Class 88 headlights

[50039]

Simple question - I was running through my photos this morning and I have a series of shots taken of 88010... but I noticed that for the pair of low-down lights, in some shots both are on, in another one neither is on, and on others, only one is one... all shots taken in a couple of seconds as it passed through... so question is - are they flashing lights, or do they flicker faster than the eye can see?
For reference, the light up top is on in all of them...

 

[Domh245]

Flickering faster than the eye can see, but caught by the digital camera.

 

[driver9000]

That's the LED refresh rate being picked up by your camera.

 

[50039]

Thanks - glad I’m not going mad! Haven’t noticed it on anything else - o
presumably the 68s are the same?
Is it a different type of light they use / different refresh rate?

 

[hexagon789]

Thanks - glad I’m not going mad! Haven’t noticed it on anything else - o
presumably the 68s are the same?
Is it a different type of light they use / different refresh rate?

A number of trains with LED lights but the refresh rate on 68/88s seems to be such that it's more noticeable than elsewhere

 

[fgwrich]

Thanks - glad I’m not going mad! Haven’t noticed it on anything else - o
presumably the 68s are the same?
Is it a different type of light they use / different refresh rate?

The new "bugeye" LED Lights fitted to Great Western's Turbo fleet (and some 66s) also do the same. It's less noticeable on standard cameras (like mobile phones) but very noticeable on higher end and CCTV cameras.

 

[rebmcr]

LED road traffic lights and LED pedestrian crossings do it too.

 

[superjohn]

Does this allow the lights to illuminate at variable brightness? For example, the same lamp used for day and night headlights.

 

[TRAX]

Does this allow the lights to illuminate at variable brightness? For example, the same lamp used for day and night headlights.

Yes I believe these locos have that feature, although it doesn’t have anything to do with the effect LEDs have on cameras.

 

[pompeyfan]

The same thing happens when people try to photograph LED destination blinds. I’ve never understood the science behind LEDs flicking.

 

[hexagon789]

The same thing happens when people try to photograph LED destination blinds. I’ve never understood the science behind LEDs flicking.

Something to do with the pulse rate of the LEDs compared to the frame rate of a camera when you take a picture.

 

[route:oxford]

The same thing happens when people try to photograph LED destination blinds. I’ve never understood the science behind LEDs flicking.

Technically a standard domestic filament bulb flickers 50 times a second too. It's just that the filament takes longer than 1/50th of a second to cool that you never notice.

 

[John Webb]

Does this allow the lights to illuminate at variable brightness? For example, the same lamp used for day and night headlights.

The brightness of LED lights can be varied by using 'Pulse-width modulation'. An LED at full brightness has a constant voltage across it. By providing pulses of full voltage at a sufficiently high frequency, the average light output from the LED can be reduced as seen by the human eyes due to the 'persistence of vision' that the human eye has. This is done by varying the length of the On time to the Off time.

The same thing happens when people try to photograph LED destination blinds. I’ve never understood the science behind LEDs flicking.

Destination blinds and platform displays contain large numbers of LEDs. To connect each one to a separate circuit would lead to a huge number of connections. The LEDs are therefore arranged in a grid of horizontal and vertical connections. Driver circuitry converts the required information into which LEDs should be on and off, and rapidly scans the vertical and horizontal connections and turns the individual LEDs on as required by energising the appropriate horizontal and vertical connection. By doing this at high speed the human eye again perceives a steady light but cameras, both still and video, see the scene without 'persistence of vision' and will catch the LEDs off when the human eye thinks they're on!
Hope this helps,

 

[malc-c]

Yes, pulse width modulation and multiplexing at work, captured by the camera's shutter speed. Not my video, but this is a good example

 

[jh64]

A similar camera 'trick' can be seen with helicopter blades, if the framerate and the rotor blade rotation manage to line up right:

 

[xotGD]

Yes, pulse width modulation and multiplexing at work, captured by the camera's shutter speed. Not my video, but this is a good example

I've taken videos of 68s with the same blinking light effects. Physics is interesting!

 

[furnessvale]

That's why factory lighting runs (or did do when I was a lad) off different phases so that rotating machinery doesn't look stationary at certain revolutions.

 

[endofthetunnel]

Refresh rates are still taken into account with lighting regarding machinery etc yep, it’s in the regulations.

And it’s why so much planning goes into TV and film lighting/cameras to make sure the various refresh rates of cameras and all lights are synced or so fast that it’s not noticeable. Super slow mo replays in sports often really look flickery because of this effect.

 

[50039]

Thanks folks - really interesting stuff... never noticed before, not on any of the 66s I’ve photographed..

 

[Peter Sarf]

If the rotation is at the right speed in relation to the frame rate a wheel or rotor blades can even appear to be going backwards !. Can see this with the naked eye if the lighting is flickering at the right speed I believe.

What did puzzle me is that the OP said the top light appeared to always be on. Is that then an incandescent bulb maybe ?.

 

[Domh245]

What did puzzle me is that the OP said the top light appeared to always be on. Is that then an incandescent bulb maybe ?.

It looks like it's also LED, but If I remember rightly because it is a marker lamp (as opposed to a head lamp) it only has one brightness, so doesn't need to be driven by PWM and can instead just be normal DC, avoiding the flicker