Wire question

[chuck]

Silver is a better conductor but it's reactive.  Gold is almost as good a conducctor and is fairly inert, especially in its purer forms.

[ChiefEagles]

Silver is a better conductor but it's reactive.  Gold is almost as good a conducctor and is fairly inert, especially in its purer forms.

Why our Pro Tournament Chargers have all gold contacts and rings.  My son always used gold for his Show Cars and electronics.

[kpolak]

Silver is a better conductor but it's reactive.  Gold is almost as good a conducctor and is fairly inert, especially in its purer forms.

I believe gold to be a relatively poor conductor...but it is excellent in resisting corrosion inheritant in copper, and silver.  Ask Chief about his gold hooks! Still shinny!

Also, the fancy schmancy audio cables used in high priced audio actually improves the sound of the system.  (I believe)  You'll have to try it for yourself if you even care...this stuff can get expensive.  Compare a small piece of THHN (14 gauge house wire) with just about any audio cable.  But, if you like the sound of your stereo now, don't fix it.

With the trains you only care about electricity getting from point A to point B..this is simple physics, audio cable is for audio.  If you can't HEAR the difference don't buy it.

Kurt

[lionelsoni]

Gold comes in third, behind silver and copper, just ahead of aluminum.  Not too bad.

Silver 15.9 nanoohm-meters

Copper 17.2 nanoohm-meters

Gold 24.4 nanoohm-meters

Aluminum 28.2 nanoohm-meters

[wrmcclellan]

Interesting set of posts.

Stranded vs solid wire for DCS. There seems to be missing important information regarding the gentleman who rewired his layout 3 times. Our former club layout was wired with solid copper conductors back in 2002. I used #12 romex style house wire (yellow jacket 12/4 NM-B - 3 conductors with ground - white/black/red insulated conductors and bare ground wire - white and black used for track 1, red and grd used for track 2, etc.) from home depot on a 30 x 45 foot layout in a home run/star pattern wiring with 6 distribution panels that then had #16 solid copper wire to the rails. Each of these 4 wire NM-B cables terminated at the TIUs (2 TIUs used). We had DCS signal levels of 10 all the way around.

What is missing from the description is the wire geometry or other factors. Was paired wire used (As we got in the NM-B) or was it loose/individual conductors. Regardless, in addition to the wire change, something was intentionally or unintentionally changed between tries 2 and 3 that affected the DCS performance. Maybe it was addition of light bulbs to terminate the signal and control reflections. Maybe it was something else. It is not the wire itself.

As I have written in the past, DCS is a radio signal that uses the twin lead (balanced pair) wire as a transmission line. Recall the old twin lead TV antenna wire. Note that it had a controlled spacing between the two conductors (to achieve a 300 ohm impedance to match the yagi/log periodic antenna on the roof). DCS needs the same basic geometry for best performance. The DCS patent is quite broad, but describes a 100's of kilobaud spread spectrum signal that would not be a very high frequency. Thus the need to tightly control the twin lead spacing and line impedance are not as critical as if it were a higher frequency television signal. Thus any good quality paired conductor wire (appropriately sized for the necessary current) will work well for DCS.

BTW - Gold is only shiny when it is lightly plated over nickel which is a poor conductor (69 nano-ohm*meter) when compared to other common conductors. Great for jewelry and connectors that are not used for high frequency applications.

[lionelsoni]

Roy, I think you meant 12-3 WG.  (I would hate for someone to waste his time finding 12-4...;-)

[A&Y Ry]

The fourth conductor is for for debugging DCS Bob.

[Anonymous]

Roy,

If you get a chance and have the issue maybe you could read the article in the July '05 issue and report back on your findings.  The author gives the exact types of wires used and if I recall makes no mention of any other modifications or add ons other than switching from solid to stranded core. 

Bruce Webster

[wrmcclellan]

Bruce,

Read the article when it came out and looked at it again just now. I do not believe the premise that only stranded wire works well for DCS. It goes against well established theory. Again as evidence - our own layout which is larger than the one mentioned in the article used solid copper wire with no problems.

BTW - the article mentions 14 ga twisted solid copper wire. I have not seen such a thing myself. A quick internet search did not produce a hit. Maybe Lionelsoni has seen such a thing?

[lionelsoni]

I have never seen it for sale; but it would be very easy to make simply by twisting a couple of individual solid wires together.

[Anonymous]

Roy,

  Thanks for reviewing the article.  All I can say is, while no real reason was given as to why stranded works better for the author I did think his process of elimination proved a point.  This and the fact that MTH advises it's use explains why I recommended it in my original post. 

Obviously you have blown his theory out of the water by using solid on a similar size layout utilizing 2 TIU's.  Add to that the combined knowledge of you and Bob and it becomes clear that DCS will work either way.

I wonder if the author was having the signal problems he mentions due to accessories prior to the installation of stranded wire.  This might be why he blamed solid core wire as the cause.  Let's ask him to rewire his layout with solid core again and see if he still has signal problems

Bruce Webster

[ChiefEagles]

YUZ GIZ talking about Jamie?  If so, I think he runs a communictaions company and is very knowledgeable about these things.  Very nice guy.  Talked to him several times [about something else].  Interesting. 

[Anonymous]

Yes Chief,

  Jamie Haislip

Bruce Webster

[wrmcclellan]

Bruce,

At the time we were building our layout and for a couple of years people were really struggling with getting DCS to work on a large layout (Including our own group). The number of posts and theories being offered for DCS operation on the very active OGR forum (particularly this subject) took a lot of time to wade through. People were literally pulling their hair out and until the light bulb trick showed up as an easy way for a modeler to terminate and impedance match the DCS signal - nothing worked reliably. We can only speculate on what steps Mr. Haislip implemented as he found his final solution with twisted wire.

I should add that we added a DCS signal termination at each distribution panel which is where the NM-B cable terminates in a strategic location for each subsection of the layout. From the distribution panel is where the 16 gauge wire (two individual wires - not twisted - tie wrapped together) emanates in a spoke like pattern to the track drops (every 4 - 6 feet) in that section. And to keep it interesting (and because I grew very weary of soldering upside down), I used small wire nuts to connect the 16 ga feeders to the wires that were soldered to the rails and dropped through the table. In fact - our entire layout track feeder system is mechanically connected using electrical panel neutral buss bars and wire nuts - very little soldering. Same stuff as used in your house wiring - it has been very reliable (and it is located in a non-temperature controlled warehouse - Chief has been there).

The use of the wire nuts make it easier to isolate a section of rail should we have an electrical problem. At each track feeder, we dropped the center rail and both outside rails. We haven't had any electrical problems that required isolating any track to date. We have soldered lots of jumpers around the early production Atlas switches which loose their internal electrical connections. I have recently read that Atlas has beefed up those connectins on newer production.

[Anonymous]

Roy,

  Mr. Haislip's article was the first on the subject to come to my attention right after I got back into the hobby.  Heck, the last time I was involved with O the big innovation of the period was Railscope! So, I wasn't savvy to the previous problems others had experienced with DCS.  I followed his wiring instructions pretty closely with the idea of some day buying DCS or if I chose the option of going exclusively with TMCC I wouldn't have to rewire.  It also gave me the best senario if I chose to use both CC systems.  I'm currently waiting for my order of TMCCII as my first CC system.

I hear you about the soldering too.  The only thing I soldered was connections to rails....even got good enough at it to avoid melting ties.   Issues with my Atlas switches were daunting but nothing serious to fix.  Once you've played with N or HO, continuity problems with O are childs play.  Everything else is connected with crimped folk type wire ends attached to buss bars and terminaters.  I have a control panel but plan on modifying and removing some block section toggles once TMCCII arrives. 

One difference in wiring is that I have isolated one outside rail on all my trackage except on switches.  This I did in hopes of eventually installing automatic signals, crossings, etc., using the operating trains for the activation of these accesories instead of using IRD.  Hopefully CC signals won't be affected by this.

Bruce Webster

[wrmcclellan]

Bruce,

One thing we can and should clearly say - following Mr. Haislip's advice is good practice. It is good advice.

I was an HO modeler way back and that means you definitely have the track skills for today's O track systems!

Depending on the size of your layout, I can verify (as have others) that smaller layouts tend to work fine with DCS and older style block wiring (where only the hot leads go to the block power switches - as on my current 8 x 12 layout which works fine). My advice is to leave the block wiring in place and see if you have a DCS problem or not. The blocks are very handy for isolating shorts (I had a new piece of track that when ballasted shorted out - but it is tubular) and even with the blocks I pulled up 6 feet of center rail until I found the problem. Without the blocks I may have pulled up even more track.

If you have a problem with the block power switches, route a common wire with each hot wire to give the DCS signal a paired wire "transmission line". Tie wrap the wires together. This is what we did on the club layout - no problems.

I forgot to add above that the reason we dropped both common wires was (1) if we were not signaling - then the extra common is useful and does not harm DCS or TMCC or (2) for signaling - we already had the wires dropped for the signaling rail (we added signaling on the main line on the club layout).

The real test for DCS on a large layout is if one can add a new loco sitting anywhere on the layout.

BTW - Block switches are handy for turning off idling locos.

[dwiemer]

Hey guys, though the thread took a turn from its initial discussion, it has been one of the most informative, thought provoking, and well mannered discussions so far.  It has been a pleasure to read through.

I would like to add to Bruce, and anyone else wiring for TMCC/potential TMCC in the future, I believe that two issues ago with OGR, they had some tips.  One tip, was that if you were having a layout in where one track passes over another, or in close proximity, you should have a true earth ground between them to provide proper signal grounding.  The demonstrations that they had were going through a tunnel, and in a yard situation.  They had small gauge wire (22), that was run along the roof of the tunnel and then, disguised as a power cable on telephone poles through the yard and then eventually to the ground slot on a outlet.  I am in the process of wiring my layout and plan on doing this.

Dennis

[Anonymous]

Thanks for the advise, Dennis.  When the new layout begins I'll definitely need to add ground wires for signals in a number of areas.  That's not to say I have anything on paper yet.   Right now it's very difficult to imagine multi-level tracks, well thought out yards and tunnels that suit the scenery when I haven't even got the room roughed in yet.  I think I'll be 3 dimensionally challenged until I start to see some benchwork assembled.

Luckily the existing layout is all on one level.  Up and running long enough to have all the bugs worked out too.  It will be my test bed for TMCCII once it arrives.  The steel and aluminum arch bridge should be the only problem area with signal transmission, if at all.  I hear simply grounding the whole metal structure should solve signal loss here if it does occur.

Thanks to everyone who has posted on this topic.  I've come away much wiser in the process. In no way does this mean the end.  Any other tips would be greatly appreciated.  Oh, and I appologize again for hi-jacking the original posters topic.

Bruce Webster

[ChiefEagles]

Yes Chief,

  Jamie Haislip

Bruce Webster

He is considered as the "IT" in DCS by a big group [most hanging out on another forum].  Interesting.  I ran all wires to the lockon from the buss bars [two exceptions where I soldered on to the wires of another lockon to keep from having so many long runs into the Walk-in].  I ran my PS2 switcher over all my track.  Signal strength was 9 and 10 all over.  I used 14 ga stranded speaker wire [Frankie gets it wholesale].