Festiniog Railway signals

[Trevor Thompson]

It has taken me a while to get here but I now have 10 working disc signals.

I have modified Rick's programme (what is called a "sketch" in Arduino) to work with the ESP32C3 boards which I purchased online.

Otherwise electronically it is a copy of Rik's work. If anyone is interested in the "sketch" I can add it.

Here is a short video to demonstrate all 10 of them working on the bench: https://www.dropbox.com/scl/fi/j6y3h94m ... ezbxc&dl=0

You can see the lamps showing red for stop and white for clear.

I have to print off 10 sockets to mount in the garden before I can show them working on the railway.

Here are photos of the "caution" signals I made to go alongside the steps where the footpath crosses the railway. These don't move, and are intended as fixed "caution" signals with "W" written across them to indicate that the driver should sound a whistle. The green light indicates "caution" (well it did in 1880):

IMG_4345.jpeg (2.97 MiB) Viewed 41099 times

IMG_4346.jpeg (2.31 MiB) Viewed 41099 times

These are really an indulgence because I can find no evidence of yellow discs with a "W" from the 1880's, but it is something which was done in the early preservation era, so why not?

You might also notice that the two types of signal are slightly different - because after printing the 10 disc signals I found a drawing of these signals and modified the drawing to more accurately reflect the prototype. The handrail, the footplate, and the footplate brackets are different. Also the base of the column has the correct webs and ornamentation. I have not implemented the scalloping on the column itself.

So now on to the double arm signals.
I am hoping to use a different approach to these by using an ESP32C3 board to act as a transmitter using low energy Bluetooth, and another ESP32C3 board to act as a BLE (Bluetooth low energy) receiver to decode the bluetooth signal and control the two servos on the signal. I will let you know if I mange to get that to work.

Trevor

[philipy]

One word... SUPERB!

[Paul_in_Ricky]

....Also the base of the column has the correct webs and ornamentation. I have not implemented the scalloping on the column itself.

On the Fespedia page about these signals https://www.festipedia.org.uk/wiki/Disc_signals it looks like the bases are actually square, not round.
A shame not add the scalloping as it adds great character.

Festdisk.jpg (439.66 KiB) Viewed 41060 times

[Trevor Thompson]

Yes you are correct - the square base isn't obvious from the drawing but it is clear in the photo.

I would have added the scalloping if I hadn't finished printing the 10 before I found the drawing.

I like your print - can we see the rest of it- and a bit more information about how you are using it?

Trevor

[Peter Butler]

Looking back to the beginning of this thread I see you experimented with a variety of shapes for the base, including square but, right or wrong, I like the look of the final rounded version best of all.

[Paul_in_Ricky]

This is the full signal. Roughly assembled as it's awaiting the weather to warm up so I can spray it in my, unheated, workshop. The black PET-G masks some of the details, plus there's things like the lamp lenses already printed for fitting after painting.
Not happy with the ladder yet. It should have round rungs 1mm dia which aren't easy to print, or more accurately getting supports off has too much collateral damage. I'll probably have another try at that when painting time is closer.
The column is hollow with brass tube bearings top and bottom to allow it to become practical in time. Not quite sure whether I'll use servos, steppers or ordinary motors. It's lead me into Arduino territory for control, so that's a new thing to learn. I quite like the idea of making it wirelessly controlled with either Bluetooth or WiFi.

FR sig.jpg (101.41 KiB) Viewed 41016 times

[SimonWood]

Here is a short video to demonstrate all 10 of them working on the bench: https://www.dropbox.com/scl/fi/j6y3h94m ... ezbxc&dl=0

There is something deeply satisfying about this video. I could watch it on loop for hours! Fantastic work.

[Trevor Thompson]

Paul,

Very clean and detailed print. What printer are you using to get these results? How was the column orientated when you printed it?

As a result of your comments on the fluting I have had a go with it. Drawing the fluting was straight forward, but getting a quality print from it is a different matter!

A photo to illustrate:

IMG_4348.jpeg (2.04 MiB) Viewed 40957 times

In the background is the point indicator with no fluting. It was printed on its side with the steps uppermost. It required some cleaning up with a file - but it is quite satisfactory for use. In the foreground is the fluted version of the point indicator, again printed on its side. The bottom would be OK, when the remains of the supports have been filed away, the sides are crisp, but the fluting has disappeared on the top - lost in the difference in height for each print layer.

In the background on the right is the fluted version printed upright. The fluting is just about visible, but as the column rises it gets lost in the flexibility of the column.

I have found that printing columns upright is possible, up to a point, but at a height of about 180mm the print falls off the bed whatever base I add to it. In the same way the taller it gets the more flexible it becomes and the definition deteriorates. Printing a square column becomes oval as it gets taller.

I suppose it might work if I added buttresses attached at various heights, to prevent the column flexing, which could be removed afterwards like supports.

I will come back to experiment further with fluted columns when I start on the point indicators properly. In the meantime the two arm signals progress:

IMG_4349.jpeg (1.79 MiB) Viewed 40957 times

I have printed the tapered rectangular columns flat on the bed in two halves. The upper face is the inside face, which has a half of the wire duct in it. The halves are at the front in the photo. At the back are the sockets that the arms fit into, again printed in halves but this time with the slot for the arm vertical and the wire duct on one side. I tapered the ends of the slot to avoid having to add supports.

Finally the first signal so far:

IMG_4350.jpeg (2.79 MiB) Viewed 40957 times

The four pieces that make the assembly have been cleaned up and bonded together.I have added a non prototypical socket to reinforce the bond between the column and the base. The base has a slot in it on its underside to continue the wire duct towards the camera. That section of the base nearest the camera is to take the servo/battery box which will attach to the end of the base extension.

[Paul_in_Ricky]

Mine were printed on a Bambu Labs A1 with 0.4mm nozzle with BBL PET-G HF onto a PEI plate.
Optimal 0.16mm layer settings with normal supports in Bambu Studio (That's just a rebranded and slightly cut down version of Orca slicer).
The column is 192mm high and printed vertically without problems. I was surprised and delighted how easily & cleanly the support structure peeled off leaving little clearing up to do.

I did consider using a 02.mm nozzle to improve the fluting and gradients quality, but the x3 increase in printing time was something I hadn't the patience for at the time. I also haven't been satisfied with the PET-G HF with the 0.2 nozzle, but may have another go with PLA+ with the 0.2 sometime in the future.

In theory printing horizontally ought to give increased strength, but with default settings getting the supports off would leave a lot of clean up work to be done. As I don't intend to use the signal anywhere it could get handled roughly it's probably not worth the effort.
With the multi filament AMS unit it would be possible to use the water soluble PVA filament for support, but that's not something I have access to.

[Trevor Thompson]

I have made a bit more progress on the signals, in that I have the first two point indicators ready to go out onto the railway. They will be attached to the points I have rebuilt for the lower terminus. They work, but the way they connect to the point needs fine tuning:

IMG_4371.JPG (1.97 MiB) Viewed 40515 times

That will have to wait until it stops raining!

Trevor

[ge_rik]

Love the fluted columns

Rik