LED shunt signals

[Deleted]

I've noticed that some of the shunt signals on the Central (and elsewhere) are starting to get "tatty", i.e. some of their LEDs are going out.

Considering how reliable LEDs are supposed to be, why are these shunt signals so susceptible to failures?

[piccadillypilot]

theonekea said:

Considering how reliable LEDs are supposed to be, why are these shunt signals so susceptible to failures?

How long have they been installed?

[Tom]

theonekea said:

Considering how reliable LEDs are supposed to be, why are these shunt signals so susceptible to failures?

Because they aren't LED's, they're fibre optics.

The lamp used is a 12v 55w halogen lamp, almost identical to the ones used as car headlights, and they tend to have a very short life.

[Deleted]

Oh!

I didn't know they were fiber optics! Why fiver optics instead of LEDs?

[piccadillypilot]

theonekea said:

I didn't know they were fiber optics! Why fiver optics instead of LEDs?

When the Central was resignalled some ten-twelve(?) years ago LED's were not in their current state of development and the railways hadn't found out how to protect the signalling circuits from any spurious emmisions, or even if protection was needed.

[Deleted]

piccadillypilot said:

theonekea said:

I didn't know they were fiber optics! Why fiver optics instead of LEDs?

When the Central was resignalled some ten-twelve(?) years ago LED's were not in their current state of development and the railways hadn't found out how to protect the signalling circuits from any spurious emmisions, or even if protection was needed.

In what way would LEDs need protection from spurious signalling circuits? My knowledge of LEDs stop at the fact that they are binary diodes that only work one way at one voltage...

[Phil]

Wrong way round- it's the signalling that needs protecting from LED emissions. The latter are not continuous but pulsed (like a laser) and it must be proved that this pulse frequency will not affect signalling or track circuit frequencies

[Deleted]

The pulsing is not inherent in the way that LEDs work. They will run on pure DC of suitable voltage and in this mode do not produce anything that could interfere with other apparatus.

In displays that need to change their appearance (such as, for example, the 7-segment displays used in clocks, calculators), it simplifies the driving circuits if the LEDs are "multiplexed", which involves turning them on and off very rapidly. This multiplexing produces emissions which could affect other equipment.

For a railway signal head, there is no need for multiplexing: either all the red LEDs are on or they are all off. The signal head could therefore be run off DC and would not produce emissions.

However, it is only in the last four or five years that LEDs have been available sufficiently large (10 mm) and sufficiently bright for use in signals. If the Central signalling is a decade old, there would not have been any suitable LEDs then.

In Sydney, we are starting to get road traffic lights and railway signals that use LEDs. The units seem to be just bolt-on replacements for the previous tungsten lamp units. But here we don't have the UK H&S requirements.

[Tom]

piccadillypilot said:

When the Central was resignalled some ten-twelve(?) years ago

A bit longer than that - the first site was West Ruislip the best part of fourteen years ago.

[Phil]

sydneynick said:

The pulsing is not inherent in the way that LEDs work.

the emissions themselves are particles of light at high frequency. If sub-frequencies were to happen they COULD affect signalling frequencies unless proved otherwise.

[russe]

the emissions themselves are particles of light at high frequency. If sub-frequencies were to happen they COULD affect signalling frequencies unless proved otherwise.

Phil - visible light, which has a frequency in the region of 10¹⁴Hz, doesn't produce or contain harmonics.

Russ

[Phil]

russe said:

the emissions themselves are particles of light at high frequency. If sub-frequencies were to happen they COULD affect signalling frequencies unless proved otherwise.

Phil - visible light, which has a frequency in the region of 10¹⁴Hz, doesn't produce or contain harmonics.

Russ

Fair comment , but as in all these situations the burden of proof is that the new system does Not interfere with the existing , rather than the other way round