How do you prove an LED is actually lit?

[jmh]

I suspect the answer to this to have something looking at it but I need to find a reliable way to find out if a LED is actually lit.

The (12V) LED bulb in question is replacing a filament bulb (I say 'a', there are actually 9 or so in various places). Finding out if a filament bulb is lit is more a matter of seeing if it is taking current and having a fuse able to blow if the bulb fails short. I had hoped we could 'see' what current the LED was drawing but the bulbs we have can take ac or dc and either polarity of the latter, so there is circuitry inside and this may well still draw current even if the LED itself has gone dark (assumed,  not experimented). So sensing current flow may not be a reliable indicator.

Having an LDR and circuitry (or similar thing to sense light) fitted in such a way that external light cannot influence is only way I can think of to do this unless I am missing something obvious / being daft / or there is a more elegant solution? (going to look at each one is not an elegant solution, nor is the series of cleverly placed mirrors I wondered about!)

Thanks

[NiHaoMike]

Use multiple LEDs with their own driving circuits, that way you have redundancy.

[themadhippy]

(going to look at each one is not an elegant solution, nor is the series of cleverly placed mirrors I wondered about!)

run a bit of fiber optic cable from each led to one point were you can if all the leds are lit

[SeanB]

Normally the LED retrofits have a bridge rectifier, a resistor and a led in the package, or simpler ones use 5 chip LED units and a resistor, 4 led in a bridge and the fifth between positive and negative of the bridge, so 3 at a time light. LED's are normally robust, so they rarely go short, but do go open, so simply detecting current flow works, easy for DC by using a sense resistor, that develops around 0.8V at rated current, and a simple resistor to the base of a NPN transistor, so you have an open collector output that pulls low with enough current flow, or use a mux to feed these small voltages into a MCU analogue input, and read the current in.

[tom66]

You can't do it with purely electronic means, you need to meter the light output.  I've seen LEDs fail "dim" but still have about 2.5V across them in the forward condition, usually this is due to the device reaching end of life but can be due to high temperature, etc.  For packaged bulbs, I've seen multiple ones fail in a condition where they are partially / dimly lit (especially common with the GU10/MR50 'halogen replacement' ones).  Would one bad LED be enough to consider the device failed - probably not - but current would still read out of range.  So I think optical testing may be the only way.

[Fraser]

Sometimes reverting to old fashioned technology remains the best and most appropriate solution….. go back to filament indicator lamps  that may easily be monitored for function. The filament lamp has been used as an indicator for many years and should not be shunned when an application benefits from its use. LED’s are excellent technology but they are not a perfect substitute for the humble filament lamp in all applications.

Fraser

[Bud]

It should draw the rated current I'd say. Anything obviously below or above the rated current would indicate a fault.

[SiliconWizard]

It should draw the rated current I'd say. Anything obviously below or above the rated current would indicate a fault.

Monitoring the current should do the trick in most cases and would be the easiest to implement.

The direct, but hardish way would be to use a photodiode and some kind of optical splitter to get light from the LED without getting light from the outside. Probably overkill.

[Siwastaja]

Sounds like a weird requirement. Do you think it's a common failure mode to stop producing light while continuing to pull similar amount of current? I'd say it's far-fetched.

But if you have to absolutely prove that, for some weird legal reasons for example, then you of course have to sense the light, using a photo sensor of some kind. But that's the same for proving an incandescent bulb. And then you need to prove that your light sensor cannot fail, and you need to prove the amount of ambient light coupling to it, ever, in any natural (maybe even supernatural?) conditions, does not exceed your detection threshold.

TLDR; what are you actually doing?

[mariush]

I suspect the answer to this to have something looking at it but I need to find a reliable way to find out if a LED is actually lit.

The (12V) LED bulb in question is replacing a filament bulb (I say 'a', there are actually 9 or so in various places). Finding out if a filament bulb is lit is more a matter of seeing if it is taking current and having a fuse able to blow if the bulb fails short. I had hoped we could 'see' what current the LED was drawing but the bulbs we have can take ac or dc and either polarity of the latter, so there is circuitry inside and this may well still draw current even if the LED itself has gone dark (assumed,  not experimented). So sensing current flow may not be a reliable indicator.

Having an LDR and circuitry (or similar thing to sense light) fitted in such a way that external light cannot influence is only way I can think of to do this unless I am missing something obvious / being daft / or there is a more elegant solution? (going to look at each one is not an elegant solution, nor is the series of cleverly placed mirrors I wondered about!)

Thanks

Such 12v LED bulbs typically have one bridge rectifier to make them compatible with 12v AC input, and then you either have enough leds in series and a resistor to limit current, or there's a led driver.

Usually if one led brakes down (often the bonding wires get loose or burn out inside the package) then the whole series of leds is broken so you don't get light anymore.

Sometimes, you may get two independent "chains"  of leds, and in that case if one led fails open breaking the chain (therefore no longer producing light), the other "chain" would still work.

Rarely you have multiple leds or filaments in parallel

If there's a led driver and the circuit fails, at most you're gonna have pulses of current draw, but average the current consumption would be very low, I'd say maybe a couple mA at most... much less compared to at least 20-50mA or whatever such bulbs would use.

[Fraser]

A Google search for “LED failure detector” revealed that On-Semi produce a component specifically designed to detect LED failures and it can even activate a back-up LED. Likely not what the OP wants due to his use of a 12V LED indicator package but it demonstrates that such detection devices exist as a COTS component.

https://www.onsemi.com/pub/Collateral/DN05066-D.PDF

https://www.onsemi.com/pdf/datasheet/hbl5006-d.pdf

Fraser

[jmh]

Do you think it's a common failure mode to stop producing light while continuing to pull similar amount of current? I'd say it's far-fetched.

My issue is I have no idea how these things may fail. A single LED yes, but all the gubbins in these bulbs, no. I would not expect it to pull the same current of course but it may pull some and I don't know the difference between good and bad. I may be worrying too much! This is why I asked rather the my usual cobble bits together and experiment.

If there's a led driver and the circuit fails, at most you're gonna have pulses of current draw, but average the current consumption would be very low, I'd say maybe a couple mA at most... much less compared to at least 20-50mA or whatever such bulbs would use.

Thanks, in which case then current sensing seems to best way to go.

TLDR; what are you actually doing?

It's to modify railway signal lamps to use LEDs instead of filament bulbs. The bulbs, each has 2 filaments / 3 contacts, take over an amp at 12V and are very expensive things and don't last that long. The trouble here is, with a filament bulb we simply use a relay to determine if it is dead, switch to the second filament and raise an alarm. We need to replicate this for the LED - not to switch to a backup LED but to drop a relay. What is supposed to happen then is all the further signals go to danger, and that bit is easy. These signal lamps are somewhat ancient and will not take the 'real railway' drop-in replacement LED units and anyway we could never afford them.

(I am of course asking in other relevant places but for me I want to understand how it works / make it work myself rather than just being given a widget)

[IanB]

I have some LED bulbs that take anything from 12 V to 24 V, AC or DC.

When I tested one of them, I found that it had an internal regulator behaving something like an SMPS, so that it would draw constant power at any supply voltage and maintain more or less constant light output. So if I doubled the voltage from 12 V to 24 V, the current drawn was reduced by half.

I would guess, if the bulb failed, that the current would change from the nominal value. Most likely it would decrease. If it increased then some internal component would probably overheat and fail, and then the current would decrease after that. Or the electronics might fail short circuit and blow a supply fuse. There are all sorts of possibilities here.

The reliable solution I suppose, in a safety critical application, is to continue using the traditional incandescent bulb, or to completely replace the signal lamp fixture with a modern replacement designed to meet regulatory standards for performance and reliability?

I believe there is some similar issue with traffic lights on the highway. You might research how they handle the same question?

[Tomorokoshi]

I'm lit, therefore i amps?

Sorry!

[SiliconWizard]

Note that the current sensing method may not work if the failure mode of the LED is a short-circuit (never seen it myself, but I guess that could happpen) and the LED is driven with a constant current source. Obvious, but hey, what do ya know.

If it's just powered with a voltage source and a limiting resistor, it should work fine though.

[Marco]

Having an LDR and circuitry (or similar thing to sense light) fitted in such a way that external light cannot influence is only way I can think of to do this unless I am missing something obvious / being daft

External light will always influence it, you just need to make sure that the threshold is high enough that only only the lamp can actuate the circuit. So either due to proximity or a lens the lamp needs to put significantly more light on the sensor than stray sunlight.

[Nusa]

For your application you're going to be using higher power LEDs mounted on a heatsink. So a thermal sensor on that heatsink would be an alternate way to sense if a light has failed. No thermal change when a LED is turned on = failed LED.

[NiHaoMike]

It's to modify railway signal lamps to use LEDs instead of filament bulbs. The bulbs, each has 2 filaments / 3 contacts, take over an amp at 12V and are very expensive things and don't last that long. The trouble here is, with a filament bulb we simply use a relay to determine if it is dead, switch to the second filament and raise an alarm. We need to replicate this for the LED - not to switch to a backup LED but to drop a relay. What is supposed to happen then is all the further signals go to danger, and that bit is easy. These signal lamps are somewhat ancient and will not take the 'real railway' drop-in replacement LED units and anyway we could never afford them.

(I am of course asking in other relevant places but for me I want to understand how it works / make it work myself rather than just being given a widget)

For that specific application, it might end up easier to design a custom board instead of trying to make it work with off the shelf modules. Maybe PWM the LEDs at a high frequency (several kHz) and detect that the LEDs are in fact pulsing at that rate?

Another solution could be to convert the lights to use cheap 12V automotive bulbs.

[thm_w]

For your application you're going to be using higher power LEDs mounted on a heatsink. So a thermal sensor on that heatsink would be an alternate way to sense if a light has failed. No thermal change when a LED is turned on = failed LED.

The idea is good but at that point the logic is too complex that you'd want a micro: waiting for heatup, taking into account ambient temperature, etc.

I think the current consumption is fine, especially if its a string of LEDs or something with a smart driver. OR if you want to get crazier, monitor current draw and voltage across the LED, and ensure they are both within appropriate windows. Could be done with a comparator or a micro.

[T3sl4co1l]

It's to modify railway signal lamps to use LEDs instead of filament bulbs. The bulbs, each has 2 filaments / 3 contacts, take over an amp at 12V and are very expensive things and don't last that long. The trouble here is, with a filament bulb we simply use a relay to determine if it is dead, switch to the second filament and raise an alarm. We need to replicate this for the LED - not to switch to a backup LED but to drop a relay. What is supposed to happen then is all the further signals go to danger, and that bit is easy. These signal lamps are somewhat ancient and will not take the 'real railway' drop-in replacement LED units and anyway we could never afford them.

Ah, now this is something to work with.

These would be outdoors, then?

It should be possible to arrange a photodiode (no one uses LDRs anymore, they're old, bad, and non-RoHS) to sense some of the optical output, close enough that the intensity of the LED chip(s) overwhelms almost any possible interference.  Interference would have to be fairly extreme I think; a white LED these days is typically about as luminous as the sun(!), so any interference would have to be much more intense: a very close welding arc perhaps; lightning strike; or highly focused light (think: Archimedes death ray).

As long as the diode senses light of sufficient intensity, turn on a MOSFET to switch a shunt resistor onto the circuit.  This replicates the incandescent's current draw.

At such intensities, the PD probably looks more like a small resistance than a diode.  Or use it photovoltaically, i.e. apply a load resistor and sense the voltage.  Maybe use two PDs in series so it can bias a BJT on (instead of using a MOSFET).  (Some Schmitt trigger action may still be desirable for clean switching, or the switching transistor can be heatsinked to handle the dissipation during indeterminate levels.)

Note that this solution necessarily negates the LED's efficiency; only reliability is affected.  (And maybe not all that much, if the resistor can't be placed somewhere away from the LEDs.)

This also assumes the LEDs aren't dimmed by variable AC or DC.  (PWM is OK.)

If changing the sense relays is an option, then I would simply suggest using the LEDs as-is.

Also mind that transients can fry LEDs easily; a TVS diode may be handy as well.  Just a little SMAJxx something should be fine.  How important this is, depends on where we're talking.  If this is like trackside infrastructure, I would anticipate lightning induced transients, and maybe use a bigger TVS.  If this is on rolling stock, ESD or smaller surges are a more likely threat, which may not need much if any additional hardware; but beware if alternator load dump is required by the customer (in which case a voltage limiter circuit is probably adequate).

Tim

[james_s]

I'm usually all in favor of LEDs, but in this case it seems like continuing to use a filament lamp is probably the most ideal solution. If you do decide to go with LED then add a photodiode to monitor the actual output. Since railway signals (that I'm familiar with) lack a reflector in order to prevent reflected light from making it appear lit, you should be able to mount a photodiode off to the side where it can see the LED but will not be exposed to direct light from elsewhere. I don't think anything will reach the reliability of failure detection you can get with incandescent, but the LED should be much less likely to fail.

[TimFox]

Before retirement, I used commercial x-ray generators that had a large red 24VDC indicator light to indicate "X Rays ON".
The control system was interlocked to prevent x-ray power unless that bulb drew current.

[Brumby]

..... take over an amp at 12V .....

These would be outdoors, then?

If we are looking at an incandescent running in the vicinity of 12W or so, I doubt they would be for outside use.

Seems more like a mission critical safety indication situation.  As such, any solution would require rigorous testing and validation with failsafes.

[james_s]

The OP said they're for a railway signal, so I'm taking that as meaning outdoors, unless they mean an indicator on a control panel somewhere. Railway signals are traditionally relatively low wattage incandescent lamps with very directional optics. The train is on a track so you know right where it's going to be and can aim the signal accordingly.

[eti]

They tend to emit light - the clue is in their very name. 😁