Things have been a little slow on the blog front of late in part because my hobby time has been spent elsewhere and because I’ve been making tools to help my cause which naturally takes away from actual layout construction. One such gizmo is the rail bender. Essentially an item older than the hobby itself, I remember seeing one of these in my very first issue of Model Railroader magazine. In that article they used an Orr bender to produce very neatly radiused bends for a traction layout. The opportunity to poke at a Fasttracks bender got me out of the arm chair and pouring through boxes looking for appropriate bearings and hardware. While I wasn’t able to find the right bearings I found most of the materials in my stack of off-cuts. A trip to Princess Auto gave me everything else and left me seven dollars lighter.
The design chosen closely follows the Fasttracks model with accommodations for manufacturing process and personal skill level. Three bearings are employed, two fixed with a third bearing race moving between the fixed pair. I used a piece of 1/2″ 6061 aluminium plate for the body with a 3/8″ X 3/8″ slot up the middle to take a movable steel die block and on it the third bearing. I kept the tool body square rather than spend time making it a more ergonomic shape. Doing so makes it easier to machine and makes it easy to hold in a bench vice. The die block is captured on the top by an 1/8″ piece of plate with a slot just wide enough for the bearing spigot and bolt to pass through.
Rather than employ a spring to hold the die block firm against the leadscrew, I opted to make a T-slot in the die block that slides over a correspondingly T-shaped end to the leadscrew. I’ve been exploring this method for use on another type of miniature bender and figured it was worth it to try it out here as well. Here you can see the die block (Cold rolled, 1018) and cutter( W-1 drill rod). Some other construction photos follow.
hardened T-slot cutter and along with bottom face of the die block
Boring out slot ends in the top plate to .440″. Slot will be completed with 3/8″ slot drill running first up the centre and then each side; necessary for the non standard slot width.
Assembly starts with screwing the leadscrew into the lower body piece , dropping the die block over top of the screw followed by the top plate, then ancillaries such as bearings, spacers and bolts.
Functional rail bender
The rail roller is functional but unfinished in the following picture. The 33″ radius curved rail having just been formed. Items that need seeing to include shortening of the stationary bearing bolts , finding or making a die block bolt with appropriate shoulder, drilling and tapping for top plate fixing bolts. A leadscrew knob is also in the cards though it’s a bit up in the air whether I’ll knurl it or machine grooves. Some form of marking would also help quicken set up and aid repeatability.
Sometimes while making a bunch of parts you realise things could be done differently. Today while making additional end of module track supports I took stock of everything and decided to make a cutter and see if it wasn’t possible to mill all of the gullets in the code55 rail. I got started by mounting some .125″ W1 Drill rod (water hardening tool steel) in my trusty Taig lathe. The photo below shows the beginnings of an end mill mounted up in my four jaw. I turned the rod down to .072″ which just so happens to be the same amount as the space between ties on the mainline.
While I was rummaging around for my torch I found a cutter similar to what I had in mind that was already filed down in thickness and with all of the cutting edges backed off. It was hardened to boot so I stopped looking for the torch. These cutters work very nicely in brass and I wanted to see how they would perform in nicklesilver which while soft requires a very keen edge, keener even than for brass. This particular double edged (can’t really say it has flutes…) cutter was a little larger than the one in the chuck at .082″
It worked! My setup wasn’t all that rigid and the rail’s unique profile didn’t help matters but it DID cut. I think I’ll finish the other cutter and when I do I’ll maybe mill it to give it less rake rather than just filing it in the vice. Still for a quickie cutter it saves time.
Here’s the end-of-module joiner that needs installing before the position of BR No.388 can be finalized. The rails get soldered to the PC board, the PC board gets epoxied to the wood and the wood glued to the ply roadbed with carpenter’s glue. Five are required before the trestle can drop into place. The rail in this instance is HO code 70 (Micro Engineering I think) that will sit just proud of the wooden ties and provide a nice flat surface to solder to the base of the code 40 running rail. I’ve used this method before, inspired by something similar in an old issue of the P87 journal. Having gone over to milled acrylic for my switch sliders the stacked rail method is still a robust method of maintaining alignment through module joins.
Prior to soldering the running rail, three piece wooden ties are be added and the works filed to height. Once the height is fixed the rail will be scalloped wit a keen file to allow daylight between the ties and running rail. In the past I did them with a moto-tool and abrasive disc but I find the effect much more pleasing when worked by hand. I start with a triangular file and then move to a rectangular one to clean.
Just another in-progress photo.
There’s something up with my blog settings as everything comes out so much darker and with much greater contrast than in the original images but you can get an idea of what is going on with the model.
Here is an updated picture of the painted goings on one week after the fact.
Note the artifacts above the last bent. I flooded too much paint into the corner while painting the X-bracing. Easy enough to correct but tough to get enough light into the corner to see what is going on.
Most of the work was on the piles and got obscured by glare from the gloss black I ended up using. Seems my Vallejo setup was missing anything darker than grey-black so Golden fluid acrylic (gloss) black was used in its place but the effect is not the same. I’ll need to tone it down when I play with the final sheen. I’m still working from the far end to the near but it is starting to look better. Touching on the sheen once more, it isn’t simply a matter of spraying one thing and it is done. Doing so would negate a lot of the nice boundaries formed through multiple layers of the Vallejo paint. Instead, the outer piles will be hand painted with some very light glazes.
A black and white version of the image above. Monochrome images can be used much the same as a painter or draftsman might use a mirror: the “flip” allowing the modeller to see something they might otherwise have missed.
Nicely weathered X-Great Northern BR 68.08
Painting is one of my favorite activities so it is a shame in a way that I don’t do more of it. It just happens that I like fabrication as much or more and so I naturally spend more time working towards painting than actually pushing it around. With BR 388 finally stuck together it is time to start painting but before that can happen I decided to look at some local X-GN bridges for inspiration.
One thing that stuck out looking at the bridges in question was the variety of colours on display. That should come as no surprise but it still sort of does. With that in mind I decided to spray a base coat of a burnt sienna and raw umber and go from there.
X-Great Northern Railway trestle over the Serpentine River
Here is the trestle after a night of spraying together with some early morning doodling with a paintbrush. The spray job turned out to be a bit more opaque than necessary but I’m working with it. As often as I paint I’m not surprised that things went a little sideways. The upper right bent is the only area to be worked through but still needs a bit of ultramarine and Quinacridone to liven up the 6X12 ends. A few of the visible bents were stippled at high pressure (45PSI). Hopefully further treatment will bring this effect forward in places. Playing around with various types of mark making is one of the best parts of the process and thankfully doesn’t require a lot of brain power.
10X eye loupe
After some deliberation I went back and added 500 odd bits to the top of the existing NBW which by the time I was part way done were more like plain washers. While not the best photo this shot through my mobile loupe arm shows some of the detail that was fixed and some work that still needs addressing. The little loupe arm really aids artifact checking, vital for model making in all scales. Due to my aversion to solvents I found it handy to do finish work on the NBWs right in the spraybooth. The arm was up to the task where my own eyesight fell well short.
The bolt/nut portion of the revised NBWs were made from .005″X.005″ strip cut from sheet and bobbed to length by eye. The little pieces were placed in location with a small paintbrush having loaded it with an appropriate amount of solvent. The washer was similarly wetted with solvent. The resulting bond is actually pretty decent but just in case I decided to wipe the works with a small amount of solvent afterwards to ensure everything stayed in place.
A few of the NBWs were out of position by more than what was reasonable and in these cases I made new NBWs complete on a sheet of glass flooded with a small thin layer of solvent. Once the washer portion was dry I used the brush again to place the nut/bolt part in the middle. It takes some amount of practice but with a solvent licked No.11 blade the whole deal can be picked off the glass and placed on the model. Were I starting over (again!) I might have done all of them this way.
The little bit of strapping modelled was done with .002″ styrene. To get .002″ styrene you attack with a razor blade and scrape it down till it mics out and then stick sand it lightly to even things out. I cut it to width by eye with a scale ruler using the various ruler marks and my memory of scale sizes to get within range. Dipping the styrene in a small puddle of solvent and then brushing it lightly with a damp paint brush seems to do the trick. The process has a lot in common with decaling. A lot of folks seem to have trouble working with small amounts of styrene and solvent cements ( I use Testor’s in the glass bottle) but in most cases I get the feeling that the individuals in question tried it once or twice and went home. Like most things practice and forethought go a long way in achieving usable results. Try, make notes and try again. Most things are possible provided you don’t give up.
Going with the top rows for the 40′ deck girder.
This being a holiday, I took the opportunity to start in on BR 387. The bridge is a pair of open deck girders; one 85′ the other 40′. Umpteen years ago back when the GN goat yahoo group was the only source of info I had I posed a question regarding the bridge’s dimensions. Glen Haug very kindly provided me with most of what I needed to start the project. And so I did. You can see the result in the header. That was back in the days of my 4’X8′ layout. Sections of Atlas plate girder bridges were cropped and elongated to produce a model that fit the envelop.
Fast forward to today and I’m on version #3, version #2 being stillborn owing to poor technique. Using a modified Taig/NWSL riveter, row upon row of rivets are easily embossed on what ever pitch and schedule I desire. Basically, If I can turn the punch and die I can emboss the rivet. No more trouble aligning rows or using registration rivets to work out offsets, the X-Y table, eliminates the need to ever reposition the workpiece. This new found utility does come at a price though for one needs to know what size rivet to emboss in the first place. Luckily there are some bridges available locally to inspect. On my walks I’ve most often seen 1.25″ and 1.375″ rivets on bridges designed with similar load limits. If the rivets are to scale out after paint then best to make them a touch undersize. With that in mind I went with a .006″~ rivet shown at the top of the image.
First step will be to produce the top and bottom vertical chords and then fettle everything else into position.
Back to riveting,
Test bent with larger piles
After some deliberation and another round of consultation I took the trestle apart and started over with larger piles and cap strips. This necessitated a further round of washer and NBW punching but in the process I got better at handling and gluing them together. I’ve been using a No. 11 blade to pick the washers out of the bottom of the die, into a small puddle of solvent and then onto the model. This method seems to work alright however I need to use a glass surface in the future as puddling the glue on the brass that just happened to be handy led to oxidation making its way into the solvent and onto the model. I can’t see it harming anything but it marrs an otherwise half decent finish. The next step will be to finish up the wood grain and cancel out as many artifacts as possible. Along the way I’ll be picking out the odd line and trying to encourage the model to adopt a more organic posture. Then it’s to the paint booth.
Save for some touch-ups this is what the trestle will look like.
A little over halfway there with the hardware. I made two sets of punches for the washers.
Blurry, this is a close-up of the small washer punch. The punched out washers fall out the side of the die.
Washer punch and die overview
Lots of NBWs!
Before I go too much further I’ll need to finish repositioning the far-side braces third bent in to a slightly higher location. At the time (it was the first bent braced) I was thinking the road would be at a lower elevation and that the brace would be partially buried. I’ve since changed my mind.