating two trains in the opposite directions simultaneously. This is where the trouble starts. A railroad that is wired with DPDT or Double Pole Double Throw electrical toggle switches (Bi-wired, as mentioned early on) never has a common ground wire or common rail. The ground is built into the DPDT switch and changes with the reversing polarity of the transformers reversing switches. Add in a reversing loop or wye and it does complicate things. Since, we tend to use the reversing switches either on Analog DC, transformers and digital commands on DCC the common wire is pretty much a thing of the past. DCC doesn't use a common wire or rail. Over the years it has proven to be problematic and an extremely frustrating way and means to wire a home or club layout. Most club layouts dumped the common wire opting for or changed to DPDT wiring specifications. There are a few die hards out there who are still stubbornly hanging onto to a fallible, problematic and frustrating common problem. On operations night, if you want to see some interesting running problems, stop in for a visit. The basic problem with Common Wire/Rail is the opposing signals are fighting each other. Never mind what happens when you throw the reversing switch on the power source. I've already said this before but I will repeat it here for Ben. The positive from both transformers goes searching for it's own home base, some have called a home run. When it encounters a locomotive pulling in the opposite direction, other then the one it has been assigned to pull, it actually works against the motor causing it to rotate backwards as both plus's fight for control of the motor. What you see "Symptomatic Behavior" is the locomotives moving forward, in opposite directions, in a slowing, stuttering and halting performance. If this is ok with you then...it works. I prefer a smooth opertion and absolute control of my locomotives. So, I choose to build my layout to DPDT specifications. It runs smooth as any you could ask for. Well, I've spent enough time here on this thread. Try it for yourself and don't take my word for it. Prove it to yourself...if you must! Just be carefull and don't fry your new transformers with the memory functions Ie., Momentum and Brake.
No Rick, that isn't the basic problem. There is no basic problem, because common rail can work just fine, on a small layout or on a layout with hundreds of blocks. If common wire/rail is problem prone, those problems are complex, not basic. I don't think anyone in this thread has adequately explained the complexity. I think we've already proven in this thread that "your mileage may vary." If you like, next time I'm at the club I can record video of trains travelling in opposite directions (DC or DCC) and meeting at sidings or double track on our common-wire layout.
Yes, it's complex and problematic. Actually I'm not sure we've proven anything...as you would need all kinds of supportive documentation and wiring diagrams with illustrations... to do so. Did I just read that you operate on a layout that has a common wire for DC and DCC? Why? And just how does that work? Most DCC systems advise not to use a common wire as this may interfer with the Digital Command Control and the signaling of the unit. You need to use a bi-wire bus. Furthermore DCC is constant AC on the rails and Analog DC is Variable DC or Direct Current on the rails. When it comes to operating trains the two don't mix very well together. Common wire? I don't think so. I know you are yanking my chain...again...right?eek: You can make me out to be the world's biggest dummy if you wish. Only...you are in charge of your own perception, with regards to what and who you think I might or might not be Dejavu, as it seems you and I have been here and had this discussion sometime in the past.
Rick, the reason our club's N scale layout uses common wire for both DC and DCC is that for the first decade and half of it's existence, our layout was common wire with DC only (9 cabs, 5 mainline and 4 yard cabs). Re-wiring the entire layout to direct-home wiring in order to add DCC would have been a massive and probably prohibitive undertaking. So the question was, could we install DCC using the common wire? (This was all before I joined the club, but anyway...) Our current wiring guru (a German who designs computer chips for a living!) called Digitrax to ask some questions. As it turned out (and I mentioned this earlier in this thread), they said that DCC should work fine on the common rail, but that they couldn't support installing it without using optically isolated connections for the Loconet on the boosters. This is to prevent any ground loops from forming through the Loconet connections. Getting the optically isolated boosters from Digitrax cost some extra amount of money ($25 or something, iirc). With this safeguard, DCC on common rail works like any other power pack on common rail. And no, the DCC messages aren't affected by the other cabs using the common. If we were doing it all over from scratch we wouldn't use common wiring. It does create a few weird side effects when our DCC power sections shut down due to a short from a derailment or other such things. Sometimes another section shuts down as well, which defeats the purpose of sections a little. OTOH, I've been told the common-wire may prevent voltages from adding when we overrun blocks. So apparently it acts as a safeguard in some ways. I'll try to take some good video at the club just to prove that I'm not yanking your chain.
Hey Ben, Interesting thoughts about the common wire. When I first saw the books on the DCC wiring plan, the bus they recommended, it reminded me of the common wire. As I read into the chapters I did bump into a caution or warning against using the common rail set-up as used by some model rails. I'm just guessing here but I think your electrical engineer is using the common rail or bus as the low side of DCC's, A.C.. That much would work. I can't possibly visualize using Analog DC and operating DCC, AC at the same time without isolating the two from each other. I have problems on my home layout when I over run a train operating on Analog DC into a block occupied by the DCC, AC. There's no way I would wire or operate the two on the same bus. The club layout I operate on did change from common rail to DPDT spec's. It was an easy switch over to DCC. A simple flip of a DPDT switch and trains were up and operating. Yep, I will look forward to the videos and I would appreciate any info, knowledge or how to's you can share on this subject. Not my place to be critical of what your engineer has created and if it works and works well for you...how lucky can you be...right?
just to "stir the pot" After re-wiring the west wall to common rail, Both me and Phil have noticed the low end speed of our engines and their torque has improved DRAMATICALLY. The engines run smoother and seem to perform better since we've swapped for bi wire on the west wall. No more power drop on the grade either I'm really liking the looks of this now.
Huh, I'd never heard of bi-wire setup before now; I'm a bit intrigued. Anyone have a guide to how it's done handy? All Google gets me is stuff about speakers and this thread :tb-err:
