Palmer Industrial Fascia

I mentioned at the end of my Foam on the Palmer Industrial Park post that I was planning on following the recommendation in Lance Mindheim's book "How To Build A Switching Layout" to use 1/4" MDF for fascia.  The bending experiment I showed at the end of that post was a success.  Here's a photo of that piece of MDF after I unclamped it and left it sitting on the bench for a couple weeks:

The bend holds even when unclamped

 You can see that the bend has relaxed a little bit, but not much.  Success!  I bought a sheet of 1/4" MDF to do the rest of the fascia.  I promptly broke two strips trying to put both bends in a single piece for the end of the peninsula.  Sanity prevailed while I still had enough left of the sheet.  I went back to my original method of doing just one bend at a time.  I used a trash can lid, sitting on a piece of 2" foam with a hole cut in it for the handle, so the lid would sit flat.  I filled the lid with water, and soaked an old towel in it.  Then I wrapped the towel around the section of an MDF strip that needed to be bent.  I propped the ends of the MDF up a bit to give it just a little bend, and left the towel wrapped area sitting on the trash can lid with the water deep enough to keep the towel soaking.

Heavily soaking at the beginning of a bend

The water needed to be refilled as it soaked in.  And pouring it over the top of the towel made sure the whole towel wrapped area stayed soaking.  Over the course of several hours, I propped up the ends higher and higher, wedging them in a little too to keep increasing the bend.  It's very important not to rush this process, or the MDF will break.  As the bend progressed the towel wrapped section sat further down in the water in the trash can lid, which made keeping it soaking easier.

The bend is progressing nicely

Once the MDF will bend to the desired radius without much force, it's time to go clamp it on the layout.  I did each bend in a separate piece.  Some planning as to where the joints in the MDF will be is necessary, so you can get the bend in about the right place with enough extra at each end so you can trim to fit.  It does take a little time to do, but it's kind of fun and I think the end result is worth it.


I fastened the MDF to the layout with yellow glue and brads.  At the joints between pieces of MDF I just made sure both ends were cut square, butted them up, and glued them like crazy.  After all the glue dried I went back and filled in the imperfections at the joints with wood filler, and sanded it flush.  It didn't take much.  The dark color of the fascia hides any remaining imperfections.  The joints between the MDF sections have a 1x4 behind them below the plywood surface, which helps keep them firmly aligned.  The top 2 1/4" is unattached though - all that's behind it is the 2" foam.  The rigidity of the MDF combined with it's solidly glued base seems to make the top end of the joints between strips plenty strong enough, even though they are just butted together and glued.

Fascia in place and painted

Above is a picture of the end of the Palmer Industrial Park peninsula with the fascia installed and painted gloss hunter green.  I painted the layout supports satin black too, which was a big improvement.  I used valspar latex enamel paint - the color, finish, and brand were selected by what was available at the local Lowes pre-mixed in a quart can.  The colors turned out pretty nicely.  The green is a pretty dark color, and it doesn't seem to be possible to get a picture of it that really shows it's color properly - it looks darker than the above picture in person.  I used forest green on the O scale layout (in the background), and in person that doesn't look as nice as the hunter green.

The paper track plan, cut back to not overhang the edges, did a good job of keeping the green fascia paint off the layout surface.  After the fascia paint was dry I pulled off the paper track plan for good.  The most recent thing I've done is finish going over the rows of indentations in the foam left by the pounce wheel track plan tracing so the track center lines are easily visible.  We lost power for four days in an early but nasty storm here.  It eventually occured to me that since the best way to go over the rows of indentations is to shine a flashlight along the layout surface, it was a task perfectly suited to a cold and dull evening with no power.

Fascia done, backdrop done, trackplan traced - time for track?

The current state of the layout is shown in the photo above.  It's now time to quit waffling on some final track decisions.  When I started this project I planned to hand lay code 40 turnouts and use ME code 40 flex track.  I eventually convinced myself that if I ever wanted to actually get the layout running I should use Atlas code 55.  I've purchased all the Atlas code 55 track and switches I need, but I must admit there has been a little nagging voice in the back of my head saying "it's still not too late to do code 40".  I also need to finalize a decision on how to control the turnouts (ground throw, just plain friction, tortoise, something more exotic), and practice soldering point jumpers to those tiny little points a bit more on my two sacrificial practice turnouts.

Up to now, all of those seemed like decisions I could string along until the time came.  The time has now come.

Aeronautical Distractions

Model railroading has been my main hobby for as long as I can remember, if not longer.  But I do have other hobbies, and occasionally one of them occupies enough time to be a serious distraction from railroading.  Recently that's been the case with radio controlled airplanes.  Specifically, small (about 15 inch wingspan) battery powered electric RC planes.

The aeronautical distractions, bottom to top in order of preference

Of the six micro rc airplanes I've had, five are still flyable and are shown in the above photo.  Micro in this case means they all have a wingspan in the range of about 15 inches.  First was the Parkzone Vapor (center in photo).  With 3 channel control (throttle, rudder and elevator), it makes a good beginner plane.  Very rugged too.  Second was a Parkzone Cessna, a little faster but still 3 channel.  It's not in the above photo, because it had a terminal encounter with a tree.  Third was the Parkzone Sukhoi (top plane in photo), the first micro rc plane with ailerons on the market.  It's "twitchy", and was too much for my abilities at the time.  Ailerons allow for more variety in crashing - cartwheeling turned out to be a specialty.  I got overwhelmed, shelved it and stopped flying for a while.  The fourth plane I got is the Parkzone P-51 (second from top in photo).  Also a 4 channel plane (throttle, rudder, elevator, and ailerons).  It's not twitchy like the Sukhoi.  I learned a lot with it until a tree jumped in front of it and bent the prop shaft.  While waiting for a replacement I bought the T-28.

The bottom two planes in the above photo are my favorites.  The Parkzone micro T-28 (second from bottom), and an E-flite micro Beast (bottom).  The T-28 is a very fun plane to fly.  It's very well behaved, yet still pretty maneuverable - loops, rolls, and other stunts are pretty easy.  Relaxing and fun.  It also has the distinction of being the only micro RC plane I haven't crashed (if you don't count a few bad landings).  The Beast is, well, a beast.  It's the only plane in the bunch to have a brushless motor and a 2 cell (7.4 volt) lipo battery - the others are all single cell.  It's got very large control surfaces.  It goes fast, can turn on a dime, and can roll fast enough so you loose orientation (i.e. you think it's rightside up when it's upside down).  It does exactly what you tell it to do, and has no self correcting tendencies (if you let go of the controls it won't level out).  It can get into a lot of trouble very fast.  If you look closely you'll see the that left top wing has been broken off right at the inside corner of the aileron.  The bottom wing has been broken twice.  The guy wires have been popped off a couple times.  The nose has been mushed in once.  Medium thickness foam safe CA with a kicker can work wonders.  But the wing does have a slight twist to it now that messes up some maneuvers.  It's still in good enough shape to teach me a few more things though.  It is most definitely NOT relaxing to fly (at least at my current skill level), but it is a lot of fun.

To tie this back into model railroading, you'll notice the planes are sitting on the benchwork of the N scale Palmer Industrial Park - in fact they're sitting on Trans Plastics.  That photo was taken after I finished installing and painting the fascia (the subject of the next post I'll write), but before I finished tracing the pounce wheel dents in the foam with black magic marker to make the track plan more visible.

Now that the weather is colder and the wind seems to insist on being too gusty to fly these little planes when it's still light enough to see them, perhaps I'll get back to some more model railroading.

Palmer Industrial Backdrop

The backdrop for the Palmer Industrial Park goes down the middle of the layout, which sticks out into the middle of the room.  I didn't want the backdrop to change the feel of the room from one big open space into a sort of maze.  So I decided to keep the top a bit below my eye level.  A nice convenient height seemed to work out to 10 inches above the foam.  That seems like it should be quite high enough to provide a good backdrop for relatively flat N scale scenes - none of my buildings will be anywhere near 10 inches tall, and even tall trees won't quite make it to the top (10 inches is 133 N scale feet).

I described the backdrop supports in Foam on the Palmer Industrial Park.

Order of operations is always a good debate.  I wanted to get the foam painted a reasonable earth color, the backdrop painted sky blue, and the fascia painted dark green.  It seemed to me that the best way to do that and have the least potential for getting a color of paint somewhere it shouldn't be was to do things in this order:

1. Paint the foam.
2. Tape down the paper track plan, to transfer it to the foam and then to use it as a drop cloth to keep backdrop and fascia paint off the foam.
3. Put up the backdrop and paint it.
4. Put up the fascia and paint it.
5. Rip off the now paint splattered, pounce wheel punctured paper plan and stand back to admire it all.

Without backdrop or fascia to worry about, painting the foam is fast and easy.

Foam painted a light tan

I used a color from TrueValue called "Benefactor" (number 14D4).  It's not quite what I hoped it would be, but it serves the purpose of being at least somewhat ground colored, and still light enough to easily see pencil and marker lines on.

With the foam painted, getting the paper plan back down over the backdrop supports turned out to be fairly easy, with only a bit of fiddling around to get it positioned properly again.

Track plan back over painted foam

As I mentioned previously I decided to try a roll of vinyl flashing for the backdrop, so I wouldn't have any seams.  The flashing is fairly thin, which is why I used sheets of plywood to support it.  The open question was what glue to fasten it with.  I didn't want to have to try to clamp the vinyl in place while the glue dried.  I wasn't sure I could do it without denting the vinyl.  And I didn't have any clamps with long enough reach anyway.  So I read the fine print on the the tubes of glue in the glue aisle.  After rejecting the ones that had dire warnings about ventilation, I found Loctite Power Grab for tub surrounds.  The fine print says it has "very high initial tack which reduces the need to brace".  I clamped a plywood scrap vertically to a support column in the basement, spread some of the stuff on it, pressed on a chunk of vinyl, and sure enough it stayed put.  After it dried, I couldn't see anywhere that the glue had damaged the vinyl.  And I couldn't pull the vinyl off without exerting enough force to wreck the vinyl anyway.  So I bought another tube (and later discovered I should have bought two more tubes), and started on the real backdrop.

Glue spread into thin layer on support

I used a plastic scrapper to spread the beads of glue out into a more or less solid thin coating on the plywood backdrop supports.  Then very carefully press the vinyl into it.  You can pull the vinyl back off, and even slide it around a little to get it exactly where you want.  The only really tricky bit for my backdrop was the bulb around the end.  I used an extra thin bead of glue up the single post at the end, and let the curve of the vinyl do most of the work, pulling it around until it just squished into the glue bead on the end vertical a bit.  Here's the backdrop completely glued on, with the remainder of the roll sitting on the O scale layout at the far end, waiting for the glue to dry.

Backdrop glued in place

At the far end the backdrop curves out to the wall with a pretty tight radius.  I didn't actually glue the backdrop to the wall, I slipped a sheet of thin (1/16 or so) model aircraft plywood I had lying around in and glued the backdrop to that.  Saves messing up the wall.

Once the glue was dry, I trimmed off the remainder of the vinyl roll and applied the first coat of paint.

First coat of paint

I used valspar EB6-4 Sky Blue.  I chose the one color actually labelled "sky blue" after bringing home hundreds of those paint sample strips with exotic names for various shades of light blue.  I'm very happy with it.  I'd always thought the backdrop on my O scale layout was a bit to intense of a blue, this one seems much better.  In the above photo you can see a couple blue marks on the track plan, so it is serving it's new purpose as a drop cloth.



So, if I was to do the backdrop over, I'd make a continuous strip of plywood down the center, cutting a series of vertical saw kerfs deep into it where the shallow curves are to make the bends.  But I'd do the big blob at the end the same way.

Transfering the Plan

On both my new N scale Palmer Industrial Park and my older O scale No-Name Industrial Park I drew the trackplan in a cad program (Ashlar Vellum, for reasons I won't bore you with here).  The cad drawing has very accurate turnout locations with point, frog, and center lines marked, and also easements.  The challenge then becomes getting the drawing onto the layout in a form that helps you lay track.  I described the process of getting the full size printout in Mocking up the Palmer Industrial Park.

Once you've got the full size plan taped down to the layout (and in the case of foam, if you're going to paint the foam paint it before you get to this stage) you can transfer the plan from the paper to the layout.  I use a pounce wheel, which is a sharp pointy wheel in a handle.  You can get cheap ones at fabric stores, but they don't have the sharpness necessary to do a good job.  I bought the 3-piece pounce wheel set from Micro-Mark, and I mostly use the largest of the three wheels.

Using the pounce wheel is simple.  Try it out on a scrap with however many paper thicknesses you have to see how much pressure you need to get a good line of dents.  It needs a lot more pressure on plywood than it does on foam.

Using the pounce wheel

At intervals while you're doing the pounce wheeling, take a break (your hand will need it anyway), and very carefully peek under the paper without moving it to make sure you're getting a good line of dents - not too deep, and not too faint.

Once your done, make sure you've got decent lines of dents everywhere.  For laying track, you want the line to be clearly visible between ties and through a thin layer of glue, so you need to go over the dents with a pencil or permanent marker.  If there are places where the dents get a bit faint (it happens), shining a flashlight along the line will bring them out.

A flashlight makes tracing dents easy

A bit of care and patience, and the cad drawing is transferred very accurately to the roadbed.

I've done all the pounce wheeling on the Palmer Industrial Park, but I haven't yet done all the tracing.  The printed plan is now serving as a drop cloth to keep backdrop and fascia paint off the foam, after which it will be discarded and I'll finish the tracing.

I should note that an alternate approach I've seen used on the North Shore Model Railroad club layout's newer areas is to simply glue the printed track plan to the plywood, and glue the roadbed and track to the plan.

A note about NO-OX

A number of people have reported good success using NO-OX to treat track.  If you search for NO-OX track you'll find a number of references.  Sanchem is the supplier, and they even have some instructions on their web site: http://www.sanchem.com/aSpecialE.html.  Since I have had occasional dirty track troubles I decided to give it a try on my O scale layout.

Worst mistake ever.

I will freely admit that I succumbed to my ever present tendency to over do things, and used more than I should have.  But I still didn't use very much.  The problem is loss of traction.  My MP-15, which should be able to move at least 12 of the Atlas 60' Berwick hi-cubes easily now struggles to move 4, and 8 is out of the question.  I've cleaned the track dozens of times since the NO-OX application, with alcohol, goo-gone, and several other things.  I've scrubbed the top and both sides of all rails with a toothbrush and alcohol, cleaned out the flangeways with Q tips and wipes, cleaned every single wheel on the layout one at a time with a goo gone soaked Q tip, and repeated the process.  So far I haven't materially improved the situation.  Nothing feels greasy, but it evidently is just a bit slippery.  I do have hope that with more cleaning and more time I will eventually get rid of the residual greasy NO-OX carrier.  But by then I will have done more track cleaning than I otherwise would have for the entire life of the layout, so it hardly seems worthwhile.

In the meantime I can't operate the layout as designed, since my loco is unable to move the requisite cars.  Fortunately the layout has a reasonable amount of operating interest even without the Tighe warehouse where the 60' hi-cubes go.  And the rest of the industries take cars that are lighter and/or roll better, so I think I'll be able to claim the warehouse has closed and operate the rest pretty much as planned.

I won't go so far as to say you shouldn't use NO-OX, since some people seem to be very happy with it.  I will say that you should be EXTREMELY careful about using too much.  And you might want to consider alternatives.

Foam on the Palmer Industrial Park

I finally got back to working on the N scale Palmer Industrial park this weekend.  The focus was getting the backdrop supports in and the foam down.

The backdrop supports came first.  I've decided to go with a 10" high backdrop.  Since the plywood is at 48" above the floor, that puts the top of the 2" foam at 50", and the top of the backdrop at 60" - just a bit below eye level for me.  It seems to be high enough to make a nice backdrop, but low enough to leave the room more open feeling.  I wanted to keep the backdrop thickness down, so instead of building a more complicated frame I opted for using my favorite fastener - the pocket screw - to fasten 12" high pieces of 3/4 ply vertically along the backdrop center line.  The paper plan now has cut outs for the backdrop supports, the theory being that once the foam is glued in and painted I will drop the paper plan back down over the supports and trace through onto the foam.  We'll see how that plays out in practice later.

The backdrop supports pocket screwed into place.

The photo also shows what I'm going to try for a backdrop - 10" wide vinyl flashing.  The total length of my backdrop is roughly 30 feet (15 feet from the wall to the end blob, then back), so the 50 foot roll will let me have no seams, and give me a little to experiment with before I start to glue it in place.

The next step was cutting the 2" foam sheet to fit.  I tried a serrated bread knife, but that quickly got stuck.  I bought a cheap fillet knife, which worked much better but also eventually gets stuck.  So I ended up using a 6" or so sawzall blade mounted in a handle I got somewhere for the purpose.  The sawzall blade cuts on the upstroke and it's got relatively fine teeth, so it makes a decent, fast, and easy cut.  By doing all the cutting on the other side of the room on a couple saw horses and vacuuming each piece before bringing it to the layout I kept all the little blue foam dust away from the layouts.  The backdrop has several angles and curves in it, so each piece of foam required some careful layout.  I balanced the 2 foot wide sections of foam in place on the layout as shown below (the gooseneck thing is for weight to keep it from tipping off), made sure the front edge was parallel to the front edge of the layout, and measured in from the backdrop the same amount as the front edge overhung the layout to transfer the back edge marks.

Transferring the back edge line to the foam.

The worst piece to lay out was the one where the lead will go - it's got odd angles on both edges.  But everything ended up fitting OK.

Test fitting the foam.

I'm considering several possibilities for the scenery at the end of the peninsula.  One is modeling the Swift River that follows the real park track, which would cut down into the 2" foam.  Another is building up the terrain a bit so the sharp curve is a bit concealed in a cut.  While I could still do it, I traced the outline of the end foam onto a 1 inch foam sheet and cut it to match.  Not sure if I'll use it yet, but it was easier to cut it now than later.

A 1" sheet cut for raised scenery at the end.

With everything ready, I glued down the foam.  I used Liquid Nails projects construction adhesive, which is safe for foam.  I had assumed one tube of it would be enough, but ended up having to make a quick trip to the hardware store for a second tube.  A motley collection of stuff from around the basement served to weigh down the foam.

Foam glued and weighted down.

Watching glue dry is, if anything, less fun than watching paint dry since you can't even see it.  So I decided to test bending the fascia around the corners.  One corner on the peninsula is a 4" radius, the other is 6".  I don't remember why they ended up different - that was a number of plans and several years ago.  Lance Mindheim's book "How To Build A Switching Layout" (an excellent book) recommends using 1/4" MDF for fascia.  I used 1/8" masonite on the O scale layout, and avoided the warping problems I've seen other people have by gluing and nailing it to the 1x4 at the front edge of the layout.  But the top edge is a little thin to lean on, and I've noticed that people like to lean on the layout.  Preventing leaners seems to be impossible (I catch myself doing it sometimes), so with a 1/4" fascia the top edge should be considerably stronger.  My one reservation about the 1/4" MDF was whether I could get it to bend to the 4" radius corner.  I bought a 2x4 foot sample piece, ripped a 6" strip off it to try.  If I do end up using that 1" foam to raise the scenery at the end I'll need to use a 7" strip, but that's not today's problem.  I soaked the bend area using a garbage can lid and a dripping wet towel, and then tried to bend it around a roll of flashing I had hanging around that's about the right diameter.  Very awkward trying to hold everything in place and clamp it.  Eventually I gave up and clamped it directly to the layout - a little damp won't hurt that corner.  Surprisingly enough, the 1/4" MDF does seem to bend nicely around the 4" radius.  I guess I won't know for sure if it will work until it's thoroughly dried out and I remove the clamps.

Test bending fascia to 4" radius.

I'm eagerly looking forward to the next steps - gluing the backdrop in place, and painting it and the foam.  But that will have to wait for the glue to dry.

No-Name Evolution

The No-Name Industrial Park track plan and indeed the layout itself has gone through a series of identity changes, geographical relocations, time periods, and just about any other change you can think of.  I'll describe a few of the more interesting points along the way.

I started off knowing I wanted some focus on industrial switching, but I was also thought I wanted some continuous running.  I had originally been thinking about modelling the 50s, the default time period of model railroading.  Once I realized how big an O scale turntable was going to be I decided steam power was not a wise use of my space.  I slid the time period ahead to the late 60s and early 70s.  The location was set in the area of western New York where I grew up.  So my first serious O scale plans were called the Elmira Industrial, placed in Elmira, NY circa 1970.

The Elmira Industrial, an early version

I managed to squeeze in a double track mainline, some staging, a small yard, a couple industries in Elmira itself, and an industrial branch.  At the end of the industrial branch was Thatcher Glass, a large glass bottle manufacturer.  And near the end was an interchange with the PRR.  Even at the time I drew the plan I knew better than to have a duck under - I just don't like them.  So I thought perhaps a drop leaf or something might do.  But that compromise, combined with the short length of the yard tracks, staging, and the over crowding of the industrial area made me have second thoughts.  When I started thinking about the tiny yard and how many cars it would have to handle in how many different blocks I realized the plan was out of balance with itself.

The next iteration of the plan did away with the duck under and therefore the mainline as well.  It still had staging though.  The general idea was that the mainline (which actually would exist in dummy form on the front edge of the layout near the yard throat) was implied, and trains would pick up and set out blocks in the yard.  In reality that meant and engine shoving a cut of cars in from a staging track and returning with a different block.  There was staging for east and west bound EL traffic (the railroad being modelled was the Erie Lackawanna), Lehigh Valley (which ran up from Sayre PA through Elmira to Horseheads, interchanging with the EL in Elmira), and a couple Pennsylvania tracks (PRR had trackage rights on the EL through Elmira).

The Elmira Industrial, a later version

The larger yard in both length and breadth would easily accommodate handling the inbound blocks, and classifying for various industries and the 4 different outbound blocks.  Thatcher Glass still occupied what you will recognize as the peninsula devoted to the N scale Palmer Industrial Park now, but it's based on a slightly earlier configuration of the prototype before the newer set of warehouses got built.

At that point construction got seriously under way, and the bottom part of the plan along the wall was built, plus a little bit up the left and right side walls.  I hand laid code 125 rail on wood ties, using Old Pullman turnout kits.  It was exciting - nothing beats the thrill of the first train run on hand laid track.

First train runs!

After the initial euphoria wore off I started to notice a few hiccups with the Old Pullman turnout kits and in the process of tinkering with them to improve them I learned more about turnouts and what to keep an eye out for.  Some inspection revealed that the one curved turnout kit I had purchased would require so much work to get the closure rails to properly line up with the frog, etc. that it might just be easier to start from scratch.  (I believe I may have ordered my turnouts at a bad time in Old Pullmans transition between owners, I would not assume my experience from some years back is any indication of their current product which I have heard good things about.)  If you look at the trackplan closely you'll notice that there are 6 or 7 curved turnouts in the yard throat area that are critical.  Any flaws in this area would make the layout a nightmare to operate.  And you'll notice that there are a lot of turnouts in general.  Doubts about my ability to get this layout operational began to set in.

Around this time two other things changed.  I started doing a lot more rail fanning, and began to get enamored with present day railroading.  (I don't use the term "modern", because various people use "modern" to refer to anything from a 1950 F unit to a 2010 genset.)  And Atlas starting making good quality O scale turnouts.  I started thinking about how to re-arrange the plan to use Atlas O turnouts.  Without the curved turnouts in the yard throat, the yard couldn't be nearly big enough to support the general operating scheme.  Major changes seemed to be in order.

The Corning Industrial

The result was the Corning Industrial plan.  Corning is about 10 miles west of Elmira in real life, and the former EL has gone through the Conrail stage and is now the Norfolk Southern.  This time the mainline was not even hinted at on layout, it was "over there" a the other end of the yard just past the top right corner of the room.  There was also no staging.  The assumption is that through trains, one eastbound and one westbound, would pick up outbound blocks and set out inbound blocks "overnight" - in other words between operating sessions.  At the start of an operating session two of the yard tracks would have the two inbound blocks on them.  A third track allowed for some storage.  The remaining two tracks were the through track and the runaround.  Not many tracks in number, but all pretty decently long, and I figured there would be enough space to sort things out without getting too crowded.  The parts of the layout with track already laid stayed untouched, with one exception - the second track and industry in the middle of the wall got removed in favor of a single larger industry.  The one remnant of the original plan lingers to this day in the No-Name Industrial - the jog in the middle of the track.  There's still a glass plant on the peninsula, but this time it's based on the Osram Sylvania plant in Wellsboro, PA.  The area formerly occupied by the yard was now planned to be an intermodal area.  Two tracks for auto racks, two tracks for intermodal flats.

Norfolk Southern uses big power for everything, and with the length of cuts I expected to be dealing with to sort out the yard it seemed like a two unit consist would not only be prototypical, but necessary.  I bought two Atlas O B40-8s.  A couple sample intermodal flats, more articulated auto racks than I care to admit, and I figured I was in good shape.  I built the left and top wall parts of the plan, and got started on the curve around into the intermodal area.  Things went considerably faster with flextrack and ready to run turnouts.

Two thing went wrong.  First, although it seems like there is enough track to handle everything, once I started actually moving cars around I started to realize that while there was enough track for everything to be on without over crowding at any given point in time, there wasn't necessarily a way to move a cut from point A to point B without either pulling long cars through the tightest curve on the railroad (36" radius going to the glass plant), or breaking the cut in half and doubling it through the crossover between the runaround and the through track.  My long yard tracks which had seemed so nice and spacious on the plan were turnout out to have moved the throat so far around that I didn't have enough lead!  Changes were called for.

The second thing that went wrong is the B40-8s themselves.  My first Atlas O engine was an SW-8 which runs like a dream.  (I don't really count the little plymouth diesel in the above photo - an ancient Atlas unit I picked up at a white elephant table.)  Very smooth starts and stops, good smooth slow speed running.  The B40-8s on the other hand simply would not sustain any speed between 0 and 3 or 4 smph with a standard decoder - you had to either be starting or stopping.  Once you got past the initial jerk into motion they ran pretty smoothly.  But since my whole operation is switching, starting and stopping is very important.  I could not make them run what I considered to be acceptably smoothly while starting and stopping slowly.  I got reasonably close with two zimo MX69S back-emf decoders, one for each of the two motors in each unit.  There were still subtle but irritating jerky motions though.  I bought an SD40 when they came out and were reported to have fixed the running problems of the B40-8s, however I found it still suffered from the same fundamental problems.

The Corning Industrial was now teetering on brink of survival, and the thing that pushed it over the edge was the arrival of the Atlas MP15 - a new switcher which had all of the same operational qualities as the original SW8 in a more modern and highly detailed package.  The Corning Industrial got a major mission change - sparser, fewer cars to be handled per op session, and it would be considered to be the end of an industrial branch.

The Corning Industrial last gasp

I did some mockups, pulled up a couple tracks, and played around with the idea for a while.  But in the end I came to the conclusion that a large part of my problem all along had been over reaching - planning something larger than I could reasonably build and maintain, and cram too much operation onto it.  Drastic measures seemed to be in order. I decided to keep just the left and bottom sides of the plan, and rip out everything else.  Well, except for the peninsula which I decided to leave in place in case I ended up wanting to put some staging on it.  It wasn't until later that I got the idea of building an N scale layout on it.

The No-Name Industrial Park

The power tools came out, large chunks of benchwork were removed, lots of homabed glued into nicely easemented curves around the top left corner was chiseled off and construction got under way on what I hope is the final reconfiguration.  I also had a lot of rolling stock and a few engines to unload, which turns out to be a lot harder than you would think.  I've got a lot of stuff on consignment at a hobby shop, I donated some stuff to a local club, and I gave some stuff to a couple friends who used it to set up their own modular O scale switching layout. It's apparently relatively easy to sell used HO scale equipment, and N scale equipment is so small you can store a whole railroad's worth in a small drawer.  But with O scale, it was taking LOTS of space in the basement, and let me tell you trying to fit 12 O scale articulated auto rack boxes into the back of a car to take somewhere to sell is NOT easy, and you won't get anything else in that trip.  Just something to think about if you ever find yourself in the mood to buy lots of equipment up for a planned O scale railroad.

I'm pretty happy with the end result.  Operations are interesting even though I don't have the full complement of boxcars necessary for Northeast Container yet (I'm being cautious about buying more than I need, a few at a time until it seems like enough is the plan).  An op session takes somewhere between 45 minutes and 2 hours, depending on what needs to be done.  The layout feels much less crowded, which is nice too.  The smaller size also makes a little more likely that some day I might actually get some scenery done.

I've come to realize that although I really enjoy operation on some larger layouts with large yards and long mainlines, it's not what I want for my own layout.  I'd like to think I've arrived at a good lasting plan, but of course there is always that cynical little thought that pops out saying "yeah, right".  Time will tell one way or the other.

Mocking up the Palmer Industrial Park

It's been a long time since I built anything in N scale - I've been in O scale for a number of years now.  I was concerned that although my plan for an N scale Palmer Industrial Park looked good on paper, I might be surprised when I built it.  Since the benchwork was already done (re-purposed from an aborted O scale plan) I decided to mock up the N scale plan full size in place.

Trans Plastics mockup

The above photo shows my mockup of the upper side of the trackplan (see below).  The black magic marker scribbles around the Trans Plastics tracks are where the paved area will be.  The mockup is quite crude - just sticks of flextrack laid over each other and the switches are merely implied.  Still, it's enough to get an impression of what the layout will look like.  The wood sticks to the right are where the backdrop will be.

I made the mockup by printing the track plan out full size.  The first step for a full size printout is to add grid lines to the cad drawing spaced closely enough to ensure that there is at least one grid intersection on each printed sheet so you can slide them all into the right positions.  Since I was going to print on legal (8.5x14) paper that meant mostly 1 foot intervals with some lines at 6" intervals.  Planning is key to get the lines in the right places.  Or alternatively you could just use a smaller grid (say 3 or 4" spacing) and not worry about exactly where the sheets fall.  I printed enough sheets to get all of the interesting track, but not necessarily the pure straight stretches - I connected those up with a ruler and pencil later on.  In order to keep from mixing the sheets up, I numbered each grid line on a small scale printout of the plan, and then number the corresponding grid lines on each full size sheet as it came off the printer.

The next problem is getting a full size version of the grid drawn on the benchwork.  The plywood on the peninsula has at least 2 prior O scale plan variations drawn on it in heavy pencil lines, so trying to add a grid on top of that seemed to leave a lot of room for confusion.  I have a 3 foot wide roll of paper left over from the days when I had access to a large format printer on a BYOP basis (bring your own paper).  The layout happens to be exactly 3 feet wide.  So I rolled out paper down the length of the layout and securely taped it down to give me a nice clean surface to draw the grid on.  I used a laser line projector to get a reference line down the 16 foot length of the layout, marked it at intervals, then used my longest straightedge to pencil it in.  A piece of string would have done just as well (and in fact that's what I've used in the past), but I am a bit of a gadget freak and a project that uses a laser is inherently cooler than a project that uses a string.  A carpenters square got me the cross lines on the grid.

The last step in getting the full sized plan was to number the full size grid the same way I numbered the small scale one.  Then carefully position and tape down each printed sheet.  It's hard to get a photo that shows the lines well.  This one is heavily tweaked to bring out the contrast.  The printed sheets are around the edges where the track is, near the center is the straight grid on the roll paper.  The heavy black teardrop in the center is a mockup of where the backdrop will be.

Mockup with printouts only

A single line representing track, even printed full size, still doesn't make it easy to get a sense of what the finished layout will look like.  To get a better idea, I used double stick tape to stick down flextrack along the lines.  I trimmed and used rail joiners to make the main track in the park runnable, but all the other tracks are just stuck down in the right place.  Making the main track operable let me test pulling and pushing 25 car cuts around the 16" radius end curve.  And it let me get a real sense for how long certain things would take, such as doing a runaround, then shoving to Maple Leaf at a scale 10 mph (the top speed permissible in the real Palmer Industrial Park).

Here's a few more photos of the mockup.  First is an end view taken from a similar angle as the black and white photo above.

End view

Next is a shot of the other side of the layout.  This shot shows the lead (nearest the edge), the runaround, the main track with 20 50 foot boxcars positioned between the clearance points for the runaround switches - the plastic pellet cars in the foreground would be sitting on the near switch - and the Quaboag track.  The two red cars at the far end of the Quaboag track are 75 foot center beams in the lumber unloading area.  You can make out the two Quaboag building mockups, although it's not easy since I used white paper and foam core for them they don't exactly stand out from the white paper on the layout surface.

Lead, runaround, and Quaboag

Finally here's a photo of the Maple Leaf Distribution mockup, again a little hard to see in the photo because everything is white.  The photo does give you a sense of the size of the building, which is about 46" long.  Should be very interesting to switch, especially if the second track isn't stuffed full.

Maple Leaf Distribution

I'm now much more confident that my plan will actually produce something that looks like I expect.

Current state of the layout is pretty much as shown.  I've boxed up and stored all the rolling stock, and unstuck and stored the track.  The next step is to cut out holes in the center of the paper where backdrop vertical supports will go.  And then carefully roll up the paper will all the little sheets still taped in the right places.  Once the backdrop supports are fastened to the plywood, I'll cut and position the 2" thick foam down each side, leaving just a little space to slip the backdrop in.  With the foam glued down, the tricky bit will be to roll the paper plan back out, dropping the holes down over the backdrop supports, and get it re-located properly and any errant sheets repositioned and secured.  A pounce wheel should make it easy to transfer the track center lines from the paper to the foam.  Once I'm sure that's right, the paper gets rolled up again (carefully, just in case), and I can go over the pounce wheel marks with a marker to make them more easily visible.  Then mount the backdrop, and lay track.  Sounds easy... but time will tell.

However, before I take the next steps on the N scale layout I'm taking the time to get the O scale layout running again.  You may have noticed stuff piled on it in the photos.  Those piles were on the future N scale layout before I started the mockup.  Now they need to be dealt with for real.  There's nothing like having a running railroad to keep the inspiration level up.

The Palmer Industrial Park

I'm in the early stages of building an N scale layout representing the Palmer Industrial Park, served by the New England Central railroad.  Why a separate second layout?  I got to thinking about it because I found myself wanting to model some larger industries than could be fit in my space in O scale.  I also wanted to keep the O scale switching layout, which meant the remaining available space was too small to model larger industries in HO scale.  Hence the choice of N scale on a second layout.

In addition to modeling larger industries, I also wanted to get more of a prototypical feel to the track arrangement.  I used # 5 turnouts on my O scale layout - unrealistically sharp.  After spending some time with satellite photos measuring angles of switches on industrial track I was interested in, I came to the conclusion that # 10 switches were just about right.  I verified this with a tape measure on a field trip to a location with a switch in a parking lot I could measure the frog taper on.

In real life the Palmer Industrial Park is in Bondsville, MA, just north of Three Rivers and Palmer.  It's worked by a local out of the NECR yard at Palmer.  I'll write a post about the prototype at some point, but for now suffice it to say that I couldn't squeeze in an exact replica of the park.  So I took all the interesting pieces, shuffled them around to fit, and used a little imagination.

Trackplan for the Palmer Industrial Park

The peninsula is 16 feet long, and 3 feet wide at the left end.  It's slightly odd shape at the right (wall) end lets it fit between the door and the O scale switching layout.

Edit 1/18/2015: I realize I neglected to include a few details here.  All turnouts are #10. The big curve around the end is 16" radius, the minimum radius everywhere else is 24", and most curves are much bigger radius than that.  The easement offset is 0.25".  Track and switches are Atlas code 55.

The park has a long lead which comes in from the NECR main at Barrett's.  The lead is the track next to the aisle marked Lead on the track plan above.  The runaround and it's tail track head off back to the right and disappear into the weeds - the presumption is there used to be more industry over there but it's now abandoned.  The runaround holds 20 50 foot N scale cars.  One option I considered during planning was to put the runaround on the lead and eliminate the "extra" track.  However with the potential for 15-25 inbound cars and 15-25 outbound cars in a session the additional space will come in very handy.

The only industry on the lead side of the peninsula is Quaboag Reload (pronounced "kway-bog").  Quaboag is patterned after a combination of the now defunct Quaboag Transfer that was a large customer in the real Palmer Industrial Park, and Wildwood Reload - an industry on the Massachusetts Central RR in South Barre, MA.  Pretty much anything that needs to move between rail and truck can show up here.  Common loads will be lumber on centerbeams, wood pellets, paper rolls, and building materials in box cars, pipe in gondolas, sheet steel in coil cars, etc.  The first warehouse on the lead is completely enclosed, the second is open at the track end.  The large areas in between and at the end of the track are for unloading lumber, pipe, etc. that don't need indoor storage.

The industry just around the corner is initially going to be Cains - an N scale version of the same industry I have on my O scale layout.  What I really want to put there is another industry from the real Palmer Industrial Park - American Dry Ice.  However there are no N scale CO2 cars available at the moment.  Atlas makes some nice vegetable oil funnel flow tanks, so it's Cains for now, and if/when CO2 cars appear I'll replace Cains with American Dry Ice.

The next industry is Trans Plastics.  In the real Palmer Industrial Park this industry is at the end of the park track, but it seems to fit better when I've put it in my available space.  Trans Plastics has space for around 20 covered hoppers loaded with plastic pellets.  As with the real one when it was still in business, the industry will be close to full most of the time, but only release and receive 2-4 cars a session.  Since there are a lot of different grades of plastic in the various cars that get used up as needed, the cars released may be scattered around, so this industry will be interesting to switch.

The final industry is Maple Leaf Distribution.  In real life Maple Leaf now has 3 pairs of tracks serving 18 doors.  Before their addition of a few years ago, it was 1 pair of tracks serving 10 doors.  I plan to compress reality a bit to make the "original" building have 7 spots spaced for 50 foot cars (like the original 10 on the real thing), and the "new" building on the same pair of tracks have 3 spots for 60 foot cars (like the new spots on the real thing).  Maple Leaf receives lots of paper rolls and a number of other commodities in box cars and ships them out by truck.  They also receive recycled paper by truck and ship it out by box car.  They generally assign cars to doors to minimize movement through the warehouse.  There may also be respots (a car not completely unloaded needs to be moved to pull/spot another car, then returned to where it was).  A car that was on the outside track and not completely unloaded may need to be respotted at the same door on the inside track.  The combination of specific door assignments, respots, and the two tracks should make for some very interesting switching.

The main park track continues past Maple Leaf and fades into the weeds at the wall.  The almost 20 car lengths between the Maple Leaf switch and the wall will provide needed space for sorting cars and stashing outbound cars out of the way.

The current state of the railroad is bare benchwork, with a full size trackplan tapped to it (more on that later).  The benchwork is actually a re-purposed section of what was originally planned to be a larger O scale layout.

The No-Name Industrial Park

The No-Name Industrial Park is my O scale switching layout.  It represents a small industrial park somewhere in eastern/central Massachusetts.  I've had so many different ideas of what I want to call it over the past year that I've decided to just ignore the naming problem for now.  Therefore for the time being the layout is known as the No-Name Industrial Park.  The park is operated by a switching service - also unnamed as of yet.  The park lead serves as an interchange track with the nearby railroad - also unspecified (but which will be either CSX or Pan Am Southern).

Trackplan for the No-Name Industrial Park

The room is almost exactly 21 feet on each side.  For perspective, that's 1008 feet in O scale.

The connection to the outside world is in the lower right, where the tightest curve on the railroad (36" radius) curves out into the aisle where it says "interchange".  Following the main track from there, the first switch you come to is a 3 or 4 car storage track for Northeast Container.  Next is the switch for Midstate Recovery Systems, a construction and demolition debris recycling business.  Next is the switch for Northeast Container itself, followed by another short storage track.  Around the lower left corner is the runaround, which can hold 9 or 10 cars.  The runaround is a slightly more forgiving 44/48" radius.  Continuing up the left wall, the next switch is for Tighe Warehouse, and then a transload track.  The final switch is for Cains Foods.

I got the idea for a switching service operating the park from SMS Rail Lines, a company that operates several industrial parks in New York, New Jersey, and Pennsylvania.  It's a fascinating operation - you can spend quite a bit of time poking around their web site and studying google/bing maps satellite and aerial photos of the parks they serve.

Power for the park switching service is an MP-15 purchased second hand from CSX and given a patch paint job.  Or at least that's the plan - currently the Atlas O MP-15 sports factory fresh CSX paint.  The MP-15 is generally kept up near the Cains switch, where there's a small trailer office.  The switching service keeps someone on hand to respond to customer requests whenever needed.  So an operating session might be as simple as pulling an empty/load from a customer to replace it with a fresh car from storage, as complex as finding a dozen inbound cars on the interchange track that need to be shuffled into position, or anywhere in between.

Northeast Container is based on the Temple-Inland plant in Biglerville, PA served by the Gettysburg Northern RR.  The plant makes boxes.  It receives boxcars of pulpboard and an occasional hopper car of starch.  Some of the empty box cars get re-spotted at the outside dock to be loaded with scrap cardboard.  Jack Hill describes it here on his blog.  You can see it in the satellite view on google maps.  As soon as I read Jack's description of the operation at the plant, I knew it would be much more interesting than the feed mill I had originally put in that spot on the layout.  I will be using a motley collection of 50 foot boxcars, plus an as-yet un-acquired starch car to serve this industry.

Midstate Recovery Systems recycles construction and demolition debris.  This is a real industry in Portland, Connecticut served by the Providence & Worcester Railroad.  You can see it here on google maps.  This industry caught my eye when I saw this photo of it on NERAIL.  It's a relatively small building with a track going through one end of it - only two cars fit inside.  Debris arrives via truck and is dumped in the building.  It's then loaded into railcars.  If you're an O scale person used to not being able to get the type of car you need for an industry, you will probably have noticed that the cars in the photo are coalveyors (some with side extensions) - a car Atlas actually makes!  On the No-Name Industrial Park the building is positioned so the truck dumping portion of it will be implied by the on layout portion of the building missing the aisle side wall.  The building is carefully position to make it easy to see the clearance point for the runaround on one side, and to easily reach the caboose ground throws for the three nearby switches.  I sincerely hope my model doesn't collapse under the weight of a record snowfall as the prototype did this January - see photo.

The Tighe Warehouse is based loosely the Tighe Logistics Group warehouse in Mansfield, MA.  There are other Tighe facilities in the area.  The Mansfield one is at the lead end of an industrial park called "the chocolate" by the railroad for historical reasons, it's served by a CSX local.  You can see it here on google maps.  On my layout most of what this warehouse receives will arrive in NS 60 foot Berwick hi-cubes (Atlas), plus a few FBOX 50 foot hi-cubes (MTH).

The transload track was inspired by J.P Noonan operations I've seen in Mansfield and Leominster, MA.  The track serves local industries that don't have rail service of their own.  There will almost always be several plastic pellet cars here, and often a hopper or two of soda ash for a nearby water treatment plant.  The plastic pellet cars are old Weaver cars - good enough to get the point across but that's about it.  The soda ash cars are a couple MTH hoppers.

Cains is based pretty closely on the Cains plant in Ayer, MA.  You can see it here on google maps.  The plant receives tank cars of vegetable oil.  There are 4 unloading spots inside the chain link fence on their siding, which is served by Pan Am Southern.  The piping leads to a tiny cement building next to the track, from which it apparently proceeds underground to the main building.  On my layout the main building is presumed to be in the aisle, so all I need to model is the little cement block building and a suggestive patch of asphalt in the parking lot (see the satellite photo).  I'll be using Weaver 50' tank cars for this industry even though they aren't right - unfortunately there doesn't seem to be a decent funnel flow veg oil tank in O scale.

The current state of the railroad is ugly but functional.  I'll post some photos sometime soon.  And I'll also post something about that blank peninsula sticking out into the room from the right wall.

Ground throws

I like to use Caboose Industries ground throws for turnout control.  They're simple and reliable.  And they don't require you to hunt around on the fascia for controls.

The way caboose ground throws are designed to be installed is sitting on top of the head blocks, with the pin going down throw a hole in the throw bar.  Here's a 208S ground throw mounted the way caboose industries intended on wood ties on the older section of my layout.

Caboose 208S mounted as designed.

That works great for hand laid turnouts, and it looks OK too even though it's knee high to an O scale person.

Atlas turnouts don't have long head block ties, so some form of extension is needed.  While I was experimenting with that, I came up with what I think is a better approach to get both extended head blocks and less conspicuous ground throws.

Inconspicuous ground throw on Atlas O turnout.

The red paint on the ground throw handle indicates the turnout is reversed. The opposite side is painted green for normal. It's a nice idea I got from a friend that helps orient visitors.

Inconspicuous is achieved by mounting the ground throw at roadbed level instead of tie level. Since I used homabed, I notch out the beveled edge by the throw bar and fill in a wood block to match the roadbed height. The pin sticking down from the ground throw is cut off, and I drill and tap a small hole in the end to accept a small screw. The throw bar on the turnout gets shortened a little - pretty much just clip off the part on the end - so it ends up almost flush with the ends of the ties when you push it so the opposite point hits the opposite stock rail.

Modified ground throw ready to install.

Then attach the ground throw to the throw bar, center the points, center the ground throw handle, and mark where to drill holes to screw down the ground throw.

Center everything to locate holes.

And here it is in working order.

Working but ugly.

At this point you could stop, but it's not pretty. Once I'm sure it's lined up and working correctly, I add the head blocks. Since the ground throw is at roadbed level, they need to go around the ground throw. I clip out a pair of ties from some Atlas flex track leaving the bit of plastic that joins them in place for the time being, and cut off the other end at the tie plates. Here's a pic of it at that stage sitting where it needs to go, but as you can see it doesn't fit down over the ground throw. Yet.

Test fitting head blocks.

Using a burr in a dremel tool, a razor saw, an xacto knife, a mill file, and a rat tail file I remove everything that seems to get in the way of a fit. The plastic bit joining the two ties keeps you honest about alignment at this stage. Here's a shot of it sitting in place after I made it fit.

It fits!

Here's a shot of what the bottom looks like. The notches get around the "hips" of the ground throw, and the cross groove in the middle clears the tops of the #2 wood screws I mounted the ground throw with. Since my wood block ended up just a smidgen above the roadbed level (oops) I filed the entire bottom a little.

Bottom view of trimmed head blocks.

The final step is to cut off the other end of the ties at the tie plates. I used CA to glue them to the ends of the switch ties and the roadbed. Being VERY careful not to get any one something that's supposed to move! It would be nice if the ties were just a bit longer, but you can't get a longer molded plastic Atlas tie without mangling a switch - something I'm not willing to do. And I think the fact that the head blocks are the same material and surface texture makes them blend in better than using wood or raw styrene shapes. Here it is glued up.

Head blocks glued on.

All that's left to do is use some putty around the edges to keep ballast and glue from getting in where they might cause trouble.

The ground throw sticks up above rail height only about 1/8 inch, so it can be pretty close to the track with no concern about pilots or whatever hitting it. Looks better at that height too.