Jump to content
 

Waton


eldavo
 Share

Recommended Posts

  • 4 months later...
  • RMweb Premium

Been a while since anything was posted here and as Waton is due out at Uckfield next month it's time for some activity. During the original challenge only one of the required signals got built due to time constraints. Time to sort that out.

The first signal was built using surface mount LEDs and was a bit of a pain to say the least. Need to find a better and more reliable method.

The MkII signal heads are being built in a slightly different manner. First up a baseplate of clear perspex is cut and the LEDs are stuck face down into a layer of quick set epoxy. Here's a 4 aspect with a feather under construction...

post-7010-0-36915800-1504644505.jpg

With the LEDs nice and secure they are wired up using enamelled copper wire. A pretty tricky operation given the size of the things!

post-7010-0-52508900-1504644518.jpg

Good idea to test everything at this stage. Of course forgetting to use a resistor and blowing the LEDs would be a really stupid thing to do...

All being well the wires are reinforced by adding another layer of epoxy. In the pic below a 4 aspect head is at stage one alongside. Double sided tape helps...

post-7010-0-83928800-1504644535.jpg

Slow going but progress none the less.

Cheers
Dave

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

  • RMweb Premium

Bit more messing around today. Another 4 aspect head wired up after much rework. Soldering these tiny LEDs is seriously tedious. This and the one with a feather will be mounted on a gantry at the western end of the platform controlling westbound movements from the platform and goods relief road.

Having wired the heads and reinforced things with copious quantities of epoxy it's time to look at the front. Bits of brass tube have been cut and filed to produce lamp hoods. They aren't strictly to scale but have been super glued to the front side of the perspex base. I decided hoods for the 5 feather lamps was going a step too far so currently I just have a bit of brass strip with holes.

These will need testing then a coat or two of paint then it will be time to see how they look on the gantry...

post-7010-0-38719900-1504644661.jpg

Cheers
Dave

Edited by eldavo
  • Like 5
Link to post
Share on other sites

  • RMweb Premium

Further work on signals sees 2 MkII signal heads fitted to the gantry destined for the west end of the platform. They seem to have survived painting and fitting and amazingly all the aspects still work! There is a bit of a problem with light bleed through but I may have to live with that. Onwards to MkIII.

post-7010-0-28314600-1504644726.jpg

Cheers
Dave

Edited by eldavo
  • Like 4
Link to post
Share on other sites

  • RMweb Premium

Pre-wired LEDs might well make things a bit easier and more reliable. Link duly noted.

Not done much on the signals themselves today but been playing with this sophisticated test rig. Maybe not immediately obvious what it's all about...

post-7010-0-23500900-1504644850.jpg

Obviously it's a bit of Peco N gauge track balanced on a bit of 5mm ply. The interesting bits are actually underneath...

post-7010-0-54323100-1504644859.jpg

There are a couple of resistors connected to an Arduino processor board but the really interesting bit is an infrared proximity detector, actually a TCRT5000L. It's the thing with 2 bits of brass tube stuck on the top. These have been added so it can be mounted under the baseboard without detecting the board but is able to sense things through a couple of small holes.

So what is all this palaver all about you may ask. The obvious way to control colour light signals would be with a few switches but that is just too simple. I want the signals on Waton to switch to red automatically as the train passes and to reflect the state of the next signal down the line when they are cleared. Bit tricky to do with simple switches.

The infrared detector will be mounted under the track beyond the signal and will allow the Arduino board to sense that the train has passed the signal. In addition the state of the next signal down the line will be sensed so when the signalman pulls the signal lever it will change from red to green, yellow or double yellow as appropriate. That's the plan anyway.

This actually only needs a relatively simple processor so the Arduino Mega 2560 with 52 inputs is a bit of an overkill. What I will actually use is an Arduino Nano which is a lot smaller...

post-7010-0-36507200-1504644871.jpg

Lots to do.

Cheers
Dave

Edited by eldavo
  • Like 3
Link to post
Share on other sites

  • RMweb Premium

The westbound gantry has now had a ladder and safety hoops added and planted on the layout. Of course this means dismantling half the set up but...

post-7010-0-01345300-1504645028.jpg

Not much is to scale but it doesn't look too bad.

Work is underway beneath the board as well so holes have been drilled for the infrared sensors...

post-7010-0-96548800-1504645040.jpg

Cheers
Dave

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

  • RMweb Premium

With the signals installed above the board it's been wiring time. Not everyone's cup of tea I know but some of like the challenge!

The enamelled copper wires form the signal head emerge from the bottom of the gantry upright and have been terminated on a scrap of salvaged tag strip. Each of the signal heads has a current limiting resistor in the 0 volt line...

post-7010-0-50048100-1504645212.jpg

A pair of infrared detectors have been glued into holes excavated in the baseboard and wired back to the vero board with the Arduino mounted on it...

post-7010-0-61300200-1504645201.jpg

Ribbon cables run from the Arduino outputs to the signal head tag board and switch inputs run back to the off board connector.


post-7010-0-23646300-1504645194.jpg

Amazingly this lot seems to work! There was a serious face palm moment when I realised I had wired the LEDs of the detectors to the 12volt supply rather than the 5volt supply from the Arduino. Not surprisingly this destroyed the LEDs. Luckily I had spares.

The light bleed in the signal heads means I'm not happy with the yellow aspects so currently I have it set up only to use the red and green. Lots more to do...

Cheers
Dave

Edited by eldavo
  • Like 5
Link to post
Share on other sites

  • RMweb Premium

Those fed up with twaddle about signals should change thread now! A MkIII 4 aspect head is under construction. This one has the LEDs mounted on a bit of plasticard with holes albeit still using epoxy.

post-7010-0-51223600-1504645394.jpg

After wiring up the back is liberally smothered in epoxy as per the previous version...

post-7010-0-94163800-1504645407.jpg

The light bleed looks to be better so far but we shall have to wait and see. Currently I'm waiting for the glue on the brass tube light cowls to harden so everything can be painted.

post-7010-0-27751600-1504645420.jpg

A second Arduino board has been prepped ready to drive this and another signal head. These are on a different baseboard to the previous gantry. Testing of the gantry signals has removed a couple of unplanned features (bugs) in the software. The signals now nicely pop back to red once a train passes over the relevant infrared detector.

I have also added a lunchtime/lazy Sunday afternoon feature. If a train passes the signal it returns to red but if the signalman does not put the lever back in the frame within 20 or so seconds the signal will again clear to green ready for the next train. This means a lazy operator can set up a route and run a series of trains only having to select the required train in the fiddleyard.

Next up is to fit two more infrared detectors and adjust the wiring in the control panel. Hopefully once I've done that the missing circuit board header sockets I need to complete the second Arduino installation will have arrived courtesy of the postie and I can wrap it all up.

Cheers
Dave

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

  • RMweb Premium

The signals, all four of them, seem to work. To get reliable triggering of the infrared sensors I've added a strip of copper tape to the underside of all the locos.

post-7010-0-87684500-1504645579.jpg

Surprisingly I seem to be making progress!

Cheers
Dave

Edited by eldavo
  • Like 6
Link to post
Share on other sites

Fantastic work Dave and also a great photo. Track work looks great also.

 

Any plans to make the platform lights illuminate (or is that just too easy). One of the humans could also be puffing on a ciggie (glowing and puffs of smoke as well). Is that stupid or a challenge?

 

Eddie

Link to post
Share on other sites

  • RMweb Premium

...Any plans to make the platform lights illuminate (or is that just too easy). One of the humans could also be puffing on a ciggie (glowing and puffs of smoke as well). Is that stupid or a challenge?

Eddie

It's always bright sunshine in Waton so the eco friendly lighting turns itself off. The station area is a no smoking area! :)

 

Cheers

Dave

  • Like 1
Link to post
Share on other sites

  • 2 weeks later...
  • RMweb Premium

Love the work on the signals. I guess suitable plastic tube isn't available to make the lamp shades for each aspect? If it was a better bond would be possible and quicker construction, just a thought.

 

Also love the control system. When I was college spent sometime figuring out a modular signal control system based on TTL 74 series logic chips, which each module would input the next, so the aspects would work as found in automatic signals. Old hat to the technology your using!

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...