modeling 12:12 rail joints

[fwright]

rioplumas

Thanks to all for your input.

I would like to model WP c. 1954 (Feather River). I have not checked actual rail lengths for that road that era so I went with a general question based on general rail length of that period.Today UP today uses welded rail and concrete ties except at switches-also a nice look. I'm more interested in the look of the rail being right rather than the length being 33 feet or 39 feet--as I said, 6 inches (or about 43 1/2 feet) would not be a problem with me.

I don't know whether you want steam or diesel power or both.  IMHO, 1954 was a very good year.  Diesel is much, much easier to model in P87.  NWSL and others have replacement wheel sets, and many model diesel drives have enough "slop" to provide the requisite equalization for P87 operation.  Cars were still a good length to permit reasonable curve radii in P87. 

This would be more of a diorama than a layout, a vignette, so realistic rail details is a consideration. I would hope that this could grow some day to be a piece in something than can operate. The look would include tie plates, fishplates, and staggered rails. After seeing the CVT and Proto 87 web sites, it appears possible.The switch details look great.

Agreed.  Note that the details are available for both NMRA-spec and P87-spec turnouts.

I like the idea of not using rail joiners, or using half length ones. I have never heard of someone not using rail joiners, so that is new to me.

Jack Work suggested it in his April 1963 Model Railroader article, "Birth of a Turnout".  The article was my guide for hand laying turnouts, and I followed his advice on rail joiners without a qualm.  I did learn that it was much easier to properly align rail joints on curves without rail joiners if I pre-curved the rail and snipped the last 1/2" or so that I couldn't curve properly.  But not using rail joiners allowed me not to worry about recessing ties and/or bumps at the joiners.  At some point, the weight of the trains vs the rigidity of the rail translates into a need to supplement spikes in wooden ties and Homasote to hold a rail joint in alignment, but it isn't with HO trains on code 70 rail.  On my new layout I will be using code 55 and code 40 rail for HO and HOn3 respectively, so will have to see for sure.

It appears that the Proto rail is placed unto the tie plates and "spike heads" are folded over with a nail set so I guess one could solder feeder wires to the underside of the rail and pass the wires through holes between ties and make the electrical connection invisible after ballasting.

My practice has been to use 26 gauge magnet wire (varnish-insulated) soldered to the bottom of the rail between ties for feeders.  I pre-paint and attach the feeders before spiking the rail down.  With no rail joiners, every piece of rail has to have a feeder. 

Question: If one goes Proto 87 on the trackwork, rails and switches, does one also need Proto 87 wheel sets or can I keep my investement in non-Proto Intermountaim metal wheels?

 

Unfortunately, turnouts force P87 to be an all-or-nothing standard.  P87 turnouts will not pass NMRA-spec wheels, nor will P87 wheels go through NMRA-spec turnouts.  I investigated going to P87 myself.  There were 2 show stoppers.

1. I didn't have the space for the larger curve radii required by P87.  Gauge widening on curves is very limited in P87, and this makes the minimum radius at least 3 times the length of the equipment being used.  There is no cheating this standard as you can do with NMRA specs.  Most P87 experimenters have found the minimum radius for a 2-8-0 to be around 26".  I only have room for 18" or so minimum radius curves.
2. There is a dearth of P87 profile steam drivers.  I do not have the tools, inclination, or patience to make drivers and equalize the chassis of every locomotive I own before running a single train.

So I have settled on using code 88 wheels for much improved appearance where I can.  My handlaid turnouts have the gauge and flangeways set right at NMRA minimum specs so that the code 88 wheels will run without issues.  I do have to be careful about how sharp a curve I use at minimum gauge.  I mostly use curved frog turnouts to conserve space and to increase the "flow" of my track, even though they are not strictly prototypical.

FWIW, ME flex track and CV tie strips do keep track gauge within limits for P87.  Atlas (and possibly some other makes) flex track may have too much gauge widening for P87.

my thoughts, your choices

Fred W

[rioplumas]

Thanks to all for your input.

I would like to model WP c. 1954 (Feather River). I have not checked actual rail lengths for that road that era so I went with a general question based on general rail length of that period.Today UP today uses welded rail and concrete ties except at switches-also a nice look. I'm more interested in the look of the rail being right rather than the length being 33 feet or 39 feet--as I said, 6 inches (or about 43 1/2 feet) would not be a problem with me.

This would be more of a diorama than a layout, a vignette, so realistic rail details is a consideration. I would hope that this could grow some day to be a piece in something than can operate. The look would include tie plates, fishplates, and staggered rails. After seeing the CVT and Proto 87 web sites, it appears possible.The switch details look great.

I like the idea of not using rail joiners, or using half length ones. I have never heard of someone not using rail joiners, so that is new to me.

It appears that the Proto rail is placed unto the tie plates and "spike heads" are folded over with a nail set so I guess one could solder feeder wires to the underside of the rail and pass the wires through holes between ties and make the electrical connection invisible after ballasting.

Question: If one goes Proto 87 on the trackwork, rails and switches, does one also need Proto 87 wheel sets or can I keep my investement in non-Proto Intermountaim metal wheels?

[fwright]

rioplumas

Has anyone modeled 39 or 78 foot lengths of rail with fishplates and at least a nick or groove in the rail, staggered, to simulate 12:12 practice for periods that used bolted rather than welded rail? With metal wheels this ought to sound more authentic.

In looking at photos that have bolted rail, the fishplates are quite noticeable to my eye. I have never been a fan of metal rail joiners that require the eye to ignore something that should not be there, as would an exposed screw head be out of place in the side of a locomotive, for example. So, this would look also look more authentic.

Details West has four and six bolt fishplates. I was thinking of scribing a thin joint in the top of the rail, or filing a nick in it, rather than using a cutoff blade to cut partway through the rail which I would believe would be too wide.

39' would work out to about 5 3/8" in HO.You could approximate this at six inches and get six rail lengths in a 36" piece of flextrack.

In curving flextrack, one rail will extend past the other. Usually this is cut off opposite the now shorter one, but it could be cut half a rail back (or forward if it was long enough). One rail on the next piece could be shifted along the tie strip of the new piece and threaded into the tie strip of the laid piece.

Staggering the real rail joint may be beneficial on curved flextrack to help minimize kinks.

Using metal rail joiners at the real joints would look odd because they would look different than the others that would just have the applied fishplates. Has anyone soldered rail end to rail end without using metal rail joiners (a soldered version of welder rail)?

You could start off with joints opposed at switches and then have half a length on one side with full lengths thereafter to the next switch to achieve staggered joints.

It would be tedious but for rail that can be seen up close, it would have a great look.

Now, I'm sure someone will ask "What about tie plates?" That will certainly work with hand laid track on individual ties. Custom Finishing has 12 metal catings at $4.95. At 30 inches apart, 39 foot rail would need about 30-32 tie plates plus four times as many spikes, so that would be over $26 a lineal foot not counting tax or shipping, expensive versus the less frequent fish plates.

This leads to Proto 87 as an example of realistic 12:12 track laying.

proto87.com has some much less expensive and interesting looking tie plates with spike heads that need occasional real spiking. The site offers joint bars that appear to somewhat less expensive. Also offered is a tie strip with tie plates and spike heads that costs about $2 per lineal foot with ties, rail, and tieplates.

So, is there interest out there in Proto track laying with realistic components?

My question back to you is:

Are you looking to increase the realism of your own track?

Which of these ideas are you contemplating using?

There are those who do extraordinarily realistic trackwork as the photos of Westport show.

As for myself, I model a 1900 era shortline and a loggin come common carrier narrow gague line in coastal Oregon.  Trackwork of these prototypes, from studying photos of contemporary railroads, look nothing like the mainlines of today.  This is a common assumption - that all track is 132lb, 39ft long rail, laid with tie plates at 4-8 spikes per tie, uniform ballast 12" deep on creosoted ties.

Rail in my modeled era was commonly 80 or 85 lb rail on the main, with plenty of 65 lb used elsewhere.  Standard rail length was 33 ft.  Remember, rail ends were/are usually where the damage occurs first, so track crews commonly cut off the end(s) and put in a short length, rather than replace the entire rail length.  Often these short blocks are contained entirely within a long fish plate.

Creosoted ties were not used in the West until later because either the conditions were dry enough to prevent rot and decay, or there was plenty of replacement timber available cheaply.  Ballast on logging lines of that era was typically dirt, with river rock in the really wet places.  Common carrier ballast was tie high in the center, but almost nothing at the tie ends.  Hand-hewn ties were still fairly common off the main line -depending on availability of sawmills when the track was laid.  Tie plates were not used unless the rail was being upgraded to the then new 100lb rail - and only then sometimes.

My practices are to use smaller and narrower ties on standard gauge - 6"x8" or 5"x7", only 8ft long.  I prefer redwood ties (suits my region and era and requires no stain).  When my supply runs out, I will have to invest in a very small table saw to make my own, or use Mt Albert ties and stain them.  4 spikes per tie.  I am experimenting with Steve Hatch's method of using grape vine twigs for hand-hewn ties.  I have quite successfully never used model rail joiners in the past (code 70 rail in HO).  I simply butted the rail ends (no solder) and carefully spike them in alignment.  This did require pre-bending of the rail on curves.  So installation of model fish plates is quite practical where it can be seen.  Others have suggested cutting model rail joiners in half length-wise to reduce the visual impact.

my thoughts, your choices

Fred W 

[jeffrey-wimberly]

 If you want all those nicks and cuts in your railheads go for it. Th my mind that's just that many more things to grab the wheel flanges and cause derailments.

[wedudler]

 I like those details. I've build one ULTIMATE HO TURNOUT from  PROTO:87.

More at my Diamond Valley.

And I've used rail joints too.

I didn't know they shoud be staggered.  :angry:   

To simulate this I cut only the top of the rail head.

Wolfgang

 

[tomikawaTT]

On a prioritized list of essential details, cutting slots in the railheads at 20 scale meter intervals is somewhere off the bottom of the page - even below the operating TBM in a window in the fascia...

 OTOH, I DO simulate rail joints - by pushing the ties that would support them together, just as my prototype does.  Japanese practice calls for squaring the rail joints rather than staggering them, and the  two adjacent joints are supported by two closely-spaced ties.  Since tie spacing is a lot easier to see than joint bars or slots in the rail, I think that is sufficient.

Actually, my prototype was changing over to continuously welded rail during the month I model.  I just jumped the gun and installed it early...

Chuck (Modeling Central Japan in September, 1964)

[loathar]

I can't remember the last time I actually saw a tie plate with 4 spikes in it. All near me only have 2. I've got some fish plates from Central Valley. I plan to use them after I put down ballast. I think it's a nice detail.
Another thing from Proto87 are the realistic looking throw bars they sell.

[grizlump9]

i tried nicking my rails 39 scale feet apart and staggering the "joints" but the noise drove me nuts.

once i ran out of rail joiners and found that a small lenght of copper wire would serve just a well if it was laid against the web of the rail just below the ball or head.   a little flux paste on the rail and the tinned wire soldered up nice and tight.  didn't look too bad either.  i don't have the skill or patience to try butt joints soldered together but i think that would just be asking for trouble anyway.  

good thing about getting older is my eyesight is not what it used to be so fine details are becoming less of an issue.  hell, i do good to see the other end of a 35 car train.

grizlump

[Arjay1969]

 I've done this on a couple of spur tracks at my club.  Once you get everything painted and ballasted, you really can't see it all that well.

That said, I'm planning on going to that level of detail on my planned shelf layout (if I ever actually get to build it!).  Those tie strips that the P:87 store sells are actually from Central Valley (http://www.cvmw.com/).  They also make a line of switch kits that can be curved as needed.

[rioplumas]

Has anyone modeled 39 or 78 foot lengths of rail with fishplates and at least a nick or groove in the rail, staggered, to simulate 12:12 practice for periods that used bolted rather than welded rail? With metal wheels this ought to sound more authentic.

In looking at photos that have bolted rail, the fishplates are quite noticeable to my eye. I have never been a fan of metal rail joiners that require the eye to ignore something that should not be there, as would an exposed screw head be out of place in the side of a locomotive, for example. So, this would look also look more authentic.

Details West has four and six bolt fishplates. I was thinking of scribing a thin joint in the top of the rail, or filing a nick in it, rather than using a cutoff blade to cut partway through the rail which I would believe would be too wide.

39' would work out to about 5 3/8" in HO.You could approximate this at six inches and get six rail lengths in a 36" piece of flextrack.

In curving flextrack, one rail will extend past the other. Usually this is cut off opposite the now shorter one, but it could be cut half a rail back (or forward if it was long enough). One rail on the next piece could be shifted along the tie strip of the new piece and threaded into the tie strip of the laid piece.

Staggering the real rail joint may be beneficial on curved flextrack to help minimize kinks.

Using metal rail joiners at the real joints would look odd because they would look different than the others that would just have the applied fishplates. Has anyone soldered rail end to rail end without using metal rail joiners (a soldered version of welder rail)?

You could start off with joints opposed at switches and then have half a length on one side with full lengths thereafter to the next switch to achieve staggered joints.

It would be tedious but for rail that can be seen up close, it would have a great look.

Now, I'm sure someone will ask "What about tie plates?" That will certainly work with hand laid track on individual ties. Custom Finishing has 12 metal catings at $4.95. At 30 inches apart, 39 foot rail would need about 30-32 tie plates plus four times as many spikes, so that would be over $26 a lineal foot not counting tax or shipping, expensive versus the less frequent fish plates.

This leads to Proto 87 as an example of realistic 12:12 track laying.

proto87.com has some much less expensive and interesting looking tie plates with spike heads that need occasional real spiking. The site offers joint bars that appear to somewhat less expensive. Also offered is a tie strip with tie plates and spike heads that costs about $2 per lineal foot with ties, rail, and tieplates.

So, is there interest out there in Proto track laying with realistic components?