LED turn signals with ramp up period

[Ls400]

https://jalopnik.com/eagle-eyed-turn-signal-enthusiasts-note-a-fascinating-d-1845927919

Summary: Mazda LED turn signals are mimic incandescent turn signal bulbs in that both take a bit to ramp up to full intensity.

I'm not familiar with the research on LED turn signals, so I turn here and ask: is what Mazda's doing a good thing? A bad thing? Not enough data to conclusively comment?

 

[Alaric Darconville]

I'm not familiar with the research on LED turn signals, so I turn here and ask: is what Mazda's doing a good thing? A bad thing? Not enough data to conclusively comment?

While it's a "neat" thing (and something I'd wished they'd invented for buses, with their gigantic turn signals very nearly level with the roof of some of my cars), it may not be the best thing. That insta-on/insta-off of turn signals is rather jarring, but that insta-on also offers critical milliseconds of initial warning (which is perhaps not as critical as in stop lamp applications). The specialized circuit may also be another point of failure. While there's not yet data (as far as I'm aware) that these shorter rise times in LED turn signals presents a real safety advantage, it doesn't make a lot of sense to deliberately remove that assumed advantage.

For the stop lamps, they really ought to leave it alone, for that's where the extremely short rise time is an advantage over a filament bulb. At 75mph, 10ms is another 1.1 feet of missed warning time.

 

[-Virgil-]

Summary: Mazda LED turn signals are mimic incandescent turn signal bulbs in that both take a bit to ramp up to full intensity.

That's not what I'm seeing in the YouTube video. The video shows that the signals immediately go to full intensity, stay there for a time, then fade gradually to off. Instant rise time, gradual drop time. Same as I've noticed a few times now when seeing these cars on the road.

In the absence of data we can't say for sure what (if any) safety effect this has, but I lean toward calling it a good thing. It does not appear Mazda has significantly (if at all) shortened the full-intensity duration of the turn signal's illumination. Instead, it seems they have added a "tail", if you will, of gradually decreasing intensity during what would usually be the signal's unlit duration. This has the effect of increasing the odds that someone glancing briefly at the signal will see some light coming from it. Personally, I think this is a good thing, sort of a best-of-both-worlds deal: the attention-grabbing benefit of the instant-on, and the added likelihood of seeing light at any given glance.

I haven't noticed whether they've put fade-to-off on the brake lights. I wouldn't object if they did.

 

[lightfooted]

From the video I wouldn't say that they are mimicking incandescents, rather they are doing their own thing. I like that manufacturers are finally beginning to do things with LED lighting that was never really possible or easily done with old incans.

 

[PhotonWrangler]

I like this instant on/ramp off mode. It still has the advantage of the millisecond-faster on period but has a graceful decay that gives it a throbbing or heartbeat effect. Very distinctive.

 

[-Virgil-]

I like this instant on/ramp off mode. It still has the advantage of the millisecond-faster on period

It's more like 200 milliseconds faster, but other than that, yeah.

 

[Alaric Darconville]

It still has the advantage of the millisecond-faster on period

About 200ms faster. For stop lamps at 75mph, that's ~22 feet of additional warning for the driver behind them. A little extra warning time for a turn signal is also an advantage.

 

[-Virgil-]

Yeah, but the weird thing is this faster/earlier warning from LED stop lamps might or might not not translate to a significant safety benefit in terms of crash avoidance. Look at this research.

 

[Alaric Darconville]

Yeah, but the weird thing is this faster/earlier warning from LED stop lamps might or might not not translate to a significant safety benefit in terms of crash avoidance.

I'm better at reading the abstract than the rest of it, but it does seem that the model changes being ceremonially unclean*​ not being clean may have muddied it up a LOT.


*Darnit, strikethrough is not supported​

 

[-Virgil-]

Yes, that's right -- the allowable intensity, on axis, for a single-compartment stop light is 80 to 300 candela. The allowable illuminated area is 50 sq cm or more. So if we're comparing (say) a 51 sq cm, 85-candela LED stop light to a 200 sq cm, 250-candela incandescent-bulb stop light…or vice-versa on the sizes and intensities versus technology (small/dim incandescent versus large/bright LED) that's a lot of confounding factors to get in the way of trying to figure out whether LED stop lights work better at preventing crashes. And that's without even going into other factors that could affect stop light conspicuity (stop light mounting height, aspect ratio of height/width, visibility angle range beyond the minimum requirements, etc).

That being said, if there were a big crash-prevention benefit to LEDs' instant rise to full intensity, I think it would probably show up despite the "noise" from those other factors. The outcome of the linked study looks to me like the benefit, if it exists, is a small one.

 

[Alaric Darconville]

That being said, if there were a big crash-prevention benefit to LEDs' instant rise to full intensity, I think it would probably show up even despite those other factors. The outcome of the linked study looks to me like the benefit, if it exists, is a small one.

I would say that generally speaking, most rear-endings occur in the slower surface street traffic, where that 200ms rise time advantage an LED stop lamp has does not translate to very many feet of extra warning for the following driver. Sure, at 75mph you're going 110 feet per second, so 200ms gives you 22 more feet to react in. At 35 mph, you're going 51fps and 200ms becomes only 10 feet of extra warning. Sure, you're not going as fast but you're also probably within 10 feet of the car in front of you pretty often, depending on the conditions.

As speeds decline, the effect of the 200ms advantage will be lessened, essentially.

 

[Ls400]

Well, leave it to Jalopnik to completely mischaracterize the turn signals and turn that mischaracterization into an entire article! Mea culpa!

 

[-Virgil-]

Jalopnik seems hit/miss to me. Sometimes they have super investigative articles with cross-checked legitimate sources and good factual basis, etc. Sometimes their stuff is more like "Who cares, there's a DEADLINE, just fill the space with text and push it live NOW!" Sometimes it's like they're talking in their sleep, or just making it up as they go along. And sometimes they post stuff that looks like "Look how smart I am, I'm way smarter than you"; one or two of their writers seem to have a hard time setting ego aside, or maybe that's a deliberate choice of tone? I don't think it matters a lot to whoever is in charge over there. Whichever mode they're in, their content pulls traffic and makes people argue in the comments, which boosts ad revenues, and I guess that's what matters these days.

As far as this specific Jalopnik article, my guess is that they fell victim to Youtube compression effects. I've seen that a lot in videos trying to show turn signal operation...it can look like an unsteady rhythm and the ons and offs can be muddied unless it's a super high resolution video and you're on a very good/fast connection and there aren't any network clog-ups between you and Youtube.

 

[Dave_H]

While I appreciate the argument for maximum warning at high speeds, instantaneous on-time of a
bright LED light at short distances can be very distracting. I thought of the ramp-up/ramp down (PWM)
idea some time ago, considered prototyping a simple control circuit (using a PIC) but didn't get around
to it (or a round tuit .

In any case the added cost could be relatively low for some lights, but proportionally higher on cheaper lights.
These certainly would have to be reliable. The last thing you need is an indicator which gets stuck on or off,
or behaves erratically.

Dave

 

[PhotonWrangler]

I suspect that some newer cars control their turn/brake lights in (software) code. If this is the case, it would only take a few more lines of code to PWM the LEDs to get this effect.

 

[-Virgil-]

While I appreciate the argument for maximum warning at high speeds, instantaneous on-time of a
bright LED light at short distances can be very distracting.

The line between "conspicuous" and "distracting" is a blurry one. It's not something that can be meaningfully judged subjectively except in very extreme cases.

 

[Dave_H]

One could argue that a large part of this discussion is subjective. All I know is, for some vehicles I
am behind with lights like this around eye level, it is negatively distracting at times; including
stopped or low speed where the distance warning advantage seems low.

Increasingly I see non-emergency service vehicle (including snow removal in Ontario using blue)
lights having on/off with rapid strobing. Not sure how this applies to different vehicle classes
such as passenger. Jury is still out on this one for me. Colour/brightness matter of course.

I suppose software PWM control might be possible with no change to lights themselves but being a
SMOS (small matter of software) isn't always trivial in the end, given the safety and reliability required
for just about everything in the automotive environment. Just reading about some of this particularly for
autonomous vehicles can be boggling.


Dave

 

[-Virgil-]

One could argue that a large part of this discussion is subjective.

One can argue anything that might come to mind!

All I know is, for some vehicles I am behind with lights like this around eye level, it is negatively distracting at times; including stopped or low speed where the distance warning advantage seems low

Maybe so. Those sound like circumstances where there wouldn't be much/any real negative consequences of a turn signal that draws your attention more than you might like, though.

Increasingly I see non-emergency service vehicle (including snow removal in Ontario using blue) lights having on/off with rapid strobing

This started many years ago when strobe tubes replaced rotating beacons. Those strobe tubes caused problems, but then LEDs came along and pretty much fixed them.

 

[Ofelas]

This made my morning

a graceful decay that gives it a throbbing or heartbeat effect.