Can LED strobe be as fast as xenon?

Can you define pulse width min? or rep. rate? No problem except brightness is proportion to duty factor which affects smear. But I think no problem with 3600 RPM 1% d.f. or a toy at 500 RPM.

you can't exceed maximum current unless you risk fusing the tiny gold strand wire but you can always oversize the LED power in parallel.

All diodes have capacitance and resistance the effect is an RC=T time constant but it is a lot fast than you can spin art. The issue is Xenon dumps all its power in flash but LED's cannot increase the average power to the same as steady current. So if 1% or 3.6 degrees is OK that's 1% brightness.

3600 RPM = 60 Hz divide by 1% is a pulse width equal to a 1 cycle at 6000 Hz every rotation. You can do that duration with a variable "1-shot" and adjust phase or angle with another variable.

All you need to is define all these variables and maybe Grok or GPT can spit out an answer

may the force be with you.

"Normal" LEDs should work fine, but any that use a phosphor to change the light output spectrum (like most white ones) may be visibly slow cutting off in a high speed application. (I have no idea how slow though)

An RGB LED with separate terminals for each colour could probably be used to give white, using appropriate resistors for each element to balance the colour levels? Or separate high output R,G & B ones.

Most LEDs can work fine on somewhat higher currents in low duty cycle applications, such as multiplexed displays - or this. Check the manufacturers data sheet for the pulse ratings.

Clearly phosphor type used significant decay times. Couple that with human eye
persistence........




Regards, Dana.

SvenBaidenmann said:

I need to strobe a spin art machine with Hall effect sensor. I have given up trying use xenon tube. I want freeze the image in the exact same position so I must trigger with the Hall effect. I tried a made Led strobe but it the led was not fast enough. Can leds equal the xenon effect? I need speed not brightness. Help me Obewon Kanobe you’re my only hope

Click to expand...

You need to say how fast it needs to flash, and how fast the object is spinning - also, why have you failed using a xenon tube?.

SvenBaidenmann said:

I need to strobe a spin art machine with Hall effect sensor. I have given up trying use xenon tube. I want freeze the image in the exact same position so I must trigger with the Hall effect. I tried a made Led strobe but it the led was not fast enough. Can leds equal the xenon effect? I need speed not brightness. Help me Obewon Kanobe you’re my only hope

Click to expand...

YES!
Most Xenon strobe lights are limited to the 100 to 200 flashes per second and each flash is 10 to 20 uSeconds.

A white LED of reasonable quality (Osram, Samsung, Seoul Semi, LumiLed, ...) use phosphors that are fluorescent (not phosphorescent) and these have, generally, sub-microsecond decay time (typically nanoseconds).

This means your flash rate can be in the 10-100kHz - way faster than any spin art.


Ideally, get a very bright (think automotive headlight) Osram flat black or other multiwatt array LED.

Note that your claim of the LED is not fast enough is likely, "my LED flashing circuit is not suitable for a strobe". You cannot use a simple 555 flashing circuit with 50% duty cycle - you need to adjust it to be on for some very short time and off for the rest of your rotation to create the stop motion.

Cheers.

PS - Let me know if you need something. I love to support the arts and/or kids projects. DM me on the Envelope icon in the top menu bar.

Automotive timing lights converted to LED's years ago and they work fine.

As mentioned you need to specify more parameters like flash rate, pulse length, target object rotation speed to see if you application is within the range of using LED's.

IR LED in TV remotes run 30 to 40khz. I have run those to 60khz at 25% duty cycle.

Tony Stewart said:

you can't exceed maximum current unless you risk fusing the tiny gold strand wire

Click to expand...

I have run LEDs way above their maximum current, for a short duty cycle. Some LEDs have that type of data in the sheet but most do not. Lazer or IR might have the data. Most LEDs for indicators will not because almost no one uses them that way.

Here is some data for a 100mA LED. It turns on or off in 15nSeconds. TSAL6200 Vishay


More data: 100mA all day long, 200mA at 50% where Ton=100uS Toff=100uS and 1.5A for one pulse at 100uS.


Another page: They are showing light out verses current for 1mA to 1A. (left side)
Right side: shows time and current.

SvenBaidenmann said:

I need to strobe a spin art machine with Hall effect sensor. I have given up trying use xenon tube. I want freeze the image in the exact same position so I must trigger with the Hall effect. I tried a made Led strobe but it the led was not fast enough. Can leds equal the xenon effect? I need speed not brightness. Help me Obewon Kanobe you’re my only hope

Click to expand...

Hi,

How bad do you need this? If you are able to pay you can get sub nanosecond pulse times. Check out Picoquant but be ready for some really steep prices

You can get LEDs to pulse very fast but you need to use a special driver circuit that is able to shape the wave that gets to the LED. The current is very high, and so the shape has to be right or the LED burns out. Unfortunately, I did not get involved in this technology because I did not need it at the time (around 10 years ago).

Laser diodes can be pulsed very fast too. You could check into that also.

For what it is worth, I was able to pulse a white LED well enough to act as a guitar tuner. When you pluck the string you can strobe it so you can get it to look like it is not moving at all if the frequency of the pulse is right. That's easily 1000Hz. That's done with a regular microcontroller like from Microchip but an Arduino chip would do just as well.

MrAl said:

How bad do you need this? If you are able to pay you can get sub nanosecond pulse times.

Click to expand...

It's spin art! With some type of persistence of vision strobe. How fast do you think an artist-built rotating device needs to spin?

Thank you Ron for the correction. I apologize for generalization.

IR LEDs are an exception.

I doubt that any LED specs limits are based on typical user habits, but rather tested to failure with conservative thresholds and optimize LED aging performance (LM70).

IR will not be seen a visible shadow or blemish and thus have thicker wirebonds and fuse ratings.

There may be newer exceptions for SMT parts designed for high pulse mode.

For reliability reasons, exceeding the Absolute Maximum ratings will impact MTBF such as fusing, interface diffusion or gradual erosion.

This certainly made a challenge for multiplexing RED LED segmented displays in the early days and HP were the leader with the most digits per driver. An extreme example was my 1st computer at work, the HP9825 in the late 70's with 4 of 32 characters per driver with an 80 column buffer.

ronsimpson said:

IR LED in TV remotes run 30 to 40khz. I have run those to 60khz at 25% duty cycle.

I have run LEDs way above their maximum current, for a short duty cycle. Some LEDs have that type of data in the sheet but most do not. Lazer or IR might have the data. Most LEDs for indicators will not because almost no one uses them that way.

Here is some data for a 100mA LED. It turns on or off in 15nSeconds. TSAL6200 Vishay


More data: 100mA all day long, 200mA at 50% where Ton=100uS Toff=100uS and 1.5A for one pulse at 100uS.


Another page: They are showing light out verses current for 1mA to 1A. (left side)
Right side: shows time and current.

Click to expand...

Assuming 600 RPM is the fast spin rate, f = 10 Hz or a period of 100 ms.

The pulse duration must be synchronous to the start of the Hall sensor and stopped by a counter reset on each rev. with the clock PLL operating and N x f. which is easily done in several ways

The decay time on diodes is proportional to junction capacitance which is proportional to the LED power rating. So I recommend 5 mm 30 deg LEDs with an Iv rating > 16,000 mcd minimum. I have lots of these and they are blinding at 1m and irrating at 10 m with rated current of 20 mA, so you can operate these at 0.1% or 1/1000 for excellent visible in dim light effects. This always causes blur with motion, this duty cycle defines the % of the circumference that is blurred. They can all be driven in series at 3V/LED with a current limiter in pulse mode.

You just need to define how many and how fast.

This can also be slowly rotated or moved by a small frequency offset using an offset PLL for slow motion effects like wagon wheels in cinema.