Lowa

Thank you! To everyone who's following what we're doing: this is just the beginning; we want to automate every aspect of LEGO train layouts. For example, we just finalized a design for a PF track compatible decoupler: it allows you to remotely decouple railcars of LEGO trains; it's a PF version of 7862 for the old 12V system. If you like what we're doing help us spread the word by sharing posts... or place and order and share your experience

We're working on a Raspberry Pi based layout controller that can run nControl and manage the system; and then use a tablet/phone to interface with that controller. That will allow you to take the traditional computer out of the loop. What do you think? About the double crossovers, we're working on a 'modular switch system' that will allow you to make a variety of different switches with a limited number of track pieces; double crossovers being one of them. The main difference with the 7996 will be that ours has 4 fully independent switches, so you can run the trains parallel at the same time... We're making good progress and if all goes well we'll I should be able show the first 'final' prototypes beginning next week. Thanks! What is still missing for you ? Indeed, we have customers that are using our track switch motors like that! Yes, that's what we're working on!

Great, let us know how it goes! We just uploaded a YouTube video to show train automation with a tablet: we control the switch motors, traffic lights, train head lights and the train from nControl. As this is broader than 'track switch motors' I posted it in a new thread:

We have been working on expanding our train automation system. The video below provides a glimpse of the (near) future. In this video we control everything from a tablet: switches, traffic lights, train head lights and the train itself... The layout automation system is connected to a PC that runs nControl, the tablet interfaces with the PC using a remote desktop app like TeamViewer. The train uses a prototype of our WiFi based train controller. The train communicates with nControl using nControl's internal MQTT server. The track switch motors and controllers are available; the traffic lights should become available the first weeks of April. The train controller is in development: the goal is to be able to control the direction, speed, head lights, check the battery power, and use a position sensor so nControl can track the position of the train on the layout. Let us know what you think!

The sensors are working with our controllers but I still need to see how to do it for 'generic controllers'. We'll certainly add a 3 aspect signal, I have been looking into making the actual lights and it seems to be possible to get everything I need into the housing. We also ordered 4 wire cables to make those lights. 'Dark night running mode' That's a nice idea! I don't see much issues for the track layout, all the graphics are programmed so it's just a matter of changing the pen color. The tile layout is going to be more complex as the graphics are bitmap. But we have vector drawings of all the bitmaps so it's feasible. I like the idea; we could also add a light sensitive sensor so the switch happens automatically... and that could also switch one other illumination in the LEGO setup.

You're a genius! I gave it a try with remote desktop and it works great both with phones and tablets. What are you using ? I ran a couple of tests on Android and found that Google remote desktop works best for my phone and TeamViewer was the best option for my tablet. TeamViewer gives me a perfect image on my tablet. On my phone the Google solution works better because TeamViewer only support 'native touch screen controls' for tablets, not for phones. Also the responsiveness is great, it didn't see any latency. This is a great solution, thanks for sharing this! Yes, we have preparing the production of the lights. We're waiting for the cables, we cannot 3D print those... They are currently being made by a supplier, once we have them we can start making the lights! It's pretty cool to control everything with a tablet. Based on the MQTT suggestions of @skaako we managed to develop a prototype of a WiFi based LEGO train controller. So now we can do the switches, traffic lights and the train (and the head lights of the train) with nControl from a tablet! It's pretty neat, we'll make a WIP video to show it...

Ok, great!

Hi Matt, Assigning a custom color to each switch should be possible and could help to identify the switches/tiles. But we released the version 2017.2 yesterday (Windows and Mac) and it has all the features you suggested in your previous post, so now you can flip the switches by clicking/tapping on them in the train layout. Can you give that a try and let me know if with this there is still a need for color coding the switches ? We don't have any track switch motors stock in Europe at this points. The goal is to send some at the end of the month. So depending on your time frame you can get them right now from the US of wait until (half ?) March to order them from our European store. Hi Q3771. Version 2017.2 has a 'Synchronize' function; it loops over all the tile and synchronizes the actuators to the state of the tiles. Thanks for suggesting this feature; it's pretty cool to see it in action... it shows a progress bar and a preview of the tile that is currently being synchronized,so you see what it is doing. I think this is a very handy feature when you make a temporary setup... @McWaffel, @Q3671 The shipment cleared customs today, so 4DBrix Europe should have your motors by the end of the week... Thanks for sharing this. You even managed to connect to the blinking functions of the lights! Super cool!

Sorry for the late reply... the port of nControl to MacOS consumed all my time lately... Concerning your questions: So for every 4 switches i need 1 USB Port? => Indeed, one controller can handle up to 4 motors, but you can connect multiple controllers to one computer. The system is designed to be able to handle multiple controllers simultaneously. The maximum number of controllers is limited by the number of USB devices a computer can handle: 127. Also cabling length might be a problem for usb? How long can the cable from controller to switch be? => We have tested it with a 3m/10ft extension cable between the motor and controller and everything worked fine. This is most likely not the limit, we didn't test it with a longer cable. The motors have 24AWG (measure of wire thickness) while we only found 26AWG (=thinner) extension cables; that what we used for our 3m/10ft test. We custom ordered 24AWG extension cables, the are currently making them. Once we have our 24AWG cables we'll test to see how far we can really go... Maybe any plan in the future to make a Bus-System and only have 1 central interface? => That how it is when you use a USB hub. You connect all the controller to the USB hub and then you can use nControl as you central interface to control everything. You can also combine different type of controllers: motors, lights, sensors, etc. Would then also be nice instead of a USB Interface to something to run this on an iPad/Tablet App instead. => An app for mobile devices would indeed be nice, but we're not there yet... porting to mobile will have consequences on both the software and hardware level. We stared with computer applications because we wanted to go beyond simple 'remote control'; with the sensor you need to program how you want to respond to a train detection; and it seemed more convenient to do that on a computer than with a tablet. But the ultimate goal is to indeed have nControl run on mobile devices as well. I hope you make it available in europe soon. => Yes, we're working on that!

The track switch automation in action on a layout...

FYI: I received the 1.5mm thick orings (9mm outer diameter). I prefer them over the 1mm thick orings because they look sturdier. So that's what I would go with for the medium sized wheels. For the small wheels, the 1.5mm thick rings look a bit too bulky... so I ordered some 1mm thick ones to give it a try on the small wheels...

Indeed, adding the tongue and groove connection to the tiles is an interesting idea...

Hi Matt, No, I don't have experience with them in general. What's the goal of using them ?

Ok, noted. Thanks for letting us know your thought! I see your point, but the two sort of go together: you need to choose a radius for your switch. My initial idea was to start with one for R24 (existing LEGO curve) but according to some users R24 doesn't work well for narrow gauge trains, but others like @M_slug357 seem fine with it. I don't have any experience with narrow gauge trains, so my question: does it make sense to make an R24 narrow gauge switch or not ?

I'm glad you like nControl and thank you for the complements! I'm getting very close; I'm currently finalizing the design of the sensor and sensor controller. The software is ready to use the sensors, it just needs testing and debugging. I do still need to add a hook so it can process sensor feedback from your DIY board, but that shouldn't be too complex. I guess that should be possible by adding custom Python scripts to the tiles. If you hit a bottleneck, we can see what features we need to add, because the goal is to have a framework that is flexible enough to do things like this. I'm very interested in the set-up you're trying to build: have you built anything or is it still in a conceptual phase ? To get started, I would give you free nControl trial license for DIY boards for a couple of months so you can figure out how to get it to work with your system...

@mattmarsden I have been working on nControl and in the next release the state of the switches will be stored in the project file and the tiles will be initialized with these values when you open the project. You can actually get the tiles out-of-sync with the actual switches by going into simulation mode; that changes the tiles but not the actual switches. Therefore I also added a 'synchronization' function, it sweeps all the tiles and sets the switches in the layout in the state of the tile. @Q371, I guess that's (close to) what you meant with a 'reset' button, or not ?

I'm glad to hear you're happy with the motors! About your suggestions for nControl: Being able to click on the switch on the track layout to operate it rather than just the tile: That's indeed something I have to look into as finding the right tile becomes increasing challenging with the number of switch motors. Remembering the state of the switch after closing and reopening the program otherwise what is shown on the track layout is wrong unless you operate all the switches: I'm aware of that issue. At the moment the project file doesn't contain the state of the switches, we're going to add this so that nControl opens with the same state as you closed it. That's to only option I see, retrieving the position from the motors is not possible as the motors don't allow bidirectional communication, they cannot tell you the position they are in. Automatically connect to the controllers on opening the program: I hadn't thought of that but I can see this would be very convenient once your system is set-up. Most probably it makes sense to add this as a project settings. I guess it would also be handy to have a 'default project' that opens automatically when you open nControl ? Remembering the zoom and window sizes as I need to adjust everytime I open nControl: Yes, that makes sense. That shouldn't be a big deal. Thank you for the feedback! These are all very useful remarks and will improve the software. I need to see when I can incorporate that. I'm currently working on the port to Mac. So I need to be careful with making modifications to the software as I'm working on two systems.

Personally, I like the 'one part' style tracks because you can give the rails a less 'bulky' shape as you don't need space to fit a stud inside the rail. But it's a bigger effort from both the designing and printing side of things. Having a 'rail' system as you suggest has indeed a number of advantages. A 'dual gauge' system built on the exiting R40 curves would require a rail-based design anyway. I gave it a try and found out that it's a fairly straightforward modeling job. I did try to make it as 'elegant' as possible, e.g. keeping the cross section of the rail as close as possible to the cross section of the PF rails. I ended up making 3 prototypes: 2 rails to make a R30 narrow gauge track and one to make a dual gauge using an R40 curve. The conclusion is that it would be feasible to make a wide range of rails with different radii. I thought about it and I don't see an issue in having both systems: teh 'one part' style for a limited number of radii: a set of radii selected by the community the 'rail' style for a wide range of radii: any radius you need @pirzyk In the horizontal plane, I can easily customize the clutch power to get it right. These rails have 'standard' clutch power for the plates in the middle but an increased clutch power for the plates at the ends. It makes it slightly tougher to add/remove the plates at the ends but once they are on it they won't come lose! It think the increased clutch power is absolutely needed because you connect the rail with just one stud to that plate. By increasing the clutch power at the ends I can now put the end of the rail one stud of the plate and shake the rail vigorously, it doesn't come loose, it doesn't even move. So there's no need for kragle ... R30 narrow gauge rails on 2x6 LEGO plates dual gauge rail on a R40 curve A poll would be perfect! But besides the 'system' (R24-36-48- or R30-42-54) if would also be interesting to figure out within those systems, which radii are the most popular ones.

@jrathfon I'm a bit confused, a R30, R42, R66, etc radius gives 6 studs between the tracks not 8 studs like with the standard tracks. I thought we wanted an 8 stud gap between parallel tracks, or did I get that wrong?

@McWaffel, @Q3671 We're preparing a shipment with monorail track that we'll send to our European distributor this week. If you're still interested in the track switch motors and/or controller, we can add them to that shipment so you can get them from the European 4DBrix store. In Euro, the motors are €43.95 and motors €16.45.

That would be possible, but I don't really see the advantage of this one over the 4 stud wide one... That's indeed an intriguing idea! I did a quick test and put my tile on a train track and it works... trains can pass over it without any problems. Leaving the track is indeed the main issue. The only solution I see would be a 'monorail style' switch = a switch with a sliding piece of track. Or if the train track is raised, the monorail could dive and leave the loop...

Thank you for the feedback, this is very useful to know ! I'll do the test with a cross just behind a curve. I already did a quick verification of our cross with a set of LEGO train wheel and there's not enough play to hit the crossing rail, so I guess we're good. Maybe that was a 'design flaw' in the 9V track. I did notice that that gaps between the crossing tracks in the PF switches are 1mm wider than the gaps in the 9V switches, that is clearly an intended modification of the switch. Maybe the gap in the 9V cross was slightly too tight... I cannot verify that because I don't have a 9V cross track. But we'll check it on our cross and let you know... Thank you, I could not have explained it better...

I haven't worked on the small wheel but I did look at the medium wheels and bought some 1mm thick rubber o-ring. They work fine but I'm wondering if 1mm is not too thin, so I ordered some 1.5mm thick ones to compare. I should have them by the end of next week. I guess the small wheels need o-rings too, even if you don't power them. Without rings the wheels tend to slip over the tracks; they are not spinning. The LEGO tracks are very smooth and I guess the axle just gives a little bit too much friction. I'm waiting for the 1.5mm thick o-rings to arrive and than I'll decide whether I should get 1mm or 1.5mm o-rings for the small wheels. If you insist I can print you some, but I think it would be better to wait another two weeks until I have the o-rings for the small wheels.

We are working on switches. I'm not sure what you mean with 'short' switches. We're planning to release a switch with the footprint shown below. It could be used in three different ways: with a 'turn back' adapter you can use it for a parallel track (like a PF switch) with a 'curve' adapter you can use it in a continuous curve or connect two switches and make a single cross-over Is this what you're talking about ?

Yes, that should work, but you need to center the rail in between 2 studs so you would need to use a 1x2 with a central stud. But unlike with the normal trains the monorail train does not ride on the rail itself so you need to add plates around it. Those plates would not touch the rail but leave a 1/2 stud gap. I'm not sure if that would look good... To close that gap we would have to put the rail on a 2 stud wide tile, but then you're getting very close to the original 4 stud wide monorail tile design.