On my 4x8 N scale layout, I have 24 turnouts, 37 pieces of flex, and a few pieces of sectional track. I have 14 power feeds [one for every 3-4 pieces of any kind of track] . All joints are soldered using this method...
Solder Method. Join flextrack to switch/crossover with rail joiner. Use liquid flux applied to the outside of the joiners with a microbrush until wet [2 dabs is enough]. I then touch my 30Watt pencil tip iron with chisel tip attached [filed to match profile of rail web] to the inside of the rail. After 4 seconds, I touch the solder to the outside of the rail joiner. 1/8" to 1/4" length of solder is all that's needed. Get it to flow and then back off. Re-apply the solder tip to ensure that all flux has "boiled" out of the connection, and the connection is shiny. To finish each connection, I wipe the tip of the iron across the inside of the rail to ensure that it smooth and won't interfere with any wheels on my cars/locos. I've also found that during the wiping step, I can touch the iron to the top of the rail while touching solder to the outside of the rail web, and this will fill any small gaps in the rail ends where they meet. I then sand the connection with 80Grit sand paper and test it with at least 2 cars. Tedious, but pretty flawless. Another item of note regarding soldering is that the wiping method I described above results in better electrical continuity for each connection, which I check after sanding.
Happy Railroading!
After having some issues with track feeds, I have settled on soldering all rail joiners and putting feeders on every piece of track.
But I do enjoy soldering so............ and I live in SoCal so I do not worry about warpage - or at least i have not recognized it yet!
rrinker Not sure what table you used, but the one I found lists nickle silver at 9 microinches/in per degree fahrenheit. Which works out to 3 sections of flex track expanding 58 thou at a 60 degree temperature difference... --Randy
Not sure what table you used, but the one I found lists nickle silver at 9 microinches/in per degree fahrenheit. Which works out to 3 sections of flex track expanding 58 thou at a 60 degree temperature difference...
--Randy
Randy, I agree. Several years back, a materials engineer on another forum site did a calculation for Code 100 nickel-silver rail and found that a contiguous length 100' long would expand by a whopping 0.25" over that entire length with a rise of 30 deg F. Four 1/16" sliding gaps in that 100' will take care of the expansion nicely. Add another 30 degrees, and you need only one other set of gaps. Assuming the tracks were laid when the temps in the room were near 70 deg, 30 degrees in either direction is accommodated by eight small gaps over 100' of rails. Hardly a chore or worrisome.
But throw in wooden framework that is not sealed with an impermeable polymer...well, you have better keep a close eye on humidity because you will see expansions in the half-inch range over 10 feet of benchwork if the humidity rises from 50% - 90% [as a guess].
Crandell
The reply was great, as a former law enforcement officer I'll add "you can accomplish more with a kind word and a gun than a kind word alone."
Not sure what table you used, but the one I found lists nickle silver at 9 microinches/in per degree fahrenheit. Which works out to 3 sections of flex track expanding 58 thou at a 60 degree temperature difference.
Wood changes very little with temperature, but humidity is the killer here. If you have even humidity, either controlled or naturally even throughotu the year, there probably is no issue. I found a document that lists all sorts of wood properties, even electrical conductivity, btut he only moisture data is the typical shrinkage from green to kiln dried. That's 2-4%, but even the wood at the big box stores isn't green. And that doesn;t cover what happens to dried wood when the relative humidity goes up to 90% like it often does around here.
CHecking ont he Homasote web site, they list the expansion from 50 to 90% relative humidity as .25%. SO a 9 foot length of homasote under those conditions will expand by .27 inch, more than 1/4 inch. ANd presumably shrink by a similar amount, which is what happened at the old railroad club I belonged to. We had heat but no air conditionign in our building, and a large section of the layout was built in the summer, with all the doors open, in the typical 80+% humidity of the Northeast summer. That winter, with botht he outside air drier plus the heat on drying out the inside air even more, we ended up with kinked rails - and it wasn;t even soldered at all joints. The temperature was actually constant, or cooler in the winter - so the rail did not expand, if anythign it shrank. However, the wood and homasote DID shrink.
Modeling the Reading Railroad in the 1950's
Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.
It is to be remembered that the rail is layed at X temperature and at that time, that is you base temperature for all future discussions. If your NS track is layed at a rather cool 45 degees and soldered up in an attic in the eaves and then hits 105 on a hot summer's day when no AC is on, every 3 lengths of flex track will expand ~1/8" over its entire length! On a big layout with a wrap around main of 72 feet, that is a full one inch of growth! This is a real, calculable expansion of a known alloy metal. Few here will suffer such trials. I do, however.
Table frames and track tables of varying materials swelling or shrinking due to heat or moisture is effectively incalculable due to wood curing issues, wood sealing issues, differing materials, etc. All of this unknown stuff may counter, (act positively), to the metal track changes or add to ( make worse) the track problem at varying temperatures.
Track expansion and contraction is easily controlled with simple slide-gapping and insuring that the differential temperature between the day the track was first put down and your hottest or coldest day in the train table area is small. After you do this, the table and sub-base and other issues are on their own, but you have done the right thing so far as the track is concerned. In my attic, the differentials are easily 40+ degrees over the year and about 50 degrees over the intial X temp of laying the track.
Richard
If I can't fix it, I can fix it so it can't be fixed
That's another huge can of worms that comes up allt he time ont he Layout Construction group on Yahoo, but any slight lookign at information liek expansion coefficients makes it hugely obvious that it is NOT the rail expanding and contracting - unless your train room varies from extremes that are too cold for human habitation up to too hot for human habitation, with a brief period in the middle when it's OK to go in and play trains.
It is indeed the benchwork, and that depends on the material used, how it's fastened, and just what the environmental conditions are. Someone just ht eother day mentioned they have no issues but they live in an area that's the opposite of the Northeast - here we have cold dry winters and warm wet summers, so you have both heat and high humidity to expand the wood, and cold and dry to contract it. In his area, it's humid when it's cold, and dry when it's warm, so just when the termperature is making the wood expand, the humidity is makign it shrink. Balances out, apparantly.
Then there's people who use latex paint to paint the wood and think they've sealed it..not likely. I ha dno issues on either of my last two layouts, but both were built on extruded foam which had very little expansion from temperature and essentially none from humidity. Even though it's all held up with wood framework, nothign ever moved - the current layout is in a bedroom and somewhat regulated at least temperature-wise, but the previous one was in an unfinished basement and even with a dehumidifier runnign in summer and space heat3ers int he winter, the temperature and humidity varied greatly.
Even though I solder all my rail joints, I do make allowence for some expansion and contraction. As explained, I use DC and have signaling, so I have isolated track sections. Care is taken at insulated rail joints to allow for some expansion and contraction.
And, on exceptionally long sections fo track, expansion gaps are left unsoldered and jumpered around at track level - just like the prototype. But that would generally be a section of say twenty feet or more with one expansion joint in the middle. Once or twice in 40 years I may have had a 10 foot section of rail move enough to close up a gap, that's about it. Now I allways use insulated rail joiners.
I would suggest to you that on our model layouts it is not the rail that actually expands enough to cause problems, but it is the benchwork, expanding/contracting both from temperature and humidity changes. Wood expandes and contracts way more than metal - our layouts are generally built on wood frameworks.
A temperature/humidity stable invironment, choice of quality materials and carefull construction will limit such problems considerably. I use only premium Poplar framing and birch plywood for my layout. And have committed to the idea of building in modules that can move were ever I do in the future - no tearing down layouts and starting over.
Sheldon
My shelf layout is in a finished attic, as I noted. It is air conditioned via a large window mounted unit. However, when I am at work in the summer, the AC is not on until I come home, and only then, if I go upstairs to the layout and my man cave.
Accordingly, I do not solder my rail into one long mass. Expansion due to extreme temperature swings that can amount to as much as 60 or more degrees will cause the track to buckle upward and work the track nails/spikes loose over time. So, I put a gap in the unsoldered joiners at every 3 foot section of about .010 inches. Computation from the coefficient of linear expansion of the alloy shows this to be about right. Thus, all rail or at least each three foot section can expand and contract without any stresses developing. This would certainly not be a factor in a basement layout or especially inside the living space in a home or on a small pike where the temperature is controlled year around.
I have had two separate standard gauge HO (DC) pikes on this shelf system since 1980 and on the first one, I had rail lift and in one instance rail actually pop out of and break the the Atlas plastic fake spikes! Not a pretty sight.
Oddly, real narrow gauge roads had a heck of a time as the summer sun would beat down and heat the spindly 35lb to 55lb rails horribly and in winter the mountain snows could send temperatures sub zero on a regular basis. Track expansion and contraction was constantly working spikes loose and up or buckling sections of track that caused frequent minor derailments.
I hope to be starting construction on a shelf layout soon. I intend on soldering both rails on every piece of track and every necessary switch component. Not planning on soldering rail joiners, because they won't be relied on for power distribution. I figure a few minutes work per section will result in smoother operation down the road.
I've watched other people have enough problems with dead spots and weird glitches to leave things to chance.
Richard,
That is a fine solution for you, but hardly practical for most peoples layout goals.
Some of us actually plan layouts before we build them and are interested in different aspects of railroading than narrow gauge.
I too have an electrical and electronics background, but I don't personally use DCC at home. I do use, and have helped install DCC on a number of friends layouts.
I'm waiting for a really good direct radio system before I even consider installing decoders/recievers in my locos. And even that might not make me switch.
In the mean time, I am into exactly the opposite kind of railroading from you. My ATLANTIC CENTRAL is a large double track mainline Class I set in 1954. Detection, signalling and CTC are top operational requirements here.
I use a control system of my own design based on MZL control used by Ed Ravenscroft in the 1960's.
It intergrates signaling, detection, cab control, one button turnout route control, and CTC all in an intergrated system. Should DCC or direct radio be desired later it can be added "overtop" of the existing system with no problem.
Here are some photos of the controls for one section of my system as installed on another modelers layout:
It uses Aristo radio throttles and provides complete walk around control. These panels are prewired on the bench and have very minimal connections to be made under the layout.
And, I've been soldering all my rail joints for more than 40 years, I think I will stick with what works for me.
As for DCC, I have seen lots of DCC layouts without the "every three feet" feeders and they all run fine.
Every 10, 12, 15 feet, OK, that's close to what I do in DC.
And once small pieces of track are soldered, they are no longer small pieces of track.
My technique is somewhat unconventional. I have about 75 linear feet of layout on a 2 foot wide shelf that is, effectively, an oval around the slope of my entire attic roof. With 75 feet of run and only four very small towns and several mine only sidings between each, I can assemble on the fly as things please me or come to mind.
I never actually layout a layout on paper, per se. Instead, I figure where I might like the towns on the layout and lay my single main line track in 3 foot lengths. If the first township looks like it has to slip 4" or even 20 inches before that first in-town turnout to keep a 3 foot section, then that is the town's position. I hate tiny chunks of track in a mainline flow. Thus far, I have only 2 odd pieces of track in the main line to all turnouts off the main line. One piece is about 3 inches long and that one single piece is soldered to the three foot section before it is installed. The feeder goes to that now 39 inch piece at 191/2 inches The other odd piece is about 19" long and it has its own feeder off the buss going to it at dead center with no joiner soldering at either end.
I did note earlier that I rarely solder a joiner. So far, there is only one odd solder joint and the main line is close to being finished with the bulk of the switches off the main in place. Everything else is either a passing siding, all using one or two pieces of 36" flex, switch to switch or a spur which can be any length, but all spurs are over 15 inches long but none longer than 36 inches and, thus, all get a center feeder off the gapped switch.
Narrow gauge trains of 6- 30 foot cars plus caboose is considered a very long train with nothing but light 1882 C-19 consolidations pulling stuff. Narrow gauge never had two track mains, of course.
I would much rather save my fight for scenery than track. As an electronics engineer by profession we always like to connect directly from the rail to the buss in the center. All soldering of all feeder wires is done before the 3 foot section goes down permanently on the road bed for inspection and paint covering purposes. everyone has their own way of doing things. For my part, I find that about 100% of all DCC comm issues are dirty wheels, dirty track or dirty transfer contacts off wheels, since I know my track electrical hookup will never be an issue.
narrow gauge nuclear To answer a question posed, perhaps to me; My smallest isolated track section on a spur or siding is probably 15 inches. As this short section goes to a switch that has insulated joiners, then yes, I do add a feeder off the 12 gauge DCC buss to the center of that small piece of track. I do not have 3, 5 or even 10 inch spur shorty pieces on my layout as no car and engine would fit on such a small piece. In general, I do not solder rail joiners because of my policy of hitting the center of all 3 foot flex track pieces with buss feeders, frees me from this practice. Richard
To answer a question posed, perhaps to me; My smallest isolated track section on a spur or siding is probably 15 inches. As this short section goes to a switch that has insulated joiners, then yes, I do add a feeder off the 12 gauge DCC buss to the center of that small piece of track.
I do not have 3, 5 or even 10 inch spur shorty pieces on my layout as no car and engine would fit on such a small piece. In general, I do not solder rail joiners because of my policy of hitting the center of all 3 foot flex track pieces with buss feeders, frees me from this practice.
Richard, you missed my point. Surely your "mainline" is made of equal multiples of three feet?
Train leaves turnout #1 there is 42" of track between it and turnout #2 - somewhere in there I see a piece of track 6" long? Or two three foot pieces wastefully cut into 21" pieces?
Since I solder my rail joints, its a non issue - what do you do? another feeder for the 6" piece? or 21" piece?
Depends on how short. I needed a filler section to reach between two pieces of flex, right where I planned to gap the rails between power districs. My solution was to solder a short filelr section from an Atlas assortment to each adjoining piece of flex, effectively making a section fo flex track 3 feet 2 inches long, and putting the insulating joiners betwen the short pieces. Thus maintaining ever section of flex being fed from each end. I keep waiting for my layout to fail like everyone seems to insist it will since I feed power through the joiners, but it hasn't yet, despite multiple coats of paint, alcohol spray, and diluted glue getting all over the rails and joiners.
narrow gauge nuclear I can sort of see that if you have a large pike or pre-exisitng mega empire in DC and decide to go DCC that hooking to the center of every 3' section would be a task that would seem a never ending adventure. Plus, on many long built roads, scenery and very unhandy locations would abound such that boring the small holes would be a task, and in some instances, doing a decent soldering job might be nearly impossible. So if you have a pre-existing road in DC and have no real running problems in DC then whatever seems to work so far as buss tap feeders is concerned is certainly fine. If problems should crop up, you can address them individually. However, for all new builds in DCC, it would be rather silly not to just wear the belt and suspenders by tying into the center of each 3 foot piece. No need to solder the joiners. I am lucky as I build in narrow gauge. A single track main only. No rail yard at all with the biggest yard being a three strip siding group. All depots have only a single siding off the main and only one or two 8 car length passing sidings on the entire road. No turntable, just two wyes are planned Richard
I can sort of see that if you have a large pike or pre-exisitng mega empire in DC and decide to go DCC that hooking to the center of every 3' section would be a task that would seem a never ending adventure. Plus, on many long built roads, scenery and very unhandy locations would abound such that boring the small holes would be a task, and in some instances, doing a decent soldering job might be nearly impossible.
So if you have a pre-existing road in DC and have no real running problems in DC then whatever seems to work so far as buss tap feeders is concerned is certainly fine. If problems should crop up, you can address them individually.
However, for all new builds in DCC, it would be rather silly not to just wear the belt and suspenders by tying into the center of each 3 foot piece. No need to solder the joiners.
I am lucky as I build in narrow gauge. A single track main only. No rail yard at all with the biggest yard being a three strip siding group. All depots have only a single siding off the main and only one or two 8 car length passing sidings on the entire road. No turntable, just two wyes are planned
Tell me, what do you do with all the short, 8", 14", 3",etc, etc, pieces of track that are created as you spot turnouts in their desired locations? Do you feed each of them seperately?
Randy,
I know that what I do is EXTREMELY complex, and requires way more wires than even the most "overkill" power fed DCC layout -
BUT, my wiring is complex because signaling, detection, signal interlocking, cab control, colision avoidence, turnout control, CTC and wireless radio throttles are all intergated into one infrastructure.
Many of which are features that require extensive additional infrastructure with DCC.
I agree that any system needs a distributed power system, and I believe DCC should have lots of power districts and sub districts on seperate circuit breakers - because 5 or more amps at 16 volts can be dangerous.
My whole system is "distributed" around the layout to minimize length of wire runs - the throttle bus for the mainline throttles being the longest and only "layout wide" wiring.
BUT the "every three feet or less", "every piece of rail" thing just cracks me up.
And to me, it is the same work to hook up 50 pair of wires to same the bus or 50 pairs of wires to 50 control points on relay boards. The "thinking" work is not done under the layout on my DC system any more than it is with a DCC system - its called planning.
Still soldering all my rail joints in each electrial "section".
Like I said, I put power to all my feeder drops. Work? Guess it depends onw hat you call work. It's pretty easy and efforltess, really.
Like I ALSO said, on both layouts now, my initial tests were done by hooking up exactly one pair of wires. Two wires. Things ran great, no slowdowns, nothing. I STILL don;t have all the feeders hooked up on my current layout, and there are no dead or slow spots. I only recently got around to connecting a few ont he farthest side of the layout, mainly because I want to get it all done before attaching the fascia panels.
Even for a DCC layotu with signalling, if you are smart and use a distrubted block deection system instead of one that requires you to run all the wires back to some central point, there will always be less wiring than for any sort of DC system that allows multiple train operation. MZL is only 'simpler' if you decide not to treat the DCC command station and decoders are the black boxes they are. If you consider the electronics inside - then yes, it is WAY beyond switches, contacts on switch motors, and maybe a few relays. But looking at just the sheer amount of wiring - there's far less in DCC, even if you do put feeders every 3 feet. It's one pair of wires, around the layout, with short feeders connected to it. Even allowing for block detection, it needn't get any more complex than that. And if that is complicated - you picked the wrong hobby, perhaps static models would be a better choice.
I really haven't seen anyone serious about DCC say "Oh yeah, just 2 wires". Train sets with loops of sectional track don;t work very well with just two wires, and no one claims DCC is some magic that can fix this. In fact more often than not the actual saying is, if it doesn't work well on DC, it's not going to work on DCC either. Which truth be told is rather on the overly conservative side because unless you are using DC with things like PWM and BEMF, you CAN improve runnng performance by going to DCC, to a certain extent. You can't cover up severe faults in the mechanism but the little ones can be smoothed over particularly with BEMF
selector And if you ballast, you don't even need the joiners; the ballast should do an excellent job for alignment. I maintain that soldering a feeder at the junction of a pair of flext track lengths, say to a joiner, covers a multitude of problems, including limiting the work of laying and powering track reliably. The proof should be in the utility over time, and so far that has worked for me.
And if you ballast, you don't even need the joiners; the ballast should do an excellent job for alignment.
I maintain that soldering a feeder at the junction of a pair of flext track lengths, say to a joiner, covers a multitude of problems, including limiting the work of laying and powering track reliably. The proof should be in the utility over time, and so far that has worked for me.
Crandell,
Your approach is the most common sense I have read since this thread began. I realize over kill never hurt anything, but DCC or not, every three feet is overkill. Your every six feet is still likely more than needed, but seems much more practical from a labor standpoint.
I know I will be scolded for this, but once again, it all sounds like a lot more than two wires to me.
I refer all who are offended to my original post in this thread where I discribed a friends layout, originally built DC, now converted to DCC, still only fed with one drop to what was eash of his "blocks", typically 12' to 20' in length - his layout runs flawlessly on DCC, just like it did on DC.
I have a feeder at just about every section of track or every other joint.
Dennis Blank Jr.
CEO,COO,CFO,CMO,Bossman,Slavedriver,Engineer,Trackforeman,Grunt. Birdsboro & Reading Railroad
If you run DCC and not DC and if you have a heavy gauge buss wire pair running under the table as DCC recommends and if you connect to the center of each 3 foot section of flex track, every switch and every smaller, isolated piece of track, then rail joiners are absolutely superfluous in any electrical sense. They will serve only to mechanically align the rail ends. There would be no need to solder them at all unless you are doing so for curve stability.
Funny how I've never heard anyone say " I have a problem because I put too many feeder wires in". However I have heard of some who soldered all their joints and it buckled in the heat of summer. Depends on the climate/environment where you live. Also where you have your layout, like perhaps in the attic with no aircondioning in a Cape Cod style house. Could cause a problem I suppose!
Jim
If you skimp on not soldering feed wire to every piece of rail or at least solder some connections on rail you don't feed you will be sorry down the road. Where I didn't feed wire, I solder the butt joints using about a 1/4" piece of bare wire accross the joints, on the outside of course. Corrosion at the joints will cause it to fail if you only use rail connectors and especially after ballasting. I have to leave some gaps on my rails because of temperature changes. I didn't think I would, but what a difference from winter to summer I see in the gaps that I made for expansion and contraction. Glad I did that! We don't, at least I don't, like to go back and do it right after all the scenery and ballast are done, then find out we have a power loss on sections of track 5 years from now! NO FUN in that because we were in a hurry or skimped on it. Track laying, aligning switch's, and wiring are the most critically important fundamentals of trouble free performance, be a perfectionist when doing track and wiring! Jim
I also have soldered every rail joiner and in 25 years on this layout have never had a problem. My drops are about every 10 feetish connected to a 12 g buss, also no problem. We often run 10 to 15 locos (in several MU lashups) about half with sound all powered by an NCE 5amp system.
As far as removing turnouts, it's quite simple and no more of pia than unsoldered ones. You just take a dremel and cut the turnout out at the rail joiners. With a nice hooked dental tool, it's just a matter of heating the ends of the rails and pulling off the leftover joiners. Trim the new turnout to fit the space, slip on some joiners, slip it in place and resolder the joiners. It took me longer to write this than it takes to do it. I've only had to replace one bad turnout on my layout in 25 years, but I've had to move some due to design changes. All the ones I've used were built to last.
Jay
C-415 Build: https://imageshack.com/a/tShC/1
Other builds: https://imageshack.com/my/albums
Paul3 I would not solder every rail joiner. Expansion and contraction of the layout alone should cause second thoughts on that. Not to mention the pain in the rear part of trying to remove & replace switches, diamonds, and other pieces of track. Soldering on the curves is one thing, but every rail joiner? You are asking for trouble if you do. Paul A. Cutler III
I would not solder every rail joiner. Expansion and contraction of the layout alone should cause second thoughts on that. Not to mention the pain in the rear part of trying to remove & replace switches, diamonds, and other pieces of track. Soldering on the curves is one thing, but every rail joiner? You are asking for trouble if you do.
Paul A. Cutler III
I have never had a problem in over 40 years of soldering all my rail joints in each electrically isolated track section, typically 12 to 20 feet in length. And now days I always use plastic rail joiners for section gaps, not just gaps in the rail. I may "be asking for trouble", but it has not shown up yet.
Dropping feeders every 3 feet? Definite overkill. I drop feeders every 9 feet and have no issues related to that. I would recommend that one drop jumpers across every rail joint so that every piece of rail has a wired or soldered connection going to it.
One thing that should also be brought up is redundantcy. IMHO, soldering a feeder to every piece of rail, while overkill, is still a single point failure system. IOW, any one failure of solder joint or bus connection will knock out your track power. One would be better served by jumping every rail joint with soldered wire connections and adding feeders every so often. That way, if you have one solder joint failure, you still have power to the rail.