True or false, a 1 twist in every foot will help the signal quality of the BUSS?
I am using 14 AWG Stranded wire, (red, black) string it thru brass eyes under the route. Feeders are 22 awg SOLID wire every 3 ft, all joints except turnouts will be solder. Every 10 sections I will not solder one side, expansion.
Thank you
George P.
What you are refering to is a capacitance effect between current carrying wires that are in close proximity to each other. This can be a problem in some curcuits, but usually in the high frequency/RF range of signals. DCC is alternating current, but considering it can be measured with an ordinary volt meter, I wouldn’t think it would be in the high frequency range…
There’s been a lot of discussion on this and much smarter people than me suggest that it is actually inductance and interference, not capacitance, that creates problems in long buss runs. DCC is not pure AC, but a square wave signal that can be degraded by these effects in long buss runs as the “corners” of the square wave signal become rounded-off, spikey, etc.
Those are some of the reasons a gentle twist in the buss wires is suggested.
From Allan Gartner’s excellent site:
http://www.wiringfordcc.com/track_2.htm#c2
Regards,
Byron
Bottom line is, it probably isn’t required. It won;t hurt to do it, but I doubt there is any noticeable effect. With an oscilloscope, sure, but within the operating parameters of most DCC decoders, probably not. You probably shouldn’t run any given bus line more than 50 feet fromt eh booster anyway, even with #12 wire - which menas 50’ in either direction, so that covers 100’ of layout. It’s better to distribute the boosters than run really long bus lines, especially if your power requirements need more than a single booster anyway. For example, if your layout strethech around 100’, and you need 2 boosters, putting them 25’ in from the ends and 50’ apat means you have no one bus run longer than 25’. Signal degredation? Voltage drop? What degredation and voltage drop?
–Randy
As an EE, I think that twisting the wires would be way down the list of things I’d worry about. Twisting the wires is an attempt to get the same noise on both sides of the signal so that the difference can still be detected. The DCC signal is huge, and pretty slow (in electon terms), so it isn;t going to be relevent. The biggest source of noise is going to be the motor in the locomotive itself, where you can’t twist the wires. Also, twisting the tracks that the signal also runs through is a bit dicey.
Better to worry about good, solid connections, length of runs, short protection, and good power supplies, I think.
This question was based on the Nov MRM DCC corner that had them Twisted.
Thanks All
We have:
(2) PS515
(1) DCS 100
(1) DB150
(1) DT400
(1) dt100IR
All we need is:
Tonys Trains stuff and 45 mobil decoders and 18 Stationery decoders.
Oh yea, fini***he bench WORK.
Then I can start the DCC process. Just trying to have all the parts on the shelf before I start the job.
Exactly - you can twist the bus wires all you want, but every 3 feet you have the connected to what? Two nice shiny conductors perfectly in parallel to each other - one giant capacitor! So while the signal may be better on a long twisted bus, once you connect all the rail feeders, it all goes down the tubes! Hmm, tube amps have more ‘warmth’, maybe we should make a tube DCC booster. LOL!
[:D]
–Randy
I’m not sure where you see anything about twisting wires in that article. [%-)][%-)][%-)]
For me and my understanding of this twist or not to Twist buss wires issue is not a DCC thing. If you are running DCC and turnouts and such, no big deal. Either way will work just fine. The issue really doesn’t pose a potential problem until you start to do “Block Detection” and/or “Transponding”. The equipment to do the block detection is very sensitive. So twisting is recommended. The recommendation in these cases also suggest keeping the network (Loco Net) at least 6 inches from the buss to help eliminate “Cross Talk” which could confuse the block detection.
I read somewhere that this twisting and separation came from a layout who installed block detection. They could not get it to work and spent many months with lots of help from Digitrax support to find the problem Their buss was one big and very long wad of power, loconet, accessories, etc. (Like big phone cable with 50 pairs in it). Finally after separating all this wire, they got Block Detection to work correctly.
My suggestion is TWIST! It doesn’t hurt anything, and it MAY help in the future.
Bre2tSco2t
[:I][:I][:I][:I] It was in the OTHER magazine. THe MRM had the hooks showing the hanging of the buss. The OTHER one had a picture shoing the twist in the RED/BLACK Buss.
Solved my possible cross talk for future signals. Got some MIL SPEC 22 AWG solid 4 wire with a tracer and foil sheating. Talk about over kill!! Got 450 ft of it for $25.00.
OK, guys! It’s BUS not BUSS. BUS, like that city transit thing that hardly anybody rides anymore. Also, a wire is more of an inductor, and two wires are two inductors not a capacitor. I also fail to see how two parallel rails constitute a capacitor. At what frequency? I have seen a small straight piece of wire used as an inductor in VHF transmitting equipment, but at DCC frequencies? I think we’re being silly here. The axiom that seems to apply when wiring a layout for DCC is, “The better it looks, the better it works”. Leep it neat and you should be fine.
Cheers,
Ed
Exactly! I think the idea of keeping the “power” bus sepatate from the control bus (in whatever system you have) is a good idea, since the big signals on the power might try to corrupt the control. Don’t know a lot about the transponding side of things, but even then, I’m sure it is low frequency, keeps things neat and clean and things should be good. I suspect that twisting was not what made things work, but rather other things that happened when the twisting was done, like getting the control away from the power, improved connections, etc. If twisting made a difference, we’ve got a system that is too fragile. We have no termination (on the power side), rails, motors right next to decoders, switch machines, lighting, and who knows what else running around the train room. We get away with it because the DCC signal is huge and slow. The control bus (loconet, or whatever), and the signals to boosters should be better protected. Some are coax, some twisted pair. And they should probably be kept away from noise sources, like the power bus, switch machines, lighting, etc.
Too much thinking for now…