Should the main bus be straight wires or should it be twisted wire pairs. I wired an ethernet into my house and that is all twisted wire pairs. I'm thinking dcc is sending signals over the wires so probably twistetd wire pairs would work better than separated straight wires. How about feeder wires from the bus to the track? I think twisted wire pairs make a neater installation also.
The power bus wires should be kept close together to mininimize their inductance (and the resulting impedance) of the bus. They don't have to be twisted, but being kept close is just as effective
DCCWiki - To Twist or Not to Twist
Over my 18 years and 5 HO scale layouts, I have never twisted the bus wires, and I have never had a problem.
Rich
Alton Junction
betamaxDCCWiki - To Twist or Not to Twist
In fact, Alexander Graham Bell discovered that by twisting the two wires together, interference and inductance were reduced for a stronger signal and clearer communications.
so while trwisting may be unnecessary, would it make things worse?
wouldn't wrapping the wires once/twice a foot be an easy way to keep them "close together"?
greg - Philadelphia & Reading / Reading
I am with Greg on this one.
On the two DCC layouts I helped build, we twisted (or used twisted/shielded) the main bus wires.
It might have been a waste of time, but it sure did not make anything worse.
-Kevin
Living the dream.
My main bus is installed. The only way I would be able to twist it would be to take the whole thing out and redo it. I don't think I want to do it for what may be a waste of time.
Rio Grande. The Action Road - Focus 1977-1983
Mine is run without twisting or being together. From what I have learned this twisting or no twisting is an issue on large railroads with long stretches of bus lines, very few of us have anything big enough to mater.
My view is that the more we fudge on things the more likely we will have problems or maybe just a little less than optimal performance. Whether it is settling for less than perfect trackwork, or the multitude of other things we work on, taking care of the little things takes a lot of frustration out of the hobby.
I had the same question years ago and with what little knowledge of electronics I do possess, after reading a bunch I decided to twist the wire as it would only take a couple of minutes to pull 150' off the spools down the driveway, have the wife hold one end and stick the other in a drill and turn it on for a few seconds to twist the wire together.
More time has been spent reading this thread than it would take to remove any lingering doubt by twisting the wire.
Brent
"All of the world's problems are the result of the difference between how we think and how the world works."
What about a 50' run. How long is "long"?
at least with many engineering things, there is no one right answer. there should be good reasons for doing things
the circuit on the left doesn't have a capacitor across the bridge that the one on the right has. we recently installed detectors without the capacitor to find that many, but no all reported false detections. adding the cap across the connectors resolved the problem, but it was never clear why only some needed it.
engineering designs are often driven by cost. why add or do something if it's not needed, but then again, why take a risk, how much does the insurance cost ($$).
it's good to know why
gregc at least with many engineering things, there is no one right answer. there should be good reasons for doing things the circuit on the left doesn't have a capacitor across the bridge that the one on the right has. we recently installed detectors without the capacitor to find that many, but no all reported false detections. adding the cap across the connectors resolved the problem, but it was never clear why only some needed it. engineering designs are often driven by cost. why add or do something if it's not needed, but then again, why take a risk, how much does the insurance cost ($$). it's good to know why
The cap is to catch / absorb errant voltage spikes. I recall many years ago, I built a simple flasher using a 555 timer. It would work perfect using a battery, but not with my 12 volt supply I use for layout lighting. Finally, I determined, by adding a small cap across the voltage input to the circuit, it worked as expected.
Mark.
¡ uʍop ǝpısdn sı ǝɹnʇɐuƃıs ʎɯ 'dlǝɥ
Mark R.The cap is to catch / absorb errant voltage spikes.
in our case, we were wiring a multi track staging yard with 4 detectors / track. why would there be spikes on some feeds but not others on the same staging track or the one next to it?
Mark R. gregc it's good to know why
gregc it's good to know why
The twisted buss wire is in response to “ringing” in a DCC system that causes voltage spikes during shorts that will blow up decoders or cause them “lose their minds” and need a reset or replacement. The experts don’t really agree on this issue. My friends and I have had various issues over the years that were solved by a variety of methods but it is hard to tell which solutions worked….
The other issue is why were the decoders blowing up/going crazy? Is it ringing? Is it poor decoder design (don’t get me started) Is it the DCC systems design? All I can say now is that after a variety of practices were instigated, the issue has gone away.
Many/most layouts will be just fine. I didn’t start having problems until I hosted OPS sessions and start running six locomotives at once with operators tripping the breaker every once in a while. If you rarely trip the breaker and don’t do things like install decoders in old brass (or even just run old brass) your incidence of shorts will be so low as to not bother with this at all
Spikes in DCC have been measured but are so inconsistent that without a scope and some expertise, it is hard to be sure your layout has the issue and even what types of design cause the problem. Length of buss run does appear to be a factor. Twisting the pair helps mitigate the problem. The issue I have is that twisting the pair makes it a lot harder to solder feeders to the buss. Given that I have hundreds of feeders on the layout - that is an issue for me. I use a resistor at the end of runs instead of twisting the wires.
Way too much from me on the topic.
Your mileage may vary,
Guy
see stuff at: the Willoughby Line Site
RICH BONDShould the main bus be straight wires or should it be twisted wire pairs..
Twisted pair has a different meaning and a different intention than twisting the buss wires.
Larry Puckett, the DCC guy, who use to write for MR, uses wire that looks like speaker wire and says that solves the problem.
I use single wires and twist them a couple times per linear foot.
If you have an oscilliscope, you will worry about noise more than the rest of us.
Henry
COB Potomac & Northern
Shenandoah Valley
My bus is Romex standard solid wire for wall circuits in houses. I'm sure it twists a bit by happenstance as I pull it through all the knotholes I drill under the layout. But I don't inadvertently twist the rails they feed. Somehow, in all those tens of feet of parallel tracks, I don't have any problems.
trainnut1250 Length of buss run does appear to be a factor. Twisting the pair helps mitigate the problem. The issue I have is that twisting the pair makes it a lot harder to solder feeders to the buss. Given that I have hundreds of feeders on the layout - that is an issue for me.
Length of buss run does appear to be a factor. Twisting the pair helps mitigate the problem. The issue I have is that twisting the pair makes it a lot harder to solder feeders to the buss. Given that I have hundreds of feeders on the layout - that is an issue for me.
Too many people overthink this and just end up confusing themselves and everyone else in the process....
The inductance of a wire is directly related to its gauge and its length. The longer it is the more inductance it will have. A lighter gauge wire also has more inductance than a heavier one.
Easiest way to mange inductance is to keep all the wires close. They don't have to be twisted, just side by side held with tape or cable ties will work wonders. Makes it easier to tap into them too.
As to spikes, that is the nature of the beast as the inductor is charged and discharged. Too much inductance will round off the leading edge, and cause an overshoot during the discharge phase. Adding an RC circuit (if necessary) provides a method of absorbing some of the overshoot. But a lot of that is also related to the design of the booster output too. They are not designed for every specific application a model railroad may present.
Just as a bitty note: the twisting appropriate for DC or DCC datastream does not have to be tight as in Ethernet pairs, either in turns per inch or tight proximity. As a result you can easily leave enough slack in the wrap to use an appliance wedge or similar thin piece of wood inserted through the (loose) wrappings both to isolate and 'backstop' any feeder soldering or insulation-removal activity...
Guys DCC isn’t Rocket Science, the data rate of DCC falls within in the regular human audible range (20Hz to 20kHz) at about 8kHz. I hooked my DCC controller to my existing DC wiring back in 2005 and I’ve never had any problems with DCC operations in 17 years.I do not use a DCC buss. I originally wired my layout using DC blocks with #19 AWG Bell Wire to each block when I built my current layout, the longest pair of track wires is about 25’ or 7.6M. With dual locomotives drawing a total of about 1.5 Amps the voltage loss through the #19 AWG Bell Wire at the rails is under.09 volts.Early on I checked my wiring for ringing with my Tektronix Oscilloscope and saw none.The only problem I had with DCC was current detection for my signaling system, I used old school Twin T circuit from the late 60s until the mid 80s when I went to Rob Paisley current sensors.When I added DCC in 2005 I was operating dual mode, DC or DCC, and the Paisley circuit detectors wouldn’t work in DCC mode. Over the years I eventually cut over to optical beam detection and now use reflective IR detection, I like it much better than current detection. I no longer need to add resistors to the trucks and the IR detectors (less than $1 each in bulk) sense every car.Mel My Model Railroad http://melvineperry.blogspot.com/ Bakersfield, California Turned 84 in July, aging is definitely not for wimps.