Interesting debate

230v nominal well that is for the maths part.  Now speak to someone in Yorkshire and they will tell you they get 247 to 250v on a regular basis.

As for the Manufacturer who make Traffic signals using incandescent lamps they need to speak to their R&D and get working on the LED version.  The cost savings for whoever pays for the street sign / Traffic signals pence/kWh should be a convincing factor.  100w lamp vs 15w LED Lamp.

I think they started to swap out the lights in London around 2014 from incandescent to LED lamps

Sergio Fernandez said:

As for the Manufacturer who make Traffic signals using incandescent lamps they need to speak to their R&D and get working on the LED version.  The cost savings for whoever pays for the street sign / Traffic signals pence/kWh should be a convincing factor.  100w lamp vs 15w LED Lamp.

The failure mode for most LED lamps is often to a stroboscopic output as opposed to off.

I’m sure there are LED traffic signals – I presume they’d need a bespoke driver so can’t be easily retrofit. I’ve got a client who manufactures railway signals with LED’s, and a former one who manufactures LED runway lamps. Both are subject to very stringent requirements so that they fail to a known safe state.

Incandescent lamps are a lot easier on this front – their failure modes are more obvious. If the controller detects an open circuit then it knows there’s a fault and can degrade into the safest state which in most cases is to trip all signals such that motorists will interpret as caution and give way. 

To me it seem quite obvious that LED used for outdoor like the following

Traffic signals

Runway lamps

Railway signals

should fail to safe. If they don't then there is common sense lacking and the design team need to be re-educated as to what the failing are.  As you say Strobe, which is annoying enough on a garden PIR floodlight but it is absolutely dangerous for a traffic signal or a run way at night in the rain (just adding risk to emphasise the point for future readers) I would then also go 1 step further


Get that unit to report back to central office for traffic signals or rail signal or control panel at an airport of its failure.  Designers need to make kit that is fit for purpose and fit for peer review.
UK traffic lights could be fitted with IoT technology quite easily.

The airport would also get further enhancements like secondary/redundancy light sets.   

It's all very well that a person knows the difference between these states.

But you've got do design a machine that knows, and meets a SIL rating. What happens if the visible front of a lamp appears off, but the monitoring equipment which can only see current and voltage doesn't see this? That's not too difficult if you're on a clean sheet design, but if you've got an existing mature design then it's not something you can just change.

Safety engineering is very hard work unfortunatly.

The Designers will need to work on it longer until

it is fit for purpose and fit for peer review.

Too many products come to Market before they are ready for the Market.  eg AFDDs

Which is why where things really matter, we often have the equivalent of incandescent lamps longer than you might like, such as a telephone standard with voltages and current levels based on carbon granule microphones and moving iron earpieces, with complex electronics at both ends emulating the original kit. Or, if it is a railway telephone, sometimes the actual original kit.

Mike

Which is why where things really matter, we often have the equivalent of incandescent lamps longer than you might like, such as a telephone standard with voltages and current levels based on carbon granule microphones and moving iron earpieces, with complex electronics at both ends emulating the original kit. Or, if it is a railway telephone, sometimes the actual original kit.

Mike