Jump to content
 

Dongits Model Railway -- Three layers around the Study


Bloodnok
 Share

Recommended Posts

All six tracks glued at the yard exit, and point motors fitted:

Yard_Exit_2.jpg.558a0626e25dc09e1f4c79d4f64f1067.jpg


And five tracks complete through to the yard to the entrance:

Yard_5.jpg.f945079cbb14407a91c23116f8acda93.jpg

Time to get that sixth and final yard track installed...

Edited by Bloodnok
Picture Fixin'
  • Like 3
Link to post
Share on other sites

A fairly productive day yesterday. The 6th and final yard track is installed and complete:
 

Yard_Complete.jpg.a0a1fe35f3db2c601b8b34c3c07f1629.jpg


All the point motors have been built and installed at the yard exit:
 

Yard_Exit_Motors.jpg.a54bd26a9441966ac15296bc9a5ebd5e.jpg

 

The baseboard for the horseshoe into the gradient has been finished and has underlay on it:

 

Horseshoe_1.jpg.40c32c741250b75786cbe14f67a7df36.jpg

 

... and I started building the first 8 block detectors. 

Edited by Bloodnok
Picture Fixin'
  • Like 2
Link to post
Share on other sites

After a lot of cleaning up and moving stuff about, the underlay is complete at this end of the layout:

Yard_Approach_1.jpg.9df835d28624865dea3098f0490ff93d.jpg

And the two-to-three track junction in the gradient (which is one end of the reversing section) has been located and is gluing:

Lower_Gradient_Junction_1.jpg.fb5b185b90e85547363beae385cbd6fb.jpg

(... now covered in weights).

I need more servos to arrive before I can mount point motors on these points.

I also finished the track circuit detectors, but then discovered the cable I was intending to use to connect them to the layout bus has a blind pin in the socket. I needed a 10 pin cable, and this one only had 9 of it's 10 wires connected, and the tenth doesn't even have the hole in the plug, so it won't even connect the other 9.

I've ordered new cables, although the delivery estimate is early July to mid August -- doesn't fill me with confidence.

Edited by Bloodnok
Picture Fixin'
  • Like 2
Link to post
Share on other sites

12 more servos have arrived, and I have now built enough point motors to get me as far as the visible surface of the layout.
 didn't actually order 12 servos though, I ordered 30. I've asked the vendor where the other 18 got to. They have promised to actually ship them this time...

I'm expecting a couple more servo controllers soon(tm).

The next 'pinch point' in construction is that I'm running out of insulating rail joiners. I have a grand total of 7 left (I'm not counting the one that went 'ping' -- the chances of actually finding that one are minimal). Apparently I've got through all the packets I bought, while ending up with more unopened packets of metal ones than I started out with. There must have been some spare packets lurking somewhere. Shame there wasn't any of the plastic ones lurking spare really. (Or maybe there were, and I've used those too?).

While those are on order, I'll be mostly building electronics. I have a DCC booster to build, plus some asymmetric braking sections (and their controller) to build. I've also been told I should have done my block detectors differently, so I'll need to modify those.

Link to post
Share on other sites

  • RMweb Gold

I’ve been inspired by this topic to name my new layout ‘Lockdown End’. Thank you for the idea. I’ll never forget when I started building it! But running a bit behind where this layout is - still early days.

 

 

Edited by ITG
Added link
  • Like 1
Link to post
Share on other sites

"I'll be mostly building electronics".

<Cue the kazoo music>


Yard_Entrance_6.jpg.dffe74438e28d64a9850716f8495d3f2.jpg

 

Horseshoe_2.jpg.001c5579b7d38d4fb85f7d484e76c621.jpg

... Amount of electronics built: Zero.

That should change now, as I've used the last insulating rail joiner. Well, unless I find the one that went 'ping'...

Edited by Bloodnok
Picture Fixin'
  • Like 2
Link to post
Share on other sites

Yesterday I built two more servo controllers. They have been power-tested and are awaiting installation and configuration:

1322032144_CANMIO-SVO_2__3.jpg.51c62af79db948d77bb951156099bf33.jpg

 

This DCC booster kit is at the 'initial low-current testing' phase where I make sure the relevant sockets are receiving the right power voltages before I finish construction:


NB1B-Initial-Testing.jpg.86a51d040d02a4cb4106cd8a60b441b9.jpg

I built it in the 5A configuration.

I got stuck with this relay driver, as one of the capacitors is missing from the bag of parts:

CANVOUT_missing_C7.jpg.a5f87d7967be84412c2fbc580a678074.jpg

I need to source a replacement C7.

 

Once complete, I'll be using this to drive relay controlled ABC braking sections.



Oh, and I have to call this out, too. Take a bow, Rails of Sheffield. You have outdone even Amazon in the wasteful packaging challenge.

You managed to send me a couple of packets of insulating rail joiners, which would happily fit in an envelope ... in a box large enough for a large OO gauge locomotive.

Edited by Bloodnok
Picture Fixin'
  • Like 2
Link to post
Share on other sites

  • 2 weeks later...

Wow, did I ever forget to keep this updated :(

Oops.

 

Outside of the overtime I'm doing for work (which is why I've not been on here so much), I finished laying the storage yard track and I'm now wiring up the bottom level of the layout.


I have the electronics mounted in four of the concentration points. There will be more of these later when the upper levels are more complete. The servo controllers are all built, as is the DCC command station and booster. The canbus data cabling is in, but I'm now waiting for some reels of heavier gauge wire to arrive before I complete the power distribution and DCC block wiring.

Edited by Bloodnok
Picture Fixin'
  • Like 2
Link to post
Share on other sites

  • 4 weeks later...

Lessons learned today: Check points for electrical conductivity *before* laying them.

I have spent quite a while today trying to work out what was wrong with my wiring where the frog of one point was not reliably powered. Solution: One of the three rails that make up the frog was not connected to the other two. So if I tried testing connectivity on that rail, it would always fail. Hence failing about 1/3rd of the time, and not with any correlation to which way the junction was set.

  • Friendly/supportive 1
Link to post
Share on other sites

I wanted to upload some video of things like point motors moving, and that required creating a YouTube channel. Which required a name. Which kinda forced me to actually make a decision.

I put a first video there which is mostly a slideshow of what's been posted in this thread with a few extra / different shots here and there:
 


... Actual video content (of things which move!) coming soon.

Edited by Bloodnok
  • Like 1
Link to post
Share on other sites

Aaaaand ... it now works right through to JMRI. I have a configured panel, I can set junctions and routes, and the first train detector is working -- the block goes red in JMRI when a train is on the track. Well, modulo one dodgy joint I apparently completely forgot to solder. But I've fixed that now.

... Concept proven. Serial production can now commence.

  • Like 3
Link to post
Share on other sites

  • 2 weeks later...

A short video of the train detectors working:
 


The train detector is a MERG DTC-8, which detects the presence of a vehicle by current draw using the coils as transformers. Any kind of power draw (whether it's a DCC chip, a coach lighting circuit, or even just a resistor or capacitor across an axle) will trigger it.

The feedback bus is MERG CBus. I've got the DTC-8 connected to the expansion socket of a CANMIO-SVO (which is also driving my points).

The software is JMRI, which I'm fairly new to -- I haven't got it to do anything other than display occupancy so far, but the data is reading reliably.

  • Like 5
  • Craftsmanship/clever 1
Link to post
Share on other sites

Just now, FoxUnpopuli said:

Fascinating.  I'm looking forward to seeing more of the block detection and how you will use it to signal (and automate?) the layout. 


The problem I have with many layout automation systems is they are all or nothing -- you can automate part or all of a layout, but there is a hard boundary between the automated part and the manually driven part. You can't just weave a manually driven train in between the automated ones.

My plan is to use the block detection to drive a signalling system implemented on the computer, and have that signalling system include ABC for the track. This means that manually driven trains will automatically stop at red signals, but remain under manual control.

In that way I can run multiple trains without having to speed match them all manually and they will self-regulate using the signalling system. All I need to do is start them off in the first place.

Assuming I can write some ARS magic into the signalling system, I can then have trains with unique paths, which will produce an effect sufficiently close to full automation when I want to 'watch the trains go by', but still allow me to weave my own train through the other traffic.

  • Like 1
  • Agree 1
Link to post
Share on other sites

1 hour ago, Bloodnok said:

The problem I have with many layout automation systems is they are all or nothing -- you can automate part or all of a layout, but there is a hard boundary between the automated part and the manually driven part. You can't just weave a manually driven train in between the automated ones.

 

I see we have foudn the same issue.  I wish you luck finding a solution - looks like you're well on your way.  :)

  • Like 1
Link to post
Share on other sites

50 minutes ago, FoxUnpopuli said:

I see we have found the same issue.  I wish you luck finding a solution - looks like you're well on your way.  :)


I think it's the almost two decades of messing with train simulators that has ruined "traditional automation" for me. Once you get back to software, I don't see why all the same features wouldn't be possible.

Link to post
Share on other sites

  • 1 month later...

Welp. This went a bit quiet.

I claim poverty from sudden and unexpected expenses. No disposable income for about two months means no progress on layout. But I'm back now. I've built a second train detector, and I have two more kits in stock to build (I might try shooting a timelapse of making one. Perhaps people would be interested in seeing that?), and I've also been chipping things that have been in storage since forever (so I have more stuff I can run as the operational area of the layout increases).

I have had a few issues though. My chosen fleet standard decoder, the Zimo MX600R (which I selected for it's cheapness, good overall feature set, and critically, configurable ABC stopping distance and behaviour) fits really well in most diesel locos -- it's narrow and thin, and is a natural fit in many a roof void. However, at 26mm long, it's proven a bit too long to fit in some other types of vehicles I've tried it in. A Bachmann 08, for example, expects the decoder vertically in the nose, and has about 25.5mm length available. The body will go back on, but there's a distinct seam that wasn't as obvious before. The trailer cars of a Bachmann 108 expect the decoder to sit vertically in the toilet -- and provide only 24mm or so length. The body does not go back on with one there.

I also had a surprise on opening my Hornby Class 60, to be greeted by ... a Zimo MX600R. Contrary to what my records indicated, I had in fact chipped it previously, and just failed to record this fact. Oops. Spreadsheet and JMRI roster updated...

Edited by Bloodnok
  • Like 1
  • Informative/Useful 1
Link to post
Share on other sites

James,

 

I have am using  mainly Peco 75 Bullhead but have recently bought a box of SMP flextrack from Marcway in Sheffield. The sleepers are not as thick but I have remedied this with sheets of Plasticard. The difference is basically 1mm so I've just bought a pack of Plasticard to boost the height to bring it level with the Peco.

 

Not sure which method I will adapt yet but I'm thinking of cutting individual strips and gluing them to each sleeper individually.

 

The SMP range also offers a good choice of points etc....

 

Paul C.

Link to post
Share on other sites

  • 2 weeks later...

I've been installing more train detectors. I've reached the point where all the track is now powered. This means for the first time I've been able to run a train completely around the layout:
 



... Next task is to do things with the block occupancy data beyond just displaying it on a panel.

This means implementing some signalling logic, and then driving ABC modules from the signal state.

Edited by Bloodnok
  • Like 2
Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
 Share

×
×
  • Create New...