Hi all,
I decided to build a new locomotive no. 8 for the Bradford Valley Lumber Co, as I wasn't satisfied with the current one (a 35-ton diesel critter). I've always liked the look of the Bachmann On30 2-6-0, however as you can see below, the locomotive itself already takes up most of the turntable.
As such, I decided to convert it into a 2-6-2T saddle tank engine. Although not following a particular prototype, it's inspired by this Alco engine: http://narrowmind.railfan.net/262T-ALCo30-42.jpg.
After figuring out how to get the shell off, I removed the domes and other details. These were put aside for later use.
My next step was to construct the saddle tank. On the original model, there's a short section in front of the cab where the running board is at the same height as the cab floor, then the rest of it runs a bit higher. As such, my original plan was to have the saddle tank stop at the end of the higher section, then bring the cab forward to meet it.
I obtained some 4cm diameter PVC pipe. After cutting that to length, I cut it in half lengthwise. The resulting pipe was a little bit low, so I decided to raise it by about 5mm. To this end, I used some 3mm x 1mm and 2mm x 1mm styrene strips. I glued them edge to edge, then cut these to length. Once this was done, I glued two of these sections back-to-back, with the 3mm strips overlapping the join between the strips on the other section. After these dried, I glued them to the base of the pipe with CA. Once that had dried, I glued the front of the pipe/styrene assembly to a sheet of 1mm styrene.
With that done, I turned my attention to the body shell. I cut a slot just in front of the first boiler band, using the band as a guide. Once that was done, I glued a layer of 1mm styrene over the side running boards. This covered the holes in the original footplate, and widened it to match the cab floor. I also cut out the plastic betwen the two holes for the dome, then used a nibbler and file to expand the hole and clean up the edges. This is to allow the DCC decoder to sit on top of the mechanism, where it will be covered by the saddle tank.
By this point, the pipe had dried on the styrene sheet, and I cut the styrene to match. I then test-fit the pipe on the body. After looking at it in place, then at the inspriation photo, I realised that this would make the cab too big. To be precise, the cab front would be too far forward.
I decided to extend the upper running board to the front of the cab. Using the original cab as a guide, I cut some 2mm square styrene strip to length and glued it under the styrene I'd added earlier. This strip runs the whole length of the running board. This was done on both sides.
Once the glue had dried, I added an additional section of 1mm styrene to extend the running board.
While that was drying, I built a longer tank using the pipe, as I had before. I only used 3mm styrene strip on the bottom this time, as the original tank had been a little too high.
On some of my previous builds, I had experienced problems aligning the various components of the body. This time, I decided to use styrene strips to help align the tank. To this end, I glued three styrene strips on each side of the footplate. Two are 2mm x 1mm, the other is 1mm square. The tank fits over the tops of these, and is centered.
In order to provide a smooth surface on the outside of the tank, I laminated the tank with two layers of 0.25mm styrene. I marked the centerline on the outer layer, as well as two lines where the handrails are to go. I then test-fit it to the body shell again.
I then turned my attention to the chassis. I'd seen a similar conversion project on another website, which had used the rear mounting screw to hold the rear truck on. However, I wanted to have the rear truck the same distance from the rear axle as the front truck from the front axle. To this end, I drilled a hole in the base of the chassis even with the tyres of the rear axle. As I do with any chassis/mechanism modifications, I test-ran the loco immediately afterwards.
I then test-fit the rear truck. I had purchased it from the Bachmann parts store, along with the spring, screw and a few other pieces.
My attention then turned to the detailing. I didn't have any dome castings on hand, so I decided to re-use the original ones. Due to their height, these had to be cut down. In the photo above, an unmodified dome is on the left, and the cut down one is on the right. I used the 'lip' around the top and bottom to guide the razor saw as I cut out the middle section. I then glued the top and bottom sections together.
As I waited for the domes to dry, I installed the DCC decoder. I had measured the stall current of the motor, and found that it drew 1.05A. Normally, I use the Soundtraxx Tsunami decoders, but this current draw would blow the motor output on a TSU-1000. I found that the TCS WOW decoder could take 1.3A, so I decided to use one of those instead. I did consider a TSU-1100, but the TCS was better for my budget.
I mounted the decoder directly on top of the motor, using double-sided tape. With it in place, I was able to wire up the motor and power pick-up leads. I had to apply some masking tape, to isolate one of the motor pickups from the chassis.
Once the domes had dried, I applied body putty to fill in the cracks. I used a section of the original pipe with some sandpaper around it to sand the base of each dome to the contour of the tank. I then glued them in place, along with a water hatch. I then test fit the tank to the body.
This was the first time I'd fit the tank while the body was on the chassis. It turned out that the styrene wall at the front of the tank needed a slot cut in it, to go over the metal chassis. Once this was done, it fit without any issues.
I then tested the locomotive on my programming track, and it responded as expected.
My next step was to glue the tank to the body, and start detailing. I left a hole in the middle of the rear dome for the whistle. The original generator was re-used, although I replaced the original plastic exhaust on it with a more robust one made from wire.
For the headlight (and rear light), I rescued two Bachmann lights from a former project that didn't work out. I decided to make the front headlight an oil-burner that had been converted to electricity, and cut the exhaust at the top off. The hole left will be filled by body putty. I also drilled a hole immediately behind the headlight, for the bell. I'll be re-using the original bell from the 2-6-0.
My final step for this part of the build was to install plugs for the speaker and lights on the body shell. I just used some Single-In-Line IC sockets, cut to length. I decided to have two plugs of different sizes, wired in such a way that an accidental mis-connection wouldn't do too much damage.
At this point, all the work I had done was physical, and I hadn't done anything on the cab. Having bought a couple of items from Shapeways, I decided that the cab would be done with 3D printing. In part 2 of this write-up, I'll be covering how I designed a new pilot and cab in Blender, and arranged them for 3D printing.
The Location: Forests of the Pacific Northwest, OregonThe Year: 1948The Scale: On30The Blog: http://bvlcorr.tumblr.com
Hey tbdanny!
Very interesting project!
You are very brave. I'd have a hard time cutting into the original 2-6-0 but you will end up with a unique locomotive. I'll be curious to see how the 3D printed cab turns out.
Dave
I'm just a dude with a bad back having a lot of fun with model trains, and finally building a layout!
That surely looks like a fun project, especially in O scale. I look forward to progress and final photos.
Another option for the water tank is side tanks instead. It might be easier. Though, if you're following a prototype or just like saddle tanks, that's different. Anyway, my first steam loco was this little fella (as an example of a small side tank loco):
It's HO by the way. Made by Tenshodo. The only one of their products I could afford at the time.
By the way, the biggest "flaw" in the model is that it only picks up on two wheels on the left side (middle and rear). If you ever get one of these, be sure to add another pickup for the front wheel. And/or a keep alive.
Nice work!!!!!
Ed
Ed,
I have done a side tank model before, the original no. 7 for the BVLC:
However, I ended up having to do them three times over and they were slightly uneven. I also liked the look of that saddle-tank prototype, so I decided to go with that this time.
Yup.
Nice little fella.
That saddletank looks really good, and it looks like it'd be a good place for some extra weight if your operations would benefit from such an addition. Personally, I prefer headlights on the smokebox front, but you've obviously got a "look" in mind, and I'm looking forward to seeing it. Very nice work.
Wayne
Looks really good, tbdanny! Those unhappy results previously were good learning experiences and it really shows here. Looking forward to seeing things finished and painted.
BTW, Bruce Pryor's site is a great sources of inspiration for anyone needing something a bit different, like logging, mining, or other similar application.
http://narrowmind.railfan.net/
If you have pics of any export loco built in North America, he'd likely be interested in hosting it. I've got one of my pics on there. Best we could tell it's a GE U?B:
Mike Lehman
Urbana, IL
As I mentioned in part one of this project, I decided to use 3D printing via Shapeways to have the front pilot and cab made.
I have made 3D models of locomotives before, for Trainz Railroad Simulator. This is where I follow my interest in British railway modeling, in 'V scale', so to speak . Amongst these models were a few tank engines, and I realised that I could adapt the same techniques I'd used for their cabs to make this model.
I started with the front pilot. On the Bachmann Mogul, the cowcatcher is held onto the front deck by two small screws. As such, I decided to make a 'switcher' pilot, which would be a screw-in replacement. To this end, I measured the dimensions of the cowcatcher, and re-created the rear clips, screw holes and panel in Blender. Once this was done, I measured a Kadee No. 5 coupler box and created a 'dummy' one in Blender, to aid with construction.
Blender does measurements in what is called 'blender units'. They are to no particular scale, and so one blender unit can represent one foot, one metre or one millimeter, as needed. For my Trainz models, I usually use one blender unit to one metre, so no scaling is needed when exporting for the simulator. For Shapeways models, the unit is specified when you upload the model. As such, I worked with the scale of one unit = one millimeter for these Shapeways models.
Once the rear panel of the pilot had been created, and the Kadee box positioned correctly, I started creating the front of the pilot. I started with the coupler mount, making it 1mm thick around the Kadee box. I added a 2mm hole underneath, to allow the mounting screw through.
Once that was done, I added the two steps at the front. I originally had these as one step across the pilot. But when I looked at it, I realised that the single step would block access to the screw hole for the coupler. Hence, I split it into two. After this had been sorted out, I added the bolt heads. I just created one, then used Blender's array tool to duplicate it as needed.
I had measured the height of the original cowcatcher pilot above rail level, and positioned the top of this pilot at the same level. This allowed me to leave some clearance (2.5mm) between the bottom of the pilot and rail height.
After doing the pilot, I started working on the cab. My first step was to measure the cab end of the model with a pair of calipers. Using these measurements, I was able to re-create the original cab floor and boiler in Blender, as well as the location of the rear pilot truck. As with the pilot, I defined the coordinate of 0 on the z-axis (vertical) to be the rail height. I coloured the model body red, to indicate that it couldn't be changed.
My next step was to import the pilot that I'd made previously, and rotate it 180 degrees. It was already at the correct height. I also deleted the attachments from the rear, as they won't be needed. This rear pilot will be part of the 3D-printed cab floor.
I wanted to have the cab floor drop to just above the rear pilot, for most of its length. As such, I decided to trim the original cab floor behind the mounting slot for the original cab. This narrows the floor behind the slots by 1mm, which allows the cab sides to drop down to the new floor. After making this modification on the Blender model, I coloured the original cab floor green to indicate I would have to modify it on the physical model.
After this, I did the cab floor (in grey). The area immediately under the original cab floor is solid, to provide support and a wide surface for the glue. There's also a 2mm hole in the middle, which will be used for a screw for further reinforcement. This is duplicated on both sides. Given that this floor section will have the coupler mounted in it, I want it to be as strong as possible.
I also added a shape (in blue) to simulate the driver, in order to ensure I would have enough space for a figure. In scale measurements, it's 180cm tall, 60cm wide and 60cm deep.
My next step was to create the saddle tank in Blender, as part of the original body mesh. Using this, I then made the cab's front wall. I made it a bit taller than the original, to clear the height of the driver figure. The insets around the cab windows are 1mm tall.
(The black mesh is where two shapes overlap. When exported for the 3D printer, this becomes the one mesh.)
I then duplicated the cab wall, moving it to where the rear of the cab will be. I then moved the base of the rear wall down to the floor, and adjusted the windows. I then created the fuel bunker at the back. This is hollow, with 1mm thick walls and a platform to put the wood load on. When assembled, this bunker will house the speaker. There are also four 1mm mounting pegs, which go into corresponding holes on the rear of the cab.
With the bunker and end walls in place, I then made the side walls. I've always liked the look of the wood paneling on the Bachmann cabs, and so I replicated this on this one. Once I'd completed one of the side walls, I mirrored it over to the other side.
I then added the roof, by copying the curve from the top of the front walls and extrapolating. This will be the fourth section of the model.
I then added the ventilation hatch (I think. I'm not quite sure what that is) to the top of the roof, and created the side steps. These are the final separate components for the model.
I added some finishing touches to the cab walls. A grille for the speaker was added at the rear. Although I will be adding glazing to the front and rear windows, I will be leaving the side windows open. I also added a hole at the top, of the same diameter as the rear headlight casting. I then added a 'lip' inside the hole, 0.5mm narrower. This will provide a mounting point for the rear headlight. Once I install the headlight, I'll run the wires down to the cab floor, then fill the rest of the hole with body putty. I also added mounting pegs for the cab steps, and small (0.2mm) holes next to the doors. These will be drill starter holes, for the wire handrails that I'll be installing.
With this done, I broke down the model for 3D printing. The four major components were separated, and I put the cab steps inside the fuel bunker. I then linked these with a small sprue, and exported the model into .stl format for printing. I then uploaded it to Shapeways.
When a model is uploaded to Shapeways, it undergoes a series of automated tests. These check if the model meets the requirements for the material selected, such as maximum size, minimum thickness, clearances, etc. If these are passed, then the model goes for review by 3D printing engineers. During this second review, a fault was picked up in my model. When I put the sprue in place to hold the cab steps in the bunker, I had made it 0.2mm wide. However, for the material I had selected (frosted ultra detail plastic), the minimum width for an unsupported load-bearing wire is 1.0mm. Once I corrected this, the model was able to be printed.
I've now got both the pilot and the cab kit on order, and they're due to be shipped next week. I expect them to arrive early or mid-January, and that's when I'll be able to continue the model.
tbdanny:
Thankyou for that very interesting insight into developing a 3D printed model.
Huzzah, I agree, Dave.
tbdanny provides a great example of what the future of this hobby looks like. Sharing a project on a global scale via the internet, developing parts, and then making them available through Shapeways and other means.
Sure, there'll be RTR, but those folks lamenting the loss of kits should consider they are becoming more of DIY project, either finding parts already available that are useful or developing new parts that then become fodder for the projects of others.
Part 3: Misadventures in 3D Printing
At the end of my previous post on this build, I mentioned that I had designed the cab and arranged for it to be 3D printed by Shapeways. About three weeks after my first post, I received the first version of the cab kit and pilot:
Unfortuantely, there were several defects with the models I received.
I had designed the new pilot to be a 'drop-in' replacement for the 2-6-0's original snowplough, and I had copied the mounting pegs and screw holes from the original Bachmann part. These were an exact fit, and I was able to screw the pilot onto the front of the pilot deck without any issues.
However, the coupler box on the front was 0.6mm too narrow. As such, I had to remove the sides of the Kadee coupler box in order to get it to fit in. When I tested it with another locomotive, I also found that it was too low. I fixed both these issues by narrowing the inside coupler box walls by 0.3mm each, and raising the coupler box. Doing this allowed me to put an extra row of bolt details underneath the coupler, as well as enlarge the bracing underneath.
I then test-fit the cab to the locomotive. The cab walls had been designed to be approximately 1mm thick, however they came out around half that, with no details on them at all. However, the front and rear walls were strong enough for me to check that they fit the locomotive correctly. The roof and floor assemblies came out properly, although there were a few issues with the floor assembly. Part of it interfered with the chassis, preventing assembly of the locomotive. It was also a bit too flexible, and the top of the mounting tabs at the front were 0.1mm below the top of the original floor, where they were meant to be level.
The walls of the rear bunker were also too thin and had warped slightly. On top of this, the mounting pegs for the steps and fuel bunker were too small.
In order to rectify these faults, I made several changes. One of the first things I did was to thicken the walls, making them 1.5mm thick. There were some spots where they needed to be 1mm thick, in order to fit into the existing slots of the locomotive, but I tried to keep these areas as small as possible. My next step was to thicken the inside walls of the fuel bunker, making them as thick as possible while still allowing room for the speaker. I also enlarged the mounting pegs for both the fuel bunker and the cab steps, so that they were at least 1mm wide. Following this, I added some bracing along the side of the floor in order to improve the rigidity. I also raised the top of the mounting tabs by 0.1mm, in order to bring them level with the original locomotive's floor. The bits that interfered with the chassis were also removed.
I had noticed that two successful parts, the floor and the roof, were parts that were printed lying flat, instead of standing vertically like the walls and fuel bunker had been. As such, I decided to do the kit as a 'flat pack', to see if that would give better results. So I separated the end and side walls. I created mounting pegs on the end walls, with corresponding holes in the side walls, in order to aid assembly. Once this was done, I exported the kit and uploaded it to Shapeways.
A few weeks later, my next package from Shapeways arrived. As well as the updated pilot and cab kit, it also included a diamond stack in 'White, Strong & Flexible' plastic. This was also my own design, and is designed to be a direct replacement for the original Mogul stack, with the base contoured to fit the mogul's smokebox. I wasn't happy with the casting I'd purchased for the stack.
When the parts arrive, they have a bit of a waxy coating left on them from the 3D printing process. I cleaned this off by dipping them in water that had some dishwashing detergent in it, then blotting them on a paper towel. The above photo shows the two side wall sections, with the one on the right having just been cleaned.
Occasionally, large flat parts don't come out completely flat, and this was the case with the floor. It was bowed down slightly in the middle, so I used some brass bar to flatten it. However, the cross-bracing I had added across the back prevented me from putting the brass bar for the full length of the floor section.
While that was drying, I test-fit the cab walls together. Fortunately, they had all turned out well this time, however there was a slight bow inwards on the two side walls.
Both the floor and roof of the kit have ridges running along the edges, to hold the side walls straight. With the roof in place, the walls were pushed straight. There was enough tension to hold the roof in place without glue, allowing it to be removed for the addition of a driver figure later on.
I also test fit a coupler box into the pilot. It required a slight bit of filing, but it was a much better fit than the previous version.
I then assembled the locomotive, and sprayed it with primer. After the primer had dried, I went over the model and noticed a few gaps in the cab, including one at the front where it should have met the floor. It was at this point that I realised I'd made another mistake.
I had raised the top of the mounting tabs on the kit by 0.1mm, in order to bring them in line with the floor. However, the lower floor also needed to be raised by this amount. But I had overlooked this. As such, it created a gap at the front of the cab.
This morning, I dismantled the cab, returning the locomotive to the state it was in before I'd received the second lot of parts. I have updated the kit with the 0.1mm raise on the lower cab floor. I also removed the cross bracing on the rear of the floor, to allow for brass bar to run the full length if needed.
I also updated the model for the diamond stack. The contours did match the boiler, but the original spark arrestor was too small.
With this done, I painted and mounted the pilot on the locomotive. I then checked the coupler height. Given that this pilot was copied onto the rear of the cab floor, this confirmed that the coupler height was correct for both ends of the engine. With this done, I ordered the updated smokestack and cab kit from Shapeways.
My next post will cover the details of how I assembled the cab, as well as the final assembly, detailing and painting. Until the kit and smokestack arrive, however, I'll be working on the scenery for my layout.
The pilot looks great!
Interesting trials and tribulations! Thank you very much for taking the time to explain all the little glitches. I'm sure you must be a bit frustrated by them but seeing you work through them is very educational. Your costs must be mounting a bit though with having to do parts over again.
Thanks for sharing!
Dave,
The costs aren't too bad. Shapeways actually refunded the first model, as it didn't turn out properly despite being within the tolerances for the material selected (Frosted Ultra Detail). Also, due to the turnaround time (~3 weeks), the costs are spread out.
On top of this, Shapeways actually allows its users to sell their models to others, and I'm planning to do so with these. I've already made the pilot available, and I'll be doing the same for the cab kit and smokestack once I've confirmed they're correct.
After a few months working on the build, on-and-off, I've completed BVLC No. 8. It wasn't as straightforward as I thought it would be.
At the end of part 3, I had just realised that a minor correction needed to be made on the cab walls.
Once I'd received the new cab parts, I installed the cab floor as per the last time. I added some brass bar to reinforce it.
Like last time, I formed grab handles out of bent wire. However, this time around I ran into a bit of a complication. For some reason, the corners of both side walls broke off when I went to drill through them. In order to repair this, I used some styrene and body putty to fill in the gaps.
With the grab irons in place and repairs done, I then assembled the cab. Before attaching the rear wall, I added the rear headlight, which had been recovered from the previous cab.
I then attached the cab to the body, and ran the headlight wires into the cab, under the floor, then up next to the firebox. The idea was to have them out of sight under the cab, then disguised as boiler piping when entering the body shell. I then painted and lettered the model.
While that was drying, I prepared the driver figure. I didn't have a driver figure to hand, but I did have a guard figure (left). After removing the watch and repainting (right), 'Stan' was ready to act as the driver. I picked that name as the figure's face reminds me of Stan Lee, the comic book creator.
After the body had dried, I installed the speaker. I'd included a grille for it in the rear wall, as well as a raised outline to hold it in place. The wires were run down the opposite side of the cab to the headlight wires.
With that done, I installed a TCS WOW decoder and put the body on. As you can probably guess from the fact that this is only about halfway through this particular post, this was where the problems started. I tested the loco on my programming track, and got no response. In a moment of foolishness, I turned on the track power of the SPROG2. This resulted in my model steam engine emitting actual smoke.
I determined that some of the wires had shorted out, most likely against the shell of the motor. To avoid this, I built a small shelf to isolate the replacement decoder.
Once the replacement decoder arrived, I installed it, being very careful with the wires while doing so. Afterwards, I tried it on the programming track. The decoder responded and so I tried it on the layout. What I found was that the locomotive would speed up and slow down randomly, the sound cutting in and out while doing so. It appeared to be something to do with BEMF, but none of the configuration or troubleshooting I tried worked.
TCS couldn't find anything wrong with the decoder, when they examined it under their warranty. As such, I figured it was something with the particular combination of motor and decoder.
I had some Digitrax DZ123 decoders on hand, which I'd purchased to convert some of my old HO scale models for running at my club. These were able to take the motor's stall current, and so I installed one in the chassis. After test-running it, I found that it ran satisfactorily. However, when I moved into On30, I decided that all of my locomotives would have sound.
Given that the model was working with the DZ123 decoder, I didn't want to replace it. As such, I decided to use a Digitrax SFX006 decoder for sound. I installed it in the shell, instead of on the chassis. Before doing this, I programmed the decoder lock feature of the DZ123 so that I could configure it separately.
Unfortunately, when the shell was put on the body, the wires rubbed against the flywheel and prevented the motor from working properly. To prevent this, I used some styrene to build a 'shield' over the flywheel. I also relocated the DZ123 a bit further back, in order to reduce the amount of slack in the wires.
With that done, I programmed the sound and motor decoders as needed. For some reason, I was unable to get the decoder lock to work correctly on the SFX006. As such, I installed a small jumper between the air tank and the boiler on the right hand side of the engine. This allows the SFX006 to be disconnected when the DZ123 needs to be programmed.
Once I had the loco programmed, I just had to add the finishing touches. Although I had 3D printed some steps for the cab sides, they turned out to be a bit more fragile than anticipated. I ended up using the mounting holes for them as drill starter holes, to mount ladders I fabricated from 1mm copper wire.
After painting the steps, I added the wood load, whistle and touched up the paint where needed. With that done, I added the completed engine to my roster. It took me almost six months, from start to finish, but now I've got a saddle-tank engine of my own.
Outstadning work !! ... ..... Thank you for sharing.
GARRY
HEARTLAND DIVISION, CB&Q RR
EVERYWHERE LOST; WE HUSTLE OUR CABOOSE FOR YOU
tybdanny,
Makes for a husky looking logging Prairie, a real classic of the woods.