Wiring a Helix for DCC

[rrinker]

ATLANTIC CENTRAL

Water pipes are near electric lines all the time.

Yes the meter is read by transponder, either local or satellite.

If their smart meter system is fully on line they can even turn you off and on remotely.

Sheldon

 

 Not to get on the whole codes thing again, but my old house, the water came in, and the water heater was located, in the same closet with the main panel. This was OK'd by the inspector whenthe house was built, some 10 years before I bought it. I had the electric service upgraded, so the electricians doing the work added a second panel actually slightly FURTHER from the old panel, but the main disconnect was right next to the old panel (on the side AWAY from the water heater. Inspector made them rip it out, too close to the water heater, I ended up with an outside disconnect (got to love that - so anyone could walk up behind the house and cut my power - good thing the alarm system had battery backup).

 Then again, he IS in Florida, and this was in PA. I know my water and electric are no where near each other in my current house - the electric comes in at the garage, the water comes in on the opposite end of the house. Even gas and water aren't all that close - water comes in on the left front, gas comes in halfway back the left side (~12 feet). Which is a bit of a pain when it comes to using the space for a layout, electric as way at the opposite end and out of the way, gas and water, thus water heater and furnace, are right in the middle of the train space.

                                 --Randy

 

[Doughless]

railandsail

Probably another good reason for 2 feeders per loop would be the capability to provide some expansion joints in the track of the helix.
I thought this made sense,...
"Anyway, a Dremel tool and some wire to jump the gaps solved the issue with no further subsequent problems in over 20 years."

 

Considering your helix is outside of the main shed, it may be subject to expansion more than the normal helix (maybe construction materials will limit seasonal expansion issues) Perhaps the amount of expansion joints will dictate how many feeders you want rather than voltage drop.

If you use unsoldered rail joiners to keep gapped track together (why use a dremel when track comes in pieces anyway?), you'll need feeders for every section, with each section represented by several soldered pieces of flex track.  In your helix, I'd think you'd want a few more expansion joints than most and each section would have fewer pieces of flex track because of it.

[ATLANTIC CENTRAL]

Water pipes are near electric lines all the time.

Yes the meter is read by transponder, either local or satellite.

If their smart meter system is fully on line they can even turn you off and on remotely.

Sheldon

[railandsail]

How about the fact that my water service line was right there in that same bock of cement that the power pole was attached to.....water and electric right next to one another??

I asked the power company to come have a look, and they ignored it.

 

BTW, a question. They read these meters remotely, does that mean by some feedback service in the meter, or from satellite, or what??

[ATLANTIC CENTRAL]

Brain,

I don't want to be a trouble maker, but I don't think the power company is going to be too happy if they, or an electrician have to come work on your electric service.

Typical power company guidelines, and the NEC, require 3' of clear space directly in front of all panel boxes and meter cans........

I've worked in that trade off and on for 40 years, I would not want to work on that meter can hot with your metal frame helix "building" there in the way......

And I've done more than my share of "hot" service work.......

Sheldon 

[railandsail]

BMMECNYC

 

Just a quick question for the Railandsail:

Do you have field mice issues where you live?  Ive seen them chew through lots of electrical wiring (in cars for some reason).  Not sure if you have considered rodent access prevention for the helix.

Also could be a source of a broken feeder, although a mouse nest on the locomotive pilot would probably be a good clue that there is a problem.

 

@ BMMECNYC

I don't believe I have any field mice in my area,...and I also have a neighbor with a cat that loves to hang around my place. I welcome him for pest control.

 

I do have some plans to keep most critters out, but those pesky mosquitoes. and mud dabbers may need extra care.

And those geckos, they seem to get into ANYTHING.

 

BTW I got the helix frame structure hung today.

Helix is Hanging on its Own,...finally

Okay its bolted to the rear of the shed on one end, and standing on two poles on the other. Looks gooood, and its rigid.

I'm very pleased with the access gap height which is larger than I first surmised. I had not included the difference between the height off the floor of the main shed as compared to the extra 6 inches I got by the new height off the ground on which the shed sits. Pretty easy to get into, especially when I lay a piece of foam pad on that concrete pad.

One of the photos shows a chair and a step ladder inside the loop as I was bolting things up. With the roof on it will be tall enough for me to stand in (I'm 6'4")

I think my access to that 3-way turnout that will distribute trains to the staging tracks is also going to be of little concern.

 

 

 

 

[ATLANTIC CENTRAL]

I'm still reading and smiling.......I gave my opinion early on.

But twenty years ago, they told me if I switched to DCC, I would only need two wires?

That was obviously not true.

Feeder bus and drops every six feet represents more connections to the track than I have with my complex Advanced Cab Control DC system. And trust me it is complex........

Actually pretty glad I did not switch.........

What a wiring nightmare if you want to do detection and signaling like me?

Three or more feeder drops per my typical signal block that then need to go thru a detector of some sort, then get tied to the bus.

Then divided into power districts on circuit breakers. 

Actually, I have helped with and seen these setups on large DCC layouts.

Not much less wire than my DC system which includes:

Wireless throttles

Progressive Pushbutton Cab Control and One Buttom Route Control of interlockings

Dispatcher controlled CTC with leaves engineers to only control speed and direction (kind of like DCC)

Safety Interlockings like the prototype - turnouts will not throw when the train is in the interlocking limits

ATC - automatic train control - if an engineer overruns his block, runs red home signal - his train simply stops.......

Obviously - Signals

All for just a little more wire than all these drops and bus wire, several hundred ice cube relays (really inexpensive), a few custom circuit boards, a little planning, and no more money than typical boosters, circuit breakers and throttles for a similar sized DCC layout without signaling or CTC......

At a cost well below what it would take to do DCC, signaling and CTC by other means. Not to mention decoders for 135 locos.........

OK, here is what I don't have:

Correct, I can't run two engines independently in a given signal block. My track plan is designed so that is not really an issue.

I can't/don't use sound - dual mode decoders do not work with my Aristo Train Engineer wireles radio thorttles. There are no decoders inany of my locos. But I don't like onboard sound. It sounds like a nine transistor radio stuck between stations.....

I can't turn headlights on and off - BUT, my constant brightness headlights come on before my trains move. And I model a time before ditch lights and when day time running lights were just becoming the norm.

OK,

I don't usually talk so much about my control system anymore, I understand everyone loves their DCC, and for many people DCC is a great choice.

But this conversation has gone over the top, so I thought some over the top alternate view would be interesting - at least to some.

To the OP,

Brian,

I can appreciate that you are trying to build and much layout in your small space as possible. I wish you the best.

However, based on my experiance, you are headed for a lot of problems despite all this research and debate you are conducting here on the interweb.

You are proposing to build a complex layout in a small space in a very bad invironment - it would scare me - but I have a low risk tollerance.........

Again, my advice - learn to solder rail joints, and learn to do it well........

Sheldon 

 

[Anonymous]

gregc

 I don't understand he reasoning for track feeders that requires extra time and effort. it doesn't make sense (at least to me) to solder rails and then add feeders assuming some joints might break.

Soldering rail joints has two purposes

1) ensuring electrical contact

2) ensuring physical alignment

gregc

allow expansion and determine the distance between feeders based on voltage drop.

Because calculating voltage drop alone does not ensure that in the instance of a fault, the cumulative resistance generated by all factors will trip the circuit breaker.  Voltage drop is only part of the story (already discussed).  

Just a quick question for the Railandsail:

Do you have field mice issues where you live?  Ive seen them chew through lots of electrical wiring (in cars for some reason).  Not sure if you have considered rodent access prevention for the helix.

Also could be a source of a broken feeder, although a mouse nest on the locomotive pilot would probably be a good clue that there is a problem.

[Cymrych79]

gregc

of course you're free to do what you feel is best.   But is the extra time and effort necessary?

I've been taking a different take on this entire thread, but purely based on how I've wired my layout. If the OP goes with 1 feeder per level in the helix, that represents 17'. Call that 6 pieces of 36" flex. That's three times the max distance I wired feeders on my layout, going every other soldered joint. My figuring was, every piece of track had a wire feeding it. Overkill? Maybe. But my trains run, so I'm happy.

So, my thoughts on wiring feeders in a helix, often a covered and difficult area to work on after construction and scenery, are basic: can the OP get by with one or two sets of feeders? Sure, most likely. The math definitely seems to support that later, and probably supports the former.

But what constitutes overkill? After all, it's a hidden, hard to work on area after its assembled. So why would I switch from something I know damn well works, like every other join? Personally, I'd go every other join, or 3 per level. A tiny, tiny bit of extra work soldering 2 extra feed wires per level, but then I'd feel comfortable never having to think about it again.

It's not really math based, but more of the "if it ain't broke, don't fix it" idea. Just my $0.02.

[gregc]

posting the Mark Gurries pages was pertinent and interesting.  thanks

i understand looking at bus length in terms of voltage drop.   I don't understand he reasoning for track feeders that requires extra time and effort.

it doesn't make sense (at least to me) to solder rails and then add feeders assuming some joints might break.   If that's the concern, then why not solder a (22g) wire across the gap, using unsoldered rail joiners to allow expansion and determine the distance between feeders based on voltage drop.

BMMECNYC

I think there probably is a dead horse here somewhere, or we have stumbled onto the path of a philosophical wiring discussion.

i'm not fond of doing things without sound reason (anecdotal engineering).     I've had to re-implement things that were "thoroughly tested" or could be done "no other way".

of course you're free to do what you feel is best.   But is the extra time and effort necessary?

[Anonymous]

gregc

 

 

BMMECNYC

gregc

the voltage drop is 0.14V at 1A.   Even if this were tripled, would it be excessive?

https://sites.google.com/site/markgurries/home/dcc-general-best-practices/wiring-planing/max-track-bus-lenght

 

 

 

 

he tabulates values for 5 and 10% voltage drop.   0.14V is 1%.   2% assuming an equal drop in 14g bus.

 

or was there some specific section of the web page you were referring to.

 

 

 

Nope, that was it.  I really dont think voltage drop is the basis, well, unless you count the possibility of one or more unsoldered rail joiners loosing contact, causing high resistance joint or no contact. 

I think there probably is a dead horse here somewhere, or we have stumbled onto the path of a philosophical wiring discussion.

[gregc]

BMMECNYC

gregc

the voltage drop is 0.14V at 1A.   Even if this were tripled, would it be excessive?

https://sites.google.com/site/markgurries/home/dcc-general-best-practices/wiring-planing/max-track-bus-lenght

 

 

he tabulates values for 5 and 10% voltage drop.   0.14V is 1%.   2% assuming an equal drop in 14g bus.

 

or was there some specific section of the web page you were referring to.

 

 

[Anonymous]

gregc

by every other rail joiner, doesn't he mean that every piece of rail (2 rails / section) is connected to a railjoiner with a feeder? 

I think he means (based on how he worded it) every piece of rail gets a wire soldered to it, because Railjoiners do not provide long term, reliable electrical contact.

gregc

  but then he said (what you referred to) a feeder every 9'   (is he assuming a solder joint may break)?

Yes, he is assuming that the possibility exists that a soldered rail joiner or feeder joint may break at some point (inadequate solder applied, stress due to external forces).

I have not actually had this happen except at a joint between code 100/83.

gregc

the voltage drop is 0.14V at 1A.   Even if this were tripled, would it be excessive?

https://sites.google.com/site/markgurries/home/dcc-general-best-practices/wiring-planing/max-track-bus-lenght

[railandsail]

Probably another good reason for 2 feeders per loop would be the capability to provide some expansion joints in the track of the helix.
I thought this made sense,...
"Anyway, a Dremel tool and some wire to jump the gaps solved the issue with no further subsequent problems in over 20 years."

[railandsail]

So it sounds like the one feeder per loop would be adeguate, and 2 will certain cover the situation.

Certainly I do NOT have to consider feeders for every 3 foot section of track (which has been suggested by some folks)

[gregc]

good.  at least this is guidance from someone well respected in the field.    I won't argue with this, but i'd like to see his reasoning (numbers).

by every other rail joiner, doesn't he mean that every piece of rail (2 rails / section) is connected to a railjoiner with a feeder? 

1) EVERY SECTION OF FLEX TRACK SHALL GET ITS OWN PAIR OF TRACK FEEDERS.

  but then he said (what you referred to) a feeder every 9'   (is he assuming a solder joint may break)?

2) THERE SHALL NOT BE MORE THAN TWO SOLDERED RAIL JOINTS BETWEEN SAME RAIL TRACK FEEDERS.

assuming that nickle silver rail has the same resistance as 22g wire (0.016 ohm/ft) or 0.14 Ohm / 8.5 ft, but a feeder every 8.5 ft resulting in half the resistance in each path, but there are two paths, the voltage drop is 0.14V at 1A.   Even if this were tripled, would it be excessive?

this is the equivalent voltage drop in 54 ft (x 6.4) of 14g wire.

[Anonymous]

gregc

so how much overkill is necessary?

 

 

BMMECNYC

Also this:

https://sites.google.com/site/markgurries/home/dcc-general-best-practices/wiring-planing

 

 

 

https://sites.google.com/site/markgurries/home/dcc-general-best-practices/track-feeders-frogs/soldered-railjoiners-and-feeders

This says every piece of flex track or every other if the railjoiners on both sides are soldered.

So three pieces of flex track has two feeders.  If you solder one near the beginning of one piece of flex track and the other near the end of the third, that comes out to about every 8-9 feet you have a feeder.  Or more clearly stated, two per loop of helix the previously described 17' circumference helix.

 

[gregc]

so how much overkill is necessary?

BMMECNYC

Also this:

https://sites.google.com/site/markgurries/home/dcc-general-best-practices/wiring-planing

[rrinker]

BMMECNYC

 

 

gregc

 

 

BMMECNYC

I guess it might be also that there is resistance at each conection that is not accounted for in your calculation (wire to booster, wire to smaller wire, small wire to rail).

 

 

he said the booster outputs 14.2V.    A resistance of 2.84 Ohm results in 5A.

if there is 0.61 ohm of resistance in 50 ft of 14g and 12 ft of 26g, you're suggesting that the majority of the resistance is in the connections (2.2 Ohm = 2.8 - 0.6), not in the wire.

 

 

 

 

And now that I read Randy's response... and Mark Gurries' explanation:

The short itself is a resistive load, in addition to all of the above things.  

The only thing we have control over at the time of the short is the resistance of the wire that is installed.  The rest (wheel contact, rail cleanliness, locomotive internal wiring) is out of our control.  

Which is why people suggest what seems like overkill on wiring and feeders. 

 

 Exactly.

                       --Randy

 

[gregc]

BMMECNYC

Two sets of feeders 180 degrees out should be adequate to get a sat quarter test.

i believe i've shown it to be unnecessary.

[Anonymous]

gregc

 

 

BMMECNYC

I guess it might be also that there is resistance at each conection that is not accounted for in your calculation (wire to booster, wire to smaller wire, small wire to rail).

 

 

he said the booster outputs 14.2V.    A resistance of 2.84 Ohm results in 5A.

if there is 0.61 ohm of resistance in 50 ft of 14g and 12 ft of 26g, you're suggesting that the majority of the resistance is in the connections (2.2 Ohm = 2.8 - 0.6), not in the wire.

 

And now that I read Randy's response... and Mark Gurries' explanation:

The short itself is a resistive load, in addition to all of the above things.  

The only thing we have control over at the time of the short is the resistance of the wire that is installed.  The rest (wheel contact, rail cleanliness, locomotive internal wiring) is out of our control.  

Which is why people suggest what seems like overkill on wiring and feeders. 

[Anonymous]

BMMECNYC

 

 

gregc

 

 

rrinker

 Did you measure the voltage at the far end? If the resistance was 5.8 ohms, with a 1 amp load on the far end the voltage should drop by 5.8 volts.

 

 

i did a quick experiment to verify the numbers I calculated from the Great Feeder Experiement website.

i went to an extreme and used 47 54 ft of 30g wire for the entire run to verify the resistance and see if a short could be detected

 

 

 

 

I think we have arrived at this:

https://sites.google.com/site/markgurries/home/dcc-general-best-practices/wiring-planing/power-wiring/wire-current-ratings

 

 

we have no idea how clean the track was that was used to perform the "Great Feeder experiment"  and we dont know what was used to generate the short circuit. 

Suffice to say, perform a quarter test, if feeders are adequate, the booster or circuit breaker will trip.

Circling back around the helix to the original question at hand.

Two sets of feeders 180 degrees out should be adequate to get a sat quarter test.

One set may or may not be able to trip the booster.

Running 14GA Bus should be more than adequate.

You can get away with small feeder sizes up to some debatable length.  This will depend on your soldering skills and over time corrosion of the connections.

In the case of the OP, he could probably solder the 14GA right to the rail if it doesnt affect the operability of the track, given that all the helix track will never be seen by anyone but him.

[Anonymous]

gregc

 

 

rrinker

 Did you measure the voltage at the far end? If the resistance was 5.8 ohms, with a 1 amp load on the far end the voltage should drop by 5.8 volts.

 

 

i did a quick experiment to verify the numbers I calculated from the Great Feeder Experiement website.

i went to an extreme and used 47 54 ft of 30g wire for the entire run to verify the resistance and see if a short could be detected

 

I think we have arrived at this:

https://sites.google.com/site/markgurries/home/dcc-general-best-practices/wiring-planing/power-wiring/wire-current-ratings

[gregc]

rrinker

 Did you measure the voltage at the far end? If the resistance was 5.8 ohms, with a 1 amp load on the far end the voltage should drop by 5.8 volts.

i did a quick experiment to verify the numbers I calculated from the Great Feeder Experiement website.

i went to an extreme and used 47 54 ft of 30g wire for the entire run to verify the resistance and see if a short could be detected

[gregc]

BMMECNYC

I guess it might be also that there is resistance at each conection that is not accounted for in your calculation (wire to booster, wire to smaller wire, small wire to rail).

he said the booster outputs 14.2V.    A resistance of 2.84 Ohm results in 5A.

if there is 0.61 ohm of resistance in 50 ft of 14g and 12 ft of 26g, you're suggesting that the majority of the resistance is in the connections (2.2 Ohm = 2.8 - 0.6), not in the wire.

[Anonymous]

This calculator https://www.cirris.com/learning-center/calculators/133-wire-resistance-calculator-table

 

says that 26 AWG wire 12ft long is .49 ohms +.126 ohms = .616 ohms of wire resistance.

Rail resistance: https://www.w8ji.com/track_resistance.htm

 

0.00145 ohm per inch code 100

.00145 x 70" (36" piece of flex track, soldered connections about half inch from end of rail, plus quarter would not be exacty at end of rail)

.1015 ohm

.616 ohm + .1015 ohm = .7175 ohm

+2 unknowns:

resistance of connections, at least one of the three is a screw terminal

resistance of the 5/8ths inch piece of material used to generate the short circuit.

 https://dccwiki.com/Rail_Size

this gives slightly different data for rail resistance.

 

[rrinker]

gregc

 

 

carl425

With feeders too far apart a short will not trip the breaker.

See "The Great Feeder Experiment" at http://www.wiringfordcc.com/track.htm

 

 

i looked at The Great Feeder Experiment and have a hard time believing the results based on the description.

My understanding is he connected a 5A booster to a piece of 3ft flex track through

• 25' of 14g
• feeders of various gauge (18-26) and lengths (0.5 - 6ft)

am i correct?

based on wire resistance reported in the American Wire Gauge, i calculate the worst case wire resistance of 0.616 ohm for the 6 ft 26g case, total resistance for both paths.

if the booster supplies 15V, a short should result in 24A.

a more simplistic case is that 31ft of 26g wire has a resistance of 1.3 ohms or 2.6 ohms.

what am i missing?

i strung 54 ft of 30g wire, measured 5.8 ohm (5.6 according to the wire table) and 1+A bulb on my PowerCab lit up.

 

 Did you measure the voltage at the far end? If the resistance was 5.8 ohms, with a 1 amp load on the far end the voltage should drop by 5.8 volts. FAR too much, that's more than 1/3, the equivalent of running the train at 2/3 speed vs full throttle, and definitely noticeable. That's why we don;t use long runs of #30 wire for powering the layout.

 It all depends on how the layout is used. If you only ever run one train by yourself, you'll probably never have even a 1 amp load. Lots of operatores, depends on how the layout is divided and if there are bottleneck points that are all the same power districts, you can easily exceed 1 or 2 amps of load in a single spot.

 Note sure where that 24 amps comes from. If the wire resistance id 5.8 ohms, then a dead short is 15v/5.8 or 2.6 amps. A 3 amp capable system wouldn't trip. And you are assuming a 0 ohm short. Even clipping a pair of leads to the wires int's really 0 ohms. If you set a quarter on the rails, NOT press down, which is the proper way to do the test, say you had 4 ohms of resistance and ZERO ohms in the wire, so absolutely no loss, ideal wire, ideal track. A 4 ohm load at 15 volts is 3.75 amps. Not enough to trip a 5 amp system. But 3.75 amps through a 4 ohm load is nearly 60 watts of heat. A 60 watt light bulb easily melts plastics. THAT'S why you use heavier than you might need wire, and use circuit breakers instead of feeding 5, 8 or 10 amps right to the rails, at least in smaller scales. You're unlikely even on a busy layout to have the 6-12 (or even more) locos it would take to draw 3 amps, but even that relatively high resistance short would exceed the 3 amp setting and cause the power to cut off.

                                 --Randy

 

 

[Anonymous]

Greg,

I guess it might be also that there is resistance at each conection that is not accounted for in your calculation (wire to booster, wire to smaller wire, small wire to rail).

 

[Anonymous]

gregc

 

 

BMMECNYC

 

gregc

if the booster supplies 15V, a short should result in 24A.

Where are you getting that number?

 

 

 

24 A = 15 V / 0.61 Ohm

 

Is the booster in question actually capable of suppling 24A to the short in question?

Did you calculate the resistance of the track? There is an additional 6 (ish) feet of rail involved in the test.

Also I think the AWG chart assumes pure copper wire, which most (smaller) wire you buy is not (this is also stated on the wiring for dcc website somewhere)

 http://www.wiringfordcc.com/trakwire.htm