Hi all,
Following the completion of my 2-6-2 saddle tank, there's only one empty spot left around my layout's turntable. Given that it's 1948 in my On30 slice of Oregon, I decided that this last locomotive would be a diesel. I didn't want to do a small 'critter', as I'd tried one before. I hadn't been able to get it to perform the way I wanted. As well as this, quite a few of the On30 diesels I've seen are critters, and I wanted to do something a bit different. With this in mind, I decided that BVLC No. 10 would be a narrow-gauge road switcher.
Since I started modeling in On30, I've attempted 6 conversions of HO scale models to On30, only 3 of which have worked out. All three of the working ones used older mechanisms, all of which had metal chassis' and robust drive trains. As such, I decided this diesel project would use an Athearn Blue Box mechanism. Via Ebay, I was able to obtain a Blue Box GP40-2 in new condition.
After checking to see how it fit on the turntable, I tested the mechanism and stall current. It ran smoothly, and the stall current of 0.66A meant that I'd be able to reuse the TSU-750 decoder I had in the aforementioned critter. With this sorted, I started designing the new diesel.
My plan was to 3D-print the body, then add details as needed. After an evening with the calipers and the GP40-2, I was able to put an outline of the mechanism in Blender. As usual, I colour-coded the components. Red is the metal chassis, and the yellow is the space taken up by the drive train. I also imported the pilot and coupler box I'd modeled for the 2-6-2T, in order to get the size and positioning for them. I also added shapes for the decoder and speaker, after this render was taken.
I've always liked the look of the early Alco road switchers, particularly the RS-1. That became the inspiration for my design. After doing the basic body shell, I realised that it was using quite a bit of material. So I uploaded it to Shapeways to get an idea of how much it would cost. I was going to get the shell printed in 'White Strong and Flexible' plastic, but as designed it was a lot more than I'd budgeted for.
As such, I cut back on the amount of material needed. Instead of having a slight curve along the top and down the front, I decided to make the top and sides flat. This could be done using the 1mm thick styrene sheet I already have. After removing most of the material used in the hood, I created lips around the edges to hold the styrene.
However, this was still a little pricey. Further to this, the 'White Strong and Flexible' plastic has a grainy finish when printed, and needs to be sanded smooth. Given that I haven't done this before, I was a little concerned about ruining the curve of the rounded parts. As such, I came up with a third option. Instead of rounded edges for the hood, I would have a 45-degree angle on each of the corners. No support was needed for the top panel, as it would be resting on the supports, front and side panels. I also did a variation with the curved edges, to see if the reduction in height and removal of some additional material (e.g. the cab steps) would reduce the cost to something a bit more friendly on the wallet. I also added drill starter holes for the coupler lift brackets, and four screw holes to mount the footplate to the chassis.
As well as the body shell, I also modeled several detail parts, which I coloured green for clarity. These will be 3D printed in 'Frosted Ultra Detail' plastic, which has a smoother finish and higher resolution than that used for the body shell. Using the steam loco pilot as a guide, I created a coupler box at the front. After measuring a 3mm LED and re-creating it in Blender, I created two headlights. The rear one is designed to be glued onto a vertical side, and the one at the front was designed to go onto a horizontal hood. I also designed a radiator, and some side doors for the long hood.
With that done, I separated the cab from the body for printing. Locating tabs were added to the underside of the cab, and corresponding slots in the footplate. I also added holes along the sides for the handrails. I'll be making these out of wire when the time comes. With that done, I uploaded the body and details to Shapeways.
Unfortunately, there were some issues with both the body and the details. On the walkway, the walls between the handrail holes and the edges were too narrow, as were the walls between the lengthwise slot for the cab and the edge. The sharp corner on the corner posts was also too thin. After correcting these, I sorted out the detail parts. The only problem there was the hood top headlight. In order to fit the LED into it, some of the walls had to be made thinner than the minimum tolerances for the Frosted Ultra Detail plastic. Increasing them would have made the headlight too large, and so I decided that both the front and rear headlight would be mounted on the vertical end panels. As such, I printed two of the rear ones.
With that sorted out, I ordered the body and parts through Shapeways. I expect I'll have my next update in a month or so, once they arrive.
The Location: Forests of the Pacific Northwest, OregonThe Year: 1948The Scale: On30The Blog: http://bvlcorr.tumblr.com
I'm looking forward to seeing this project develop. The loco has nice lines, which you've nurtured despite the issues with the very long learning curve often found with 3D design and plastic engineering.
Mike Lehman
Urbana, IL
There has been a bit of progress on my On30 diesel, although not exactly what I was expecting.
After receiving the first set of 3D printed parts, I test-fit them to the chassis. Unfortunately, it seemed that some of my measurements were a little off, and the walkway didn't fit onto the chassis. However, the cab did fit onto the walkway as intended.
There's a 'bump' on the chassis where the cab of the original model went. This was approximately 4mm too far back, so I moved that end of the walkway forwards to correct this.
There were also some issues with the detail parts. Although the 3mm LEDs did fit into the headlights, it was a very tight fit and left no room for the thickness of the paint. Given that the headlights do need to be painted to prevent the whole part from lighting up, this wouldn't exactly work. As such, I enlarged the holes for the LEDs, adding 0.1mm onto the diameter. The bolt detailing was also enlarged, as it wasn't distinct enough on the original. I also thickened the side doors, as they were originally too thin and had warped as a result.
With these corrections done, I uploaded the parts to Shapeways and ordered them. But even as I did this, there was something bugging me about the model. My original intention had been to do a 'narrow gauge RS-1', as an homage to one of my favourite diesel designs. However, due to circumstances outlined in my previous post, the design had evolved into an angular, low-nosed road switcher. Given that the early diesels (e.g. GP-7, GP-9, RS-1, etc.) had high hoods at front and rear, this design seemed to me to be too modern for a layout set in 1948. As such, I decided to redesign the locomotive as a boxcab rather than a road-switcher. As this design was introduced in the 1920s, it would be more appropriate for my era.
I started by redesigning the footplate. Originally, the footplate just extended from the side of the chassis with the cab sitting on top. This resulted in a cab roof that was right on the height limit for my layout (as determined by the backdrop cutout). Given that all the boxcabs I'd seen had quite a bit of equipment on their roofs, I'd need to lower the outside walls. I also wanted to give the boxcab a bit of a 'hunkered down' look. The footboards at each end were also added.
I brought the ends and sides of the original walkway in, ensuring that they would clear the fuel tank. I then dropped them by 8mm and created a 2mm 'ledge' around the outside, upon which the body of the model will be resting. I only put a few supports between the ledge and the walkway, enough to hold it in place. These will be covered up when the body is in place. A small section was added at the ends to accomodate the coupler box I had made for the original version of the diesel.
With that done, I did the body itself and the roof (not shown). I'd noticed that most boxcabs had a distinct shape to the front wall, where it curved over then seemed to run flat underneath the roof. As such, I made the flat bit on eiher side 1.5mm thick, then extruded it back to form the side walls. With the four walls in place, I then put in the holes for the windows and doors by using a boolean operator to 'cut' them through the walls. I did the same for the large radiator grilles from the original model. Given that I already had these parts printed and ready, I decided to use them.
My next step was to create the doors and window frames, to be printed in ultra detail plastic. I also copied and shrunk the radiator grille, cutting holes for four of these - two in each side. I also added an interior wall, just behind the side doors. This was mainly to block off the motor and electronics from the front cab windows - the side and rear windows will be blacked out. However, it also adds a bit of reinforcement to the body.
I then turned my attention to the roof. The roof details are based on what I could see from the prototype photos, most of which were taken from the ground level. I have a few spare details (air tanks, etc.) in my parts box that I'll be adding to that flat area.
The ends of the walkway were also extended, to accomodate the coupler lift bars. Drill starter holes were also added for the coupler lift bars, handrails beside the doors, and a ladder up each side.
Owing to the height of the cab and the end doors, I wasn't able to re-use the headlights I'd printed for the first version of this locomotive. Those were designed to go on a vertical surface, but the boxcab needed roof-mounted ones. I had designed top-mount headlights for the original diesel. These were designed to have a 3mm LED inserted from the back, with a cover plate closing off the rear. However, the cover plate was too thin for Shapeways to print.
This time around, I attacked the problem from a different angle. After lowering the original top-mounted headlight, I sliced it in half horizontally. This allows the LED to be inserted from the top. As each piece of the headlight housing is larger, there are no problems with it being printed. It was also adjusted to match the curvature of the roof, and mounting pegs were added, with corresponding holes in the roof itself. A hole was also cut through the headlight and roof, to allow the wires to reach the LED.
My last step was to prepare the parts for printing. After separating the shell into its three components, I added some sprues to the body to reinforce it during printing and shipment.
The details were duplicated as needed, and laid out ready for printing. They and the body were then uploaded to Shapeways, and I expect to have them in a few weeks.
There are a few 'leftovers' from the original version of this loco. Namely, the cab, two walkways, the wall-mount headlights and the hood side doors. I'll be re-using the cab and headlights on a locomotive I'm building for my train club, for a mining line that just needs an engine to shuttle back and forth. Given that the 'white strong and flexible' plastic has a bit of a grainy finish, I'll be using the walkways to figure out how to smooth it out before doing the actual model. I'll put the hood side doors into my parts box - I could use them on a structure or another project in future.
Part 3: From Render to Reality
It's been a while since my last update on this project. But in the month or so since I received the parts, I've finished off the locomotive.
From left to right are the frame, body and roof, which were printed in White, Strong and Flexible Plastic. It was the first time I'd worked with this material. Although it is in the name, I was surprised by just how flexible it was. The thin beam across the top of the side doors was able to almost bend back on itself. To the right of the main body components we have the detail parts. For some reason, they printed three of the exhaust stacks and headlights, even though the model file had only contained two. The large vents and coupler boxes were originally printed for the 'road switcher' version of the model, and are being reused here.
I started by preparing the mechanism. After removing the motor and original headlight mount, I test fit the frame. Some minor filing was necessary to get it to fit properly.
With this done, I used the frame as a template to mark out the four holes for the screws. I used M2x5mm screws, two of which can be seen poking up through the holes in the bottom-right of the photo.
My next step was to install the DCC decoder. I started by installing the wires for the motor and power pickup. On the motor, I swapped the metal contact tabs around, so that the one with the tabs bent out was on top. This prevented it from shorting against the chassis, and provided something to solder the grey wire to. I was unable to solder a wire directly to the chassis, owing to the massive mass of metal acting as a giant heatsink. Instead, I drilled out a hole for a small brass screw, just in front of the motor. The black pickup wire was then soldered to this.
With that done I reinstalled the drive train. Along the bottom of the chassis is a bare section of metal, which originally conveyed current to the motor. I covered this with a piece of electrical tape.
A test fit of the body over the mechanism confirmed that the body cleared the mechanism.
All that was left was to install the decoder. To hold it, I constructed a shelf at the rear of the mechanism, long enough to hold both it and the speaker. There's a hole in the underside of the shelf, to allow the sound to pass out via the rear truck.
The decoder itself is a TSU-750, with the Alco 244 sound set. I'd measured the stall current of the Athearn motor before starting the project, and had been pleasantly surprised to find that it only drew 0.66A.
When designing the doors, I had included pilot holes for the door handle castings. These are from Precision Scale Models, and require a 1mm hole for mounting.
While the doors were drying, I mounted the body to the frame and glued it in place. I also filed a notch in the top of the internal cab wall, to clear the headlight wires.
Unlike the Frosted Ultra Detail plastic, the White Strong and Flexible plastic has a rather coarse texture. After reading a few posts about how to achieve a smooth finish, I decided to use a product called 'Mr Dissolved Putty'. This is putty dissolved in liquid, and has about the same consistency as enamel paint.
I first coated the body and roof in primer, to better see the texture of the parts. This also has the advantage of sealing the White, Strong and Flexible material, which is porous. Over the next two weeks, I applied three coats of the Mr Dissolved Putty to each side of the model, leaving it resting with that side up overnight to dry. It only took me a few minutes per evening. Nylon brushes were used to apply the putty, and these were washed off with turpentine afterwards. I didn't apply the putty to the inside or underside of the body, as these won't be seen when the model is on the tracks.
During this period, I also painted and weathered the truck sideframes.
Once the putty had set, I sanded it down to get a smooth texture. I used three grades of sandpaper to do so - 180 grit, 240 grit and 360 grit in that order. When finished, I painted it with another coat of primer. This revealed a few rough spots that had been missed, and these were addressed with another round of sanding. After this was sorted out, I drilled out the holes for the grab irons and other details.
I formed the grab irons out of 0.8mm copper wire, and the coupler lift bar out of 0.5mm copper wire. The mounts for the coupler lift bar were also Precision Scale Co castings. With these in place, I added the Frosted Ultra Detail 3D printed parts. This was where I ran into a problem. When arranging the parts for printing, I'd miscounted how many side windows I had. There were actually six, however I'd only printed four side window frames. In order to avoid having to order extra parts, I decided to use the 3D printed side window frames on the end windows.
For the middle windows, I scratchbuilt frames out of styrene strip. I glued some 1mm square strip on top of some 3.2 x 1mm styrene strip, along the edge. Once this had dried, I cut it to match the width and length of the window frame, and glued it in place from the inside. The top part of the 3.2mm strip stuck up above the top of the body, but this would help hold the roof in place.
My next step was to add some 3.2 x 1mm styrene strip along the top of the rear of the body, to help hold the roof in place. With this done, I painted, lettered and weathered the model. I used the same weathering technique I'd used on my work caboose and passenger coaches. After sealing it, I glazed the windows with plastic cut from an Amiibo box.
Although the cab windows were left clear, the rear windows were painted black from the inside to conceal the mechanism. The glazing was held in place with PVA glue, and reinforced with two-part epoxy. I had constructed a 'cover' for the front part of the mechanism inside the cab, to hide the drive train from view. As such, once the roof went on, the inside of the cab would be inaccessible. Hence, I needed to ensure the windows wouldn't fall out. I used epoxy on the driver for the same reason. The brass bar at the rear was just temporary, to hold the sides of the model apart while the roof was installed. One of the sides up that end was bowed inward slightly.
While the epoxy dried, I painted, weathered and wired the roof. To weather it, I initially painted the roof using Tamiya 'Gunship Grey'. Once this had dried, I took a paper towel, put some black weathering powder on it, and rubbed it over the roof. Rubbing in a straight line produced a streaky finish, while rubbing in circles gave a grimy look. The headlights had warm white LEDs installed in them, and these were wired to a plug in the rear of the body. As they're the only wires between the body and the mechanism, this will allow me to remove the shell should it be needed in future.
I then stuck the roof to the body using CA run along the top edge, and some rubber bands held it in place while it dried.
After the roof had dried, I installed the body on the mechanism and tested it on my programming track. It checked out, however a couple of issues manifested when I tried it on my layout. The pilot boards were just low enough to catch on the turnouts. To resolve this, I raised the body by placing some 1mm thick styrene strip along the top of the chassis. I also had to shorten the middle pair of steps, as they were low enough to catch on the turnout frogs and cause short circuits. Once they were brought to the same height as the front ones, the engine ran without a problem.
Originally, I said that this was going to be locomotive no. 10 of the Bradford Valley Lumber Company. However, I decided to give it the number 11. While working on my layout, I'd figured out how to fit an extra track around my turntable, which gives me space for another engine. Of course, this means I'll now need a number 10. Let's just say that I like the look of the On30 2-4-4-2, and that it could be a rather neat tank engine.
Have you thought of letting shape ways sell that as a kit? That is a nice boxcab
A pessimist sees a dark tunnel
An optimist sees the light at the end of the tunnel
A realist sees a frieght train
An engineer sees three idiots standing on the tracks stairing blankly in space
Some changes in your plans, but a very respectable result nonetheless.
Had a couple of thoughts that came to mind since I originally read the thread. Be cautious with the TSU-750 and Athearn motors. Athearn's older motor had a rep as having higher amp draws. Yours may be a good one in that respect, but its draw doesn't leave a lot of headroom vs the .75 amp rating of the Tsunami.
The TSU-750 can be a bit of a delicate flower. I've toasted -- or shall I say my operators have toasted -- two of them in Blackstone C-19s. Don't let them sit and spin would be my first advice. The second would be to keep a close eye on decoder and motor temps as you test run them in the regular service you exect to use them in. If they prove reasonably workable in the Athearn chassis, stll watch out for wheelspin. As a protective measure, I dialed my DCC output to the track back to 12.5 V from the more typical 14.5 v output. This didn't save all of mine, but made them a lot less delicate.
On a happier note, I can recommend another great HO diesel chassis, the Kato NW2, for conversion to On30. I narrow them to HOn3, then bash into a sort of roadswitcher that really benefits from the heft built-into the Kato chassis. You can just run them gauged as is, of course, and still have a reliable short diesel chassis.
Mike,
Thanks for the heads-up on the decoder. I mainly used the TSU-750 as it was what I had on hand, and I will keep an eye on it. My DCC track voltage is 12v, and I usually keep the engines turned off around the turntable when not in use. Given they all have sound, it could be a bit overwhelming otherwise
Incredible work and extremely well documented. Thank you for sharing the process with us. This is locomotive scratchbuilding in the 21st century, and I love it! I really think 3D printing is going to be the tool for a new era of scratchbuilt models of all kinds, including new and realistic figures.
Jim
Modeling the Baltimore waterfront in HO scale
tbdannyMy DCC track voltage is 12v...
Ah, good, that will go a long way toward alliveating the potential for problems. Both my failures were due to injudicious spinning of drivers. Should be OK if full slip draw is less than decoder rating, but not if the operator is obstinate about making the grade even when it's obvious they're not going to be able to do it. I assumed that Blackstone/Soundtraxx got as close as they could get on this model, gven it was an in-house project, but sustained abuse will exceed what it's capable of.
Any heatsinking you can couple to the board will also help keep it cool.