White LED with Shortest Response Time

what sort of response time do you want?

There are 2 types of white LEDs: the RGB type that contains a Red, Green and Blue die mounted together in a single package which produces a net white output, and there is the phosphor-type white LED that uses a yellow phosphor with a blue LED to produce a broad-spectrum white output.

In general, phosphors have an after-glow (fluorescence) which can cause a small but measurable decay time. Offhand I don't know of any use for white LEDs where this decay time is a problem; I've never seen it quantified. But nevertheless, the tri-color RGB is probably capable of a faster net response time since the LED's RGB dies will (or should) have a much faster rise and fall time, there being no after-glow.

Having said that, what is probably a lot more important than the LED is the circuit driving it. To get a really fast response time you'll need a circuit capable of extremely narrow pulses but also capable of supplying very high current.

There are dozens of LED manufacturers. Some of the more well-known manufacturers are Philips-Lumileds (marketed by Future Electronics), Cree, Nichia, and Osram-Sylvania. On their websites you can get specific info on LED types and on some of their website you can also get guidelines on driving them including suggested circuit diagrams.

Just to put a number on it, it's easy to push LED response down into the 40ns territory, but I experienced frustration trying to get them to turn off faster than 20ns. This was true even when strongly reversing the current within under 5ns.

Below 20ns it becomes a matter of finding the right LED. I've seen claims of results down to a few ns, but they didn't report what part they were using. I tried many parts and wavelengths, etc., without success. It's like the proverbial fish that got away.

But laser diodes, OTOH, are very fast. They're happy to respond in under 1ns, so it becomes a matter of the driving electronics and fast low-impedance wiring, etc. But it might be painful trying to create an apparent white light with multiple laser diodes of different wavelengths. They also naturally make beams rather than flooded light, greatly restricting their applicability for ordinary lighting.

White LEDs are often driven with dc-dc converters to control the current. The voltage step-up types, which can drive several 3V LEDs in series, are especially convenient. But most of these are fairly slow. One exception is TI's TPS61166 (link), which can turn the LED on or off in well under 1us.

TI's motivation for such a fast response time is to insure a linear light-level vs PWM duty cycle, e.g., down to the 1% region, even with high PWM frequencies like 40kHz. For example, 1% of a 40kHz 25us period is 0.25us, so they would benefit from a response time faster than that. Although their boost-converter runs at 1.2MHz, this is not fast enough for the desired switching rate, so the TPS61166 adds an additional series FET switch to drive the stack of LEDs.

You can add an external MOSFET to an ordinary boost converter, to obtain the same fast result as TI's part. It doesn't appear that distributors are stocking the TPS61166, so that may be the best approach.

I'm driving a Cree XM-L, and while I get decent responsiveness, I don't get anywhere near the brightness I desire; I'm driving it at 1800 mA in pulses of 100 microseconds; the initial response is fairly fast, but it takes a while to ramp up to its full brightness.

• 30000 microseconds is full brightness (about 500 lumens)
• 15000 microseconds is fairly close; above 350 lumens
• 7500 ms is still fairly bright
• 3250 is maybe 150 lumens
• 1625 is pretty close to 100 lumens
• 812 is unusably dim

(lumens were eyeballed with 500 and 100 lumen reference points)

Let's say that the brightness at 1625 microseconds is the minimum usable; the problem is that the longest usable pulse (due to duration) is around 100 microseconds.

What driving techniques might allow me to significantly decrease the effective response time?

Well, most of us increase the LED current in inverse proportion to the ON time. This helps to maintain the average light level. Most LEDs can easily handle a factor of 10 higher current. Some can handle much more, to 100x or higher. Whether the LED can handle this or not is a matter first of its thermal mass (assuming your ON time is short enough), and second of the internal construction of the LED.

Now, were you saying that you apply the high current, and then wait around for 100s of microseconds for the LED to increase its light output? Whoa! What I've seen is the light output falling with time during the pulse, as the die heats up.

I've got limited options; I'm the primary developer of firmwares for the hexbright (an arduino-compatible flashlight). There is a buck-boost converter for a high/low mode, but it's not designed to produce loads that could actually cook the LED (well, temperature regulation is an issue, but on the order of minutes, not fractions of a second).

It looks like I'd need to significantly modify the electronics in order to improve the characteristics for this application.

Thanks for your help :).

Hah, I was just answering an email question and discussed the controller issue this morning, addressing tricks that switching-supply designers have to use.

Question, if the current-regulating controller runs its PWM at say 0.5 or 1MHz, how are they going to get sub-microsecond on/off response times? Answer, run the PWM to get the right voltage for the LED, at the desired current, and store this voltage on a capacitor, and then use a second FET to disconnect this capacitor from the LED. That's what the TPS61166 (link) I wrote about does. Another part using that trick, but at the 20A level, is LTC's LT3743.

They also have two control levels and two output switches, so you can save two different current levels. And they turn off the current-control feedback loop while the LED is disconnected. Nice.

So, the hardware design can be critical in such cases.