Welcome to my new "Dave's Locomotive Controller"! (or DLC ) - Apologies for the misleading title, but I thought the pun was worth it! Without further ado, here it is: Over the last few weeks, I've been working on a few prototypes, leading up to my own custom controller, making full use of the the excellent "TSW World Raildriver and Joystick Interface" by CobraOne. It's almost complete, with just the handles needing to be tweaked, printed with metallic coloured filament (which is on order) and then several 'detent plates' creating for the different trains I own. I've tested it with trains I own and I'm very happy with the end results! To give you an idea of scale, it has the same footprint on my desk as my keyboard. There are a number of additional features over the closest rival, including an analogue joystick for looking and walking around in the world, a proper key and E-stop, controls which use linear rods and ball bearings (smoooooth ) and the big thing; DETENT PLATES! whilst developing the prototype, it was obvious that single detent track wasn't going to cut it, as there was too much variance between the different engines and rolling stock. So, I made the detent tracks removeable and replaceable with a few thumbscrews. The height of the detent track can be tuned to give different friction and/or detent sizes and can also limit the throw of each slider (which can be up-to 100mm). For example, if you want a short-throw on the reverser on the left hand slider, then you can print a detent track to limit the overall travel to say 40mm, with deep detents every 10mm. With careful programming in the "TSW World Raildriver and Joystick Interface", this could be made to match the reverser on an electric loco. Equally, there could be no detents at all with the full throw of 100mm for a Steam regulator, with gradually increasing force needed to move from one end to the other. Here's an example detent block that has end stops and a single detent in the centre: I'm happy to share my design with the community, but I was wanting to gauge interest; If there is lots of interest then I'm happy to spend a bit of time putting together a comprehensive pack and instructions. If there are only a few people, then I'd liaise with them directly. If I've inspired you to design your own, and want advice , then I'm equally happy to answer questions. The microcontroller I used is a cheap arduino pro-micro knock-off from China. It's a USB C variant. A future upgrade I might do is move to an STM32 board, li-po battery and connect via bluetooth rather than wired. As a starter, here's a quick list of major tools you'll need to manufacture my design; 3D FDM printer with a volume of at least 200mm x 205mm x 85mm Bambu Studio (if you want to us my 3D printer files - if it's compatible with your printer) Laser Engraver & Cutting machine with a working area of 400mm x 190mm OR A CNC router with a similar working area* Soldering Iron with normal soldering tips and tips for installing metal inserts in plastic Hot Glue Gun (or 2 part epoxy) Angle grinder and/or Bench Grinder (to cut the linear rods to length) An Amazon account for the COTS parts! I made my faceplate using this traffolyte, meaning I could cut and engrave the entire faceplate in a single go in about 10 minutes. I'm guessing not many people will own or have access to a laser engraver and cutters. The alternatives is to use a CNC router (like this one), which would involve moving the material several times (and requires a bit of experience or trial and error to line everything up). If you don't have access to either of these machines, 3D printing in several sections is a possibility or even cutting and drilling the faceplate by hand using hand tools (i.e. drills, jigsaw and Dremel) and using a label maker to make the decals! One thing about using traffolyte is that once you've engraved away the black surface exposing the white plastic, you can use Green LEDS or similar to back-light the panel, providing the white plastic has some translucency (mine does - so a future addition for me). I've not totted up the cost of the COTS parts, but I'd guess it was about £100 (although I had many of the consumables already, like the hex bolts, threaded inserts, wire etc.) but I can put together a full inventory and cost if there is interest. All the parts were printed over a working week and assembly took about 10 hours. I've added a vote to let you indicate if you'd like full plans or just my design files and your own initiative! Cheers, Dave.
Hi Dave It might be worth you mentioning this forum in mine for others to see, I don't have a problem with that and it might get you more responses. Chris
Thanks Chris, I'll head on over there shortly.... It seemed so obvious, once I'd made a prototype and figured out the limitations of a single detent track! The metallic filament I've had on order for the handles has been delayed, so I've run a couple off using plain black for now, and I've made my first proper detent plate for the 377/4: I'm not sure why there is a slightly larger gap between the P1 and P2 positions than the spacing between P2 through P4, but could be something to do with messing up the calibration in TSW4? The plates take about 30 minutes to print, and I can easily tweak this later, but I'm heading off on holiday on Wednesday for 10 nights and wanted to get something made and working to take with me. I'll get chance to to write up some stuff over my holiday (which is in North Wales and traditional Welsh weather has been forecast!) as there seems to be reasonable interest in something more detailed than just some step files and dxf's. With hindsight, and your new 'shift' button, I could have doubled up on the functions of the buttons. There is space for one last button on the internal button matrix, so I might get a drill out and put one last round button next to 'pause' for this. Dave.
I don’t understand the mechanical terminology but I’m impressed by your commitment to build a first-class rail controller. As someone who mostly drives German trains, is it possible for the throttle to be converted to a combination throttle/electric brake?
Epic piece of kit! One thing though, the lever that you have labeled 'brake' would be more suitable for 'horn' on British trains.
Hi Jon, That's exactly how I have it setup in the picture in my post above yours. The left hand lever in the picture controls the reverser and the right hand is a combined throttle and brake for the British class 377/4 trains. P1-P4 are the power settings and B1-B3 (and the emergency brake position) is braking, with 'OFF' as the neutral point between them. Each setting has a detent to make it easy to find the position. For true linear controls, you don't have to use detents at all! I'm going to have to politely decline to supply any parts to anyone, I simply don't have the time to make them (I work full time) or the inclination to sell something and deal with VAT and personal tax from profit (I am not a registered company and sole-trader would be a nightmare!) I am happy to release design files for use by private individuals for their own use, which I'll probably do under a creative commons license. There are paid services which can produce 3D printed parts and ship to you and I'm sure there will be local suppliers who can cut and etch traffolye for you for a price (many sign makers can probably do this) if you supply the necessary files/data. All the parts I have used are obtainable through company's such as Amazon and Aliexpress. You could do exactly that! Cobraone's software (this thread: LINK) allows for different lever and button maps, depending on the locomotive you are using, it's worth reading his manual and becoming familiar with the features of his software. I used this style of brake, as I use the Class 66 diesels to do freight work alot, and this style brake mimics the real class 66's. You could use this 'brake' for the horn for other loco's if you wanted. One of the things I considered doing was to use small 0.91" OLED displays (like these ones; LINK) in place of decals, so that controls could be re-labelled on-the-fly. I don't have the time to implement this at the moment, but might be something I do on a future iteration of my design. It would also significantly increase the cost; I've got a box of 100 of these displays that I bought in bulk from China in my garage for pence each but, as they communicate via I2C and they only have a couple of selectable addresses, you'd need a significant number of I2C muxers and the code would get pretty bulky too! It's at this point I'd consider designing a PCB and buying the chips direct, rather than breakout boards, but this is a significant investment in both time and money. As part of my write-up, I'll include a section at the end with potential future developments.
Shame really, I understand you being busy with work and not being able to build these to sell, it must be a long job to build one of these then as you say go through the selling process, packing it up and posting etc. There is a possibility some company would make some and pay you the royalties for every sale, no idea how all that works but it would be nice and I'm sure lots of us would be interested in buying. Col.
This is amazing and I would love to take a look at the design files. I have a Zuiki mascon that I got for my switch and densha de go and while it has plenty of applications that work great in TSW, it has some funky operating limitations that I don't like (basically entirely borne of the fact that it seems to use software-defined axis outputs instead of a linear one, and that CobraOne's software is truly excellent but would need considerable extension to provide mapping this way). Few questions: * Was there a specific reason you went with sliders versus rotary controls? I'm guessing that the design work for rotary controls is somewhat higher, especially for custom detents. * Did you spring-load the detents? This has been a mystery I've thought about a bunch of times, how to get a detent to actually...detent.
Also, a variant for German safety systems would be neat too. Having the three controls (acknowledge and release, single-pole-double-throw momentary switches; override is a simple button) at hand would, at the risk of stating the obvious, greatly enhance the immersion experience (and easily extend to the more basic UK/US control schemes).
Hi Peeps, Not getting as much done as I'd like on the write-up during my holiday, the weather has been much nicer, so I've been doing lots of day trips out! I have been making progress though, I expect to have something put out this time next week. Yes, I use a simple compression spring. The strength (or pre-compression) of the spring affects how strong the detent is. I'll explain this fully in my write-up. It was in part an aesthetic choice (a lot of the trains I drive use linear motion) but also relates to the space envelope. To get a similar refinement of control using a rotary, the handle would have protruded from the surface of the faceplate more; I travel quite a lot and wanted to keep the overall size as compact as I could, without sacrificing too much functionality. You could use the same method of detents on a rotary type mechanism, and is in fact what I did on my first prototype [link]. Whilst the axis of rotation is different, the principle would remain. I'm afraid I'm not familiar with the Germain safety systems, but I doubt it would be difficult to implement. I'm not going to have time to do anything other than publish notes on my current design at the moment. I'd encourage you to think of how to modify my design once I've published the notes, and I'd be happy to help you do this! Dave.
Hi All, I've just released the first version of the design guide and files in a new thread: LINK. Please head over there and grab the files! Cheers, Dave.
dave#2392 Absolutely fantastic what you're doing. But I maybe would consider to rename the thread title. Not only because it could confuse people, but the title doesn’t do it justice being way too generic and inprecise.
One of my cousins called yesterday and he noticed my RailDriver, I expalined and shown him what it was, after a brief discussion he to too said he was surprised considering the amout of ppl using the sim that another comapany hasn't made a similar train driver, sure there is a market for one. No offence to what you have done here, I wish I had your skills, my point was even my cousin who has little knowledge of this said straight away about other companies making one.
