Carl

@martin_wynne, I thought you would like to see this : A 3D printed track base based on your design. You can read about the gory details here : Incidentally, I know you are working on a 3D export for Templot. But I wonder if there are plans to be able to export a 2D .dxf with holes on the rail/sleeper centre intersections? Regards, Carl.

The big question on everyone's lips these days has nothing to do with pandemics, Meghan, or "how is that boat floating in mid-air?"[1]. It's "can you 3D print N gauge track bases?" The answer seems to be, yes! The total cost for the resin for this was about 50p (excluding some wastage, cleaning materials and of course, the time involved.) The process for that design was to export the part of my track that I wanted to print from Templot as a .dxf. First, I set the Generator settings in Templot to show the centre line of the rails. Then, I imported the templates as .dxf into Autodesk Fusion 360. The sleepers were extruded by 0.5mm, I added a 0.5mm base, then for each interception of rail centre and sleeper centre (except check rails, of course), I added a 0.95mm hole. This matches British Finescale track. This took a while, there is, like, two trillion of them. I then exported that as a .stl. I split the body into two parts at this point, since the print bed on my printer is only 115mm wide - the track base is about 195mm wide. The slicing software I used was Chitubox. I have one of the original Anycubic Photon resin printers. The settings I used were 3 base layers, 60s exposure for the base layer. Then 8s for all other layers. I angled the base up at 8 degrees so the base doesn't get stuck to the print bed (see the outtake!) and an 8mm platform was added. I probably didn't need 8mm though, certainly not after angling the base away from the bed. I printed with a 0.05mm layer height. The 8 degree slope is what accounts for the layer lines you can see on the prints. About 2 hours later, boom - half of my track base. I repeated the process for the other half, and the picture above is the result. Now to drill out the chair holes, give it a dusting of grey primer and a careful sanding. That should hide the layer lines. Then to add the two trillion chairs to the two trillion holes, then work out a way to do the various K and V crossings. Some 3D printed jigs are going to be needed I think. Here's a few more photos : And the promised outtake - printing completely horizontally meant that the base adhered more strongly to the resin vat than to the support. [1] https://www.bbc.co.uk/news/uk-england-cornwall-56286719

You’re more than welcome, thank you!

Martin, I see you have PMs disabled - thought you would like to know, after donating, the page Paypal sends you back to 404s - http://85a.uk/templot/donation_thanks.htm (I've removed the query string.)

You're a superstar, that does look a lot better, thank you!

Hi guys, I'm about to start building this bit of track, and I'm trying to get the sleepers roughly correct for it. The era is 1990-2000 BR, it's part of Heaton Norris Junction that I'm modelling where a branch from the Down Slow crosses the Up Fast and Up Slow. Obviously the junction is somewhat compressed. In reality, the diamonds are probably 1:8 and relatively standard. I'm trying to reach a balance between what the crossing looks like in reality, and what it would look like if the diamonds were this obtuse - I don't want hyper-strengthened diamonds like you would see at the likes of Newark. The track is flat bottom, and the sleepering on the rest of my track is concrete. I intend to build this with concrete-coloured sleepers, if only because the chairs I have for FB rail are concrete grey - after weathering. It's N gauge, using code 40 rail, and British Finescale chairs. With that in mind (if any of that makes any difference!), is the sleepering here reasonable? I'm pretty happy with it, but the things that really jump out are the gaps where the sleepers return to regular track - there's a difference of 10 degrees or so on a couple of the exits, and that really stands out to me. Any advice would be welcome. Regards, Carl.

Oh, I do like that.

Thanks. I wonder if I can summon the courage to actually model it right in the middle of the mainline! My compromise is probably going to be modelling a couple of completely disconnected points in a mode out of the way area, with the branch still in situ - but that is because of a lack of space and needing A6 or B7 points for them, and I think they will look wrong alongside C10 and E15 on the rest of the layout. Having them disconnected helps hide the shortness of them.

I was watching some YouTube vids this evening and have just come across this - a partially removed point in precisely the junction I am modelling - Heaton Norris Junction. Much as Phil mentioned, this was a temporary repair, and it’s on the WCML - or I am 99% sure it was temporary, a I’ve not seen any other pictures of it during my research. The video was from a couple of years ago.

If they did comment on it, it would only be to say it didn’t photograph well, it looks right in person, and is to the measurements on the official guide.

I tried the Worsley Works body a decade or so ago, but I couldn't find a suitable chassis at the time (I think I used a very bodged Tomytec chassis which started out with bogies.) I wonder if the Dapol chassis could be made to fit the Worsley body? I probably still have it somewhere. Still means Dapol get my money though!

I can live with the windows, they look close enough to my eyes, but those light clusters - just so wrong. But wasn't expecting that to change from the first run, and will still pick a couple of each of the GMPTE and Provincials up.

I love this layout, thanks for posting all the photos. I am planning a terrace of the Scalescenes houses, and having seen yours it’s now a definite. And I fancy watching Friends again! I look forward to seeing more photos and details.

It's taken a while, but I have a progress report. The north side of the fiddle yard is mainly laid, all the points are motorised and wired up, and have DCC busses running around all of the baseboards, ready for me to start laying track on the scenic sections. It started out being a bit of a nightmare for me. It was mostly caused by inadequate planning, combined with lack of experience - and skill! - but after the steep learning curve, things went more smoothly. Again, the lesson learnt was plan, plan, plan. Some areas of my fiddle yard are quite dense with point work. I had a number of Cobalt point motors in stock from a few years back, and was already planning on using servos for any extras. What I didn't plan was which motors I'd use in each location, but quickly found that I didn't have much choice in most cases. Also, by this point I'd soldered dropper wires on both rails at the toe of each point, and a dropper wire for the frog, and drilled holes for them. That's when I realised they'd sometimes clash where where the point motors had to go. After I'd re-soldered some dropper wires to be in more appropriate locations, the next issue was that I had about a million dropper wires, and had to continuously re-check I had the correct ones for each section of track. I wish I'd drawn a track diagram on the underside of the boards before I started, that would have saved me some time. After I did this, things went a lot smoother. On baseboard 1 (which is 4' x 2'), I have 6 Cobalts and 11 servos. The Cobalts are the "classic" versions, and are connected to a AD8-fx decoder. The 11 servos use Dingo Servo Micro 10 mounts, and are connected to Digikeijs DR4024 decoders. There are three more points to be laid, but I'll come back to them - they lead to holding sidings, and aren't needed at this early stage. Each section of track, points included, has occupancy detection via DR4018s, there are 47 sections on that board alone - at the start of this project, I thought 48 would be more than adequate for the entire layout! Here's a couple of photos. Two wires they said! And the top side - which seems depressingly little compared to the time it took to lay it! : The next step is to link the remaining tracks to this board, giving me 11 roads of storage.

I attached my oscilloscope to the ports, and nothing unfortunately! Thanks for the idea though, that would have done me for a few days.

That's really not a bad idea, if I can find an appropriate, cost-effective device. I can't immediately find one, do you have any suggestions? Digikeijs suggest that the DR5000 it is probably going to cost as much to fix as a new one is, since it's out of warranty, but will check for me if I send it to them. If anyone does find this and wonder about Digikeijs' out-of-warranty repairs work, the first 10 minutes is free, then they charge €12.50 per 15 minutes to a maximum of €50, plus parts and shipping. In this case they suspect it will go back to the processor. I'll send it them though, worth the shipping to find out. I found one in stock over the weekend, and have ordered a replacement anyway.

Thanks, at least my diagnosis seems correct. I don’t have a way of checking the DR4088LN for that section, but at least they are cheap compared to a DR5000! I found someone with one in stock, so that’s on the way. Hopefully Digikeijs can repair the DR5000 I have, if only to have a spare like you!

I'm in the middle of laying my fiddle yard, and have been testing each point I lay once I've installed the occupancy detection and point motor for it. My setup is : - DR5000, the "track" output is the feed for my accessory bus, for the whole layout. - DR5033, connected to the DR5000 with LocoNet, and this (and two others when I get that far) provide track power in three districts. This morning, whilst laying some track, I left the track power on after testing, and when I went to solder on of my bus wires either the DR5000 or DR5033 tripped (as you'd expect.) I'm not 100% sure if it was the track or accessory bus now, but the upshot is that the next time I went to test, I had no track power. After some diagnosis, it turns out that neither of the LocoNet ports on the DR5000 are outputting anything - so the fault is with the DR5000. The DR5000's track output is still working, and I if I connect that to my DR5033's track sniffer input, it is working too. So I can continue laying track and testing as I go. I'll order a new DR5000, my luck is that everywhere seems to be out of stock. I have also contacted Digikeijs about this. So it seems like it's just LocoNet which is not functioning. What I'm wondering is if there is some clever workaround for my DR4088LNs (which I use for occupancy detection) now not being able to communicate back to the DR5000, by utilising some other ports on the DR5000 or something. Or whether someone else had done the same, and managed to repair the DR5000. Grasping at straws, I know! I do have a Loconet-XpressNet DR5099 Diginethub which had me excited for half a second, until I Googled that the XpressNet and LocoNet sides are completely separate ...

Did you ask Worsley for those especially, or are you re-using parts he has available for other coaches/EMUs?

That's brilliant, thanks for all the information. Just what I was after.

Do you use the LokSound decoder in your 92s? Which version? I only picked a 92 up last week, and there were no sound versions left, so had to go DC. Not sure if I want sound or not in the 92 yet, but it'll be good to know the decoder you use.

Thanks Richard, I'll take a look at that one.

Hi guys, Before I buy some decoders and potentially regret it, I've got a couple of questions. I've picked up a Dapol HST-P and class 33, and a Revolution Class 92 recently. They're all Next18, and are my first locos designed for Next18. All of my 6-pin decoders are Zimo or Lenz, and would like to stick with Zimo for N18 where possible. 1) It seems that 4 function decoders (like the MX618N18) will be absolutely fine for locos which have independently switchable red/white lights in each end, and no other functionality? 2) The Revolution Class 92 doesn't actually specify a 6-function decoder, but it has cab lighting at both ends, and independently controlled red and white lights on each end - which seems to suggest it'll require a 6-function decoder for full functionality? 3) It seems that it's only the Dapol's Imperium2 and the LaisDCC Next18 decoders that have 6 function outputs - and the LaisDCC seem to have a poor reputation, and I've learnt that the Imperium2 is a re-badged version (at twice the price!) Is this reputation really deserved? Have I missed other options? I'm aiming for automated control of my layout, and don't want to be frustrated with poor quality decoders, and would rather sacrifice the cab lighting if the Dapol decoder isn't great. 4) Isn't it frustrating Dapol don't have downloadable loco instructions? Although I doubt any of their Next18 locos have a requirement for more than 4 functions, I would like to double check before any future purchases.

Following with interest. I've got the bits to make a start on one myself, but I'm busy laying track at the minute so it's probably a couple weeks before I do. I'm going to try two, one with the Electra vinyls in BR Blue/Grey, and one just BR blue for ease of painting!

Sorry to post to an old thread. Just wondering if it was a 121 or 122 you used as the motor chassis? Your posts mention both. I imagine both fit just fine, but thought I’d check!