Improving the LS Models
VTU coaches (5)

Original page created on 07/02/2021; updated on 01/01/2024.

Lighting, episode 2

As already mentioned, the new PCB has one LED per window. Here is the layout.

The lines represent the series mounting of two or three LEDs depending on the case. Due to the large number of LEDs, I have provided two 100 µF or even 220 µF capacitors in the technical rooms, opposite the toilets.

Mounting of the strip, side view.

The 220 µF (⌀ 8.2) capacitors fit well with a bit of force, as long as the clips visible at the yellow arrow are half-trimmed. I will realize later that the wires get jammed between the strip and the roof. Remedy: a ⌀ 1.6 hole for their passage.

View of the two B¹⁰tu in C1 livery equipped.

Here are two pictures of the B⁵rtu bar coach in Corail livery:

That is not bad, but it’s not the result I hoped for at the beginning, and I’m not sure if it was worth the complication compared to a classic strip.

Finally, a little technical information: despite the high number of LEDs, the total consumption is less than 5 mA.

Lighting, episode 3: electroluminescent wire

Still in quest of more realism, I’m looking for other solutions than the classic LEDs. What I need is a continuous (or almost continuous) lighting above the ramps.

I try out electroluminescent wire, called EL Wire in the English literature. I quote from the Addafruit website.

EL Wire, also known as Electroluminescent wire, is a stiff wire core coated with phosphor and then covered with a protective PVC sheath. When an AC signal is applied to it, it glows an aqua (blue-green) colour. Sometimes it’s covered with a coloured plastic shell to make it appear another colour. It looks a little like thin neon. Very bendable, it keeps its shape and you can curl it around your finger. It’s an easy way to add some glow to a project, not as bright as LEDs but uses a lot less power!

A lot less energy, of course, but not for the same amount of light! Operation requires a relatively high AC voltage, ideally more than 120 V AC, and a minimum frequency of 2 kHz (5 kHz is generally recommended; I have tried 20 kHz, with better results).

To obtain this voltage, there are two solutions:

• either a low voltage oscillator followed by a step-up transformer;
• or a specialized integrated circuit.

Disadvantages of these solutions:

• As for the first solution, the transformer, even if miniaturized — the higher the frequency, the less iron (magnetic circuit) is needed — is relatively bulky. The dimensions of the one in the circuit I bought for the tests are: L × W × H = 13.4 × 11.7 × 10.4, not easy to fit in with discretion.
• Concerning specialized circuits, they are often unavailable, some are very tiny (I guess they are designed for mobile phones), so difficult to solder, and their voltage usually doesn’t go high enough. Examples:
  • Maxim MAX14514. Output voltage 300 V peak-to-peak (106 V RMS). 3 × 3 mm TDFN-EP package, pin pitch 0.5 mm!
  • Microchip MIC4833 (no longer manufactured);
  • Supertex HV857. Output voltage 190 V peak-to-peak (67 V RMS, insufficient). MSOP package L × W × H = 3 × 4.9 × 1.1 mm, pin pitch 0.65 mm;
  • MP3801: not found. Very small too: 3 × 3 mm.

Test of the wire bought from Ebay

Length of the wire: 5 m. Powered by two AA batteries. Control by push button. First press on the push button: constant lighting; second press: flashing; third press: different flashing; fourth press: off.

The light emitted is bluish (I knew it) and really weak.

Here is the appearance of the PCB contained in the battery box.

The top side shows the step-up transformer, the control push-button, a signal LED and the battery contacts.

Power consumption: 250 mA at 3 V, 500 mA at 5 V, 1100 mA at 7 V (the transformer then starts to heat up). The brightness increases significantly with the voltage. But the current consumption is high — for five metres for sure, whereas a coach would need about 450 mm, hence a supposed current of about 50 mA, which is still much higher than for my LED circuits.

I forgot to mention that the main advantage of EL wire is that it can be cut at will, and that all the segments obtained remain usable.

In fact, one observation made me give up this idea: the wire is really not bright enough, and the light is too bluish. Of course, it exists in yellow, but then too yellow!

The quest goes on…