Let's talk about wiring. I figure I'm going to have a lot of questions. So let's start with the little I know about it. Feel free to correct me here.
1) If you are drawing a lot of amps you have to have heavier guage wire. Power tools such as worm-drive saws (draw 13-15 amps depending on manufacturer) need at least 12 guage wire to run 100 feet on a power cord or they suffer from voltage drop and can burn up. They use twisted wire.
2) The same principle applies to house wiring. If you are on a 20 amp breaker you need 10 guage wire. On a 15 amp breaker you use 12 guage wire. Both of these are solid core wire.
3) Data--phones and computer--is trasmitted over some pretty great distances using thin 24 guage solid core wire. For some reason, they twist it. That it is except LocoNet wiring which is a flat ribbon.
4) The smaller the wire, the easier it is to solder to the track and make it look good.
5) Soldering thin wire directly to a solid 12 guage bus isn't as easy as it should be.
I want to understand the why of things so I can make decisions about my layout. For instance I have a sufficient length of 12 guage romex residential wire, and a lot of Cat 5--like half a box (250 ft or so 8 strand twisted pairs). I also have some short pieces of 16 or 18 guage solid core wire in red and green--probably enough to do the feeders.
First 2 questions:
1) What is the advantages or disadvantages of solid wires vs twisted wires?
2) Why are wires transmitting data twisted? I know they suggest twisting bus wires on DCC layouts, but from what I seen in reality on YouTube, the twisting is half-assed (and mostly, I think) to say they did it.
Chip
Building the Rock Ridge Railroad with the slowest construction crew west of the Pecos.
110 volt AC 15 amps = #14 wire 20 amps = #12 wire 30 amps = #10 wireThe breaker size is determined by the wire size. If the wire is smaller than the breaker size then the wire will burn before the breaker trips.
Amps times volts = watts
15 amps X 110 volts = 1650 watts
The power from your transformer is DC for starters and it is not 110 volts. It is usually around 12 volts. Therefor you don't need #12 wire.
Twisted pair wiring came from the telephone company so that in a bundle of dozens of wires the pair of wires for each individual phone line stayed together and didn’t get lost in the crowd.
Lone Wolf and Santa FeTherefor you don't need #12 wire.
It's not a matter of need but a matter of got. Is there a disavantage in going with too big a wire?
SpaceMouse Lone Wolf and Santa Fe Therefor you don't need #12 wire. It's not a matter of need but a matter of got. Is there a disavantage in going with too big a wire?
Lone Wolf and Santa Fe Therefor you don't need #12 wire.
Bigger wire is stiffer and harder to work with. Also it costs more.
Lone Wolf and Santa FeBigger wire is stiffer and harder to work with. Also it costs more.
Thanks. Thing is, I already have the wire so there is no cost.
So if my bus is 45 ft long, what guage wire would you suggest?
SpaceMouse1) What is the advantages or disadvantages of solid wires vs twisted wires?
I get the solid wire, but by "twisted", do you mean, do you mean stranded wire, or runs of solid wire twisted together?
My L shaped dog bone continuous run is 50', I used 14 ga. solid wire, red and black, for the main bus, and for the feeders, I used 22 ga. solid wire. I did twist the red and black bus wire together, just make it a little more confined, and then seperated them were I soldered the feeders.
I would think that the 12 ga. solid wire, the house wiring, would be fine for the bus. I'm thinking the 16 or 18 ga., might be a little big for the feeders, but I don't know what your plan is for attaching the feeders to the track.
That's why I used the 22 ga. Just a smaller wire to deal with, and hide, as I soldered mine to the outside of the rail.
As far as stranded wire vs. solid, the stranded is more flexible. I think the solid wire is better for bus and feeders.
Mike.
My You Tube
.
The biggest advantage of stranded wire it that it is more flexible and easy to work with.
Secondary advantage: it will still work just fine if a few of the conductor strands break. It you use solid wire, and it breaks, you are done.
SpaceMouse2) Why are wires transmitting data twisted? I know they suggest twisting bus wires on DCC layouts, but from what I seen in reality on YouTube, the twisting is half-assed (and mostly, I think) to say they did it.
This is so one conductor does not induce a current into the other one that is run parallel. Not a big deal in 12 volt track wiring, but with DCC this becomes a concern. The little "1s" and "0s" that make up the data packets can become corrupted. If the data pack is unreadable the intended receiver will ignore it.
This is very true with heavy equipment that uses SAE 1587/1708, SAE J1939 or CAN/BUS for cummincation between devices. It is amazing how similar the controls for cranes are to the controls for model trains.
Twisting data pairs is very important. Neatness does not actually count for much.
-Kevin
Living the dream.
Lone Wolf and Santa FeTwisted pair wiring came from the telephone company so that in a bundle of dozens of wires the pair of wires for each individual phone line stayed together and didn’t get lost in the crowd.
All this time I thought twisted-pair coupling cut down on crosstalk between the pairs. Twisted-pair becomes essential in high-data-rate applications like Ethernet cabling.
If you are feeding relatively unmodulated DC down a 'legacy' length of twisted-pair, the value of the twisted-pairing probably defaults to what he said, useful color coding to pick out the circuit at the other end. If you are feeding even relatively slowly modulated intelligence over the DC, maintaining the twist right up to any connectors will still have value.
My last layout didn't have a buss,Feeders { reclaimed 24ga} were droped down and pigtailed , a jumper went to another pigtail,and so on. Never a problem.
I wanted to clean it up,ran a 12ga buss,cause I had some on hand. 12ga is a bear to work with. On the new layout I decided to spend the money on 14ga bus,way better, and reused the reclaimed 24ga feeders. Thats on just over 50ft of track not counting the yard and sidings.
Somtimes you got to justify not spending money agaist stress
SpaceMouse,Twisted data cables (like CAT/5) are twisted in pairs for self-shielding properties. They can resist outside interference better if they twist the matched pairs together.
LocoNet, however, is flat data cable. I've talked with Digitrax about this, and they recommend not using twisted pairs for LocoNet cables. Why? I dunno, but they know their system better than I do.
One good thing about bigger wire for DCC bus use is that you have less voltage drop and a therefore a better, clearer digital signal.
I believe 14g wire would be sufficient for a 45' bus but since you already have 12g go ahead and use it, recognizing that it will be a bit harder to work with due to its stiffness.
I would ditch the 16g & 18g wire for feeders. I think you will be mutch happier with 22g solid. Easier to solder to the rails and to the bus. Plus the rail connection will look mutch better.
A 50 foot run from the booster to the end of track is a pretty long run. My longest track power bus runs are about 40 feet and I used 10ga THHN stranded for them. I used 12ga stranded for the shorter 'occupancy zone' track power bus lines and I use 22ga solid wire for track feeders.
Modeling an HO gauge freelance version of the Union Pacific Oregon Short Line and the Utah Railway around 1957 in a world where Pirates from the Great Salt Lake founded Ogden, UT.
- Photo album of layout construction -
SpaceMouse So if my bus is 45 ft long, what guage wire would you suggest?
The longest bus on my layout was about 45 feet and I used 12 gauge solid copper wire based on conventional wisdom and advice from this forum. Trains ran through that section perfectly well. There were plenty of 22 gauge feeders. There was a minor drop in voltage (about 0.4 volts, I think), but the important thing was that section did not pass the quarter test. So, I rewired it from the far side through the booster.
Conventional wisdom needs to be evaluated on a case-by-case basis.
Robert
LINK to SNSR Blog
On my old layout (25'x50' w/ a 200' mainline run), I ran a 14AWG (stranded) bus 150' from my Digitrax Zephyr and suffered no problems for the 10 years I had it, even with 5 operators running all at the same time. I dropped 22AWG feeders every 9' and used ScotchLoks to connect them to the bus.
I used no other boosters or even circuit breakers. Just the DCS50, a UR91, and a handful of UP5's.
I'm quoting Robert but replying to all.
ROBERT PETRICKhe longest bus on my layout was about 45 feet and I used 12 gauge solid copper wire based on conventional wisdom and advice from this forum. Trains ran through that section perfectly well. There was a minor drop in voltage (about 0.4 volts, I think), but the important thing was that section did not pass the quarter test. So, I rewired it from the far side through the booster.
Actually, I don't need to run 45 feet. I picked that number because I thought it would just be easier to follow the track in a spiral. Truth is I could make 3 18 foot runs and shorten the runs.
ROBERT PETRICKThere were plenty of 22 gauge feeders.
A lot of people have mentioned 22 guage for feeders. What I have is 24 guage. Is that too small? I don't want to sound cheap, but I really am on a tight budget.
Paul3I used no other boosters or even circuit breakers. Just the DCS50, a UR91, and a handful of UP5's.
I'll be using the Zephyr and three UP5's.
I don't know enough about gauge of wire. But I've seen many-many model railroads using telephone wire. Isn't that 22 gauge? If it isn't and is 24 gauge, and it seems to work, why not use it. After all, that feeder from rail to bus don't have to be very long. Would I be safe to say "less than a foot". If longer than 1', then go with 22 gauge???
PennsyNut But I've seen many-many model railroads using telephone wire. Isn't that 22 gauge?
I just looked up CAT 5 wiring:
Since 1995, solid-conductor UTP cables for backbone cabling is required to be no thicker than 22 American Wire Gauge (AWG) and no thinner than 24 AWG, or 26 AWG for shorter-distance cabling. This standard has been retained with the 2009 revision of ANSI TIA/EIA 568.
My box was was purchaced through a contractor outlet and so I would guess it is probably 22 guage. I'll check the box when I get home. The reason I suspected it was 24 guage is I read somewhere on here that a guy had 24 guage phone wire. That will help teach me not to believe everything I read. Okay probably not, I'm pretty gullible whaen it comes to listening to model railroaders.
If it is 24 ga., I see no problems in using it. Even a bit easier to hide than 22 ga.
mbinsewi If it is 24 ga., I see no problems in using it. Even a bit easier to hide than 22 ga. Mike.
Hey Mike,
Not to interrupt the thread, but your posts now fit the screen. The links in your signature are gone, though.
Now . . . back to the regularly scheduled program.
I used #20 for my track feeders - I used red and white #12 bus wire (#14 on the last one because it was a much shorter run and since I had to buy the wire, #14 was cheaper). The #20 was also red and white, came as a loosely twisted pair, it's used for alarm switch wiring. The #12 and #14 was stranded - much easier to work with under the layout. The #20 was solid, better to make neat feeders. Soldering the small wire to the bus isn't hard - you need somethign bigger than a 25 watt soldering iton though. This is where that big 100 or 150 watt soldering gun comes into play.
At voice frequencies, much of the twisting IS to keep it neat and bundled together. Crosstalk isn't a huge problem except in the longest runs (like, miles along the poles or underground). For faster stuff like network, the twisting is critical, but still mainly for controlling capacitance. the impedence of a capacitor is dependent on the frequency - too much capacitive impendence will weaken the signal and/or distort it, reducing effective cable length.
Loconet gets away with flat phone wire because it actually tuns at a relatively slow speed - 16457 baud. The capacitance of flat wire is such that the signal can stay within the Loconet specifications for up to 1000 foot runs. I WISH I had the layout space where this would start to become a problem (they have Loconet repeaters if it is!).
--Randy
Modeling the Reading Railroad in the 1950's
Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.
ROBERT PETRICKNot to interrupt the thread, but your posts now fit the screen. The links in your signature are gone, though.
I was hoping for a comment. I was told that something in the signature was making my post go off page, so yesterday, I deleted them, which is whole nether story.
Back to wiring Rock Ridge!
SpaceMouse 1) What is the advantages or disadvantages of solid wires vs twisted wires? 2) Why are wires transmitting data twisted? I know they suggest twisting bus wires on DCC layouts, but from what I seen in reality on YouTube, the twisting is half-assed (and mostly, I think) to say they did it.
1. Twisted is easier to bend. Solid is a little more robust and cheaper. But it also helps hold the wires together.
For your garage sized layout, you can get away with 16 gauge with #20 feeders every 3 feet or so easy. I use stranged 16 gauge on my son's layout. Just google "Wire Voltage Drop calculator" to figure out what you need. Stay within 5% and you should be good to go.
2. Twisting wires is a way to reduce signal interferrence. (Noise)I use these to hold my power bus across the layout.https://www.lowes.com/pd/Gardner-Bender-50-Count-3-4-in-Plastic-Insulated-Cable-Staples/4634631
BTW: I color code mine
Blue = DCC+
White = DCC -
Black = ground
Yellow = +12VDC
Red = +5VDC
Green = 18VAC +
Orange = 18VAC -
Don - Specializing in layout DC->DCC conversions
Modeling C&O transition era and steel industries There's Nothing Like Big Steam!
DigitalGriffin feeders every 3 feet or so
Some people suggest feeders on every piece of track. They say that one should not trust joiners for electrical connections. I suppose I could trust a joiner up to the point of failure...I do have a multi-meter. What about turnouts, do you wire every one?
For once that you can, like Atlas. I never had to hook up my frog power wires ebcause with power feeds on all 3 legs, every other part of the turnout had a reliable power connection and even a ittle 44 tonner could creep over the dead frog area. WHat I did on my last two layouts was make every rail joiner a power feed - basically stacks of terminal joiners. Even after painting the rails, including the joiners, there was never any loss of power.
Okay, second set of questions:
1) I plan on using 3V LEDs to light my structures. I have a multi-voltage transformer with a 3V setting.
1a) Should each LED have it's own run or can I wire them in parallel?
1b) How can I tell when the transformer has reached capacity short of the white puff of death.
1c) How long a run can I make using 22 guage wire?
2) I have an 800w power supply from a computer that delivers both 12V and 5V. Has anyone used one of these suckers to power layout accessories? The LEDs I'm looking at come with resistors for use with 12V. Seems it would be a well modulated power source.
2b) for Randy--would this work with your servos?
MouseWhite LEDs consume about 3.5V. But you need a current limiting resistor too. I recommend 5 Volts.The math works like this:5V (in) - 3.5V (LED) = 1.5V left over.
Let's say you wanted to run 10ma per LED (Pretty bright)1.5V = .010amps * R
R = 150 Ohms.
So a LED + one 150 Ohm Resistor will light a single LED.1a) LEDs don't work like incandecent lights. Every time you wire one into the circuit, there's a voltage drop. If you hook two into series you would need 3.5V + 3.5V = 7 V + a resistor drop. I would recommend wiring them in parallel.1b) using 10 ma for each 5V circuit you could literally run 100 @ 1 amp.
1c) depends on how much current you pull through it. Run the numbers through the voltage drop calculator to find out. Stay within 5% voltage loss. (.25Volts drop @ 5 Volts out)
2) There are tons of videos on youtube showing up to turn a computer power supply into a regular power supply. It's easy. But you'll need a sandbar resistor to keep the power supply from shutting off. (It's looking for a power good pin to be high on voltage) But for Safeties sake, divide it into blocks and put a 3 amp fuse on each block. Also if you are only using a little power <15% rated capacity, the efficiency of said power supplies drops off a cliff. That means you'll pay more for the same juice over a dedicated low power supply.For my layout, I use a 400 Watt computer power supply. But for my sons, I home built a small power converter that uses 3 L7812 chips to supply 3 amps of 12 V power and 3 L7805 chips to supply 3 amps to the 5V. I run it off a left over 19 Volt laptop power supply.
DigitalGriffinThe math works like this: 5V (in) - 3.5V (LED) = 1.5V left over. Let's say you wanted to run 10ma per LED (Pretty bright) 1.5V = .010amps * R R = 150 Ohms. So a LED + one 150 Ohm Resistor will light a single LED.
The MFR says that the LED is rated 3-3.1V. Should I still go with the 3.5V?
So let me see if I have this straight. My Rockipeutians would have been using candles, fireplaces, and lanterns for light, so 10ma would proably be too much. I'm guessing I'll have to experiment to taste, but lets say I go with half that 5 ma.
5v-3.5v=1.5v left over
plugging into Olm's
1.5v=.005a * R
R=300 ohms
Now, can I use one resister for the entire circiut or do I need one for each LED?
DigitalGriffin Also if you are only using a little power <15% rated capacity, the efficiency of said power supplies drops off a cliff.
So you think 800w to power 25 LEDs is a tad overkill?
I have 2 DC power supplies. I have a Bachmann unit that came with my EZ track Hogwarts set which I plan to use to power my turntable. I also have a Mech II I was hoping to sell to help fund my layout project. I also have the adjustable transformer that I could set to 5 (or 6) volts. I guess any of these make more sense than using my power supply.
Will the LEDs work on AC?
If not, do I stick my multimeter on the terminals of the rheostat and duct tape it when it hits 5V?
If your spec sheet says 3.1V then trust the spec. 3.5V is a rough average for white LEDsResistor for each LED. Each circuit is wired in parallel.800 watts its a tad bit on the heavy side unless you want 8000 LEDs
Be careful of those cheap bachmann supplies. Mine is rated for 16VDC, and it was outputting 25V! I had two fail right in a row.
LEDs will work with AC most of the time. However you may notice a 60Hz flicker as it turns on and off quick. Most LED christmas tree lights work this way and you'll never notice it unless you try to film it.
DigitalGriffinBe careful of those cheap bachmann supplies. Mine is rated for 16VDC, and it was outputting 25V! I had two fail right in a row.
I planned to use the DC output for the turntable. The motor is rated at 7V, so I thought I would find a good speed for the TT on the Bachman and leave it there, then use a DPDT switch to run it. Do you think that would be an unreliable method?
DigitalGriffinHowever you may notice a 60Hz flicker as it turns on and off quick.
Those Rockipeutians do love their strobe parties. I hear the red spotted muchrooms grow in that part of the country.