Good day all,

I am in the process of trying to get a monostable LED pulsing circuit to work as part of my masters dissertation in mechanical engineering (that's right, mechanical, so I am a bit out of my depth here). Thus far I have managed to produce the required pulse duration down to a mere 500 nanoseconds however I am not getting anywhere enough illumination from the LED. All I require is around 1 microsecond of the brightest possible illumination, no limits!
The circuit is to be used in conjunction with some high speed pressure transducers to capture shock waves reflecting off various object in a shock tube. The system to produce the TTL signal to trigger the circuit is already working.
Without going into too much detail, the light is emitted from the LED, passes through some colour masks, bounces off some cleverly placed mirrors and is captured with a digital SLR.
The optics just in front of the camera allows all the deflected light rays (due to the change in density of the air in the shockwave region) to pass and blocks all the un-deflected rays.

You can see my video in the youtube attachments using the LED system in constant mode.

http://www.youtube.com/watch?v=9WxqUR84k_Q&feature=player_embedded

and

http://www.youtube.com/watch?v=L9Rm-YNxQ_g&feature=player_embedded

I cant see how to attach images to posts so Ive uploaded the circuit diagram to youtube:

http://www.youtube.com/watch?v=BU5OJkvf9KE&feature=player_embedded

IC2 is a 5v regulator, IC1 is the 74121 monostable vibrator CMOS chip, Q1 is the IRL520 MOSFET and LED2 is the high power LED. I have been using some Cree MCE 180lm multicolour LED's to get as broad and even visible spectrum as possible. I require all visible light wavelengths and not just a monochromatic light due to the color requirements of the optics system. R4 is a 1 ohm resistor to monitor current draw and S1 is just a switch to turn the LED on for optical alignment purposes. I need to fiddle with this side of the circuit still.

I have a feeling the MOSFET is not turning on fully (gate-source voltage not high enough??) and thus the LED cannot draw enough current?? How to check this??

Any comments would be hugely appreciated on anything related to this circuit

 

An RGB LED may not be the best choice if you want white light since it has gaps in it's spectral band between the three peaks. A white LED does a better job filling in the gaps.

For such a short burst, you should be driving the LED to it's peak current rating instead of limiting to it's continuous rating. Of course you must limit using the continuous rating for your alignment function.

 

I cant see how to attach images to posts so Ive uploaded the circuit diagram to youtube:

Just click "Manage Attachments" below the "Submit Reply" button, in the additional options panel.

 

Putting a scope on the gate of the IRL520 should provide some answers. Consider providing 10V, and up to 300-500ma, to the gate using a 2N7000 as a switch.

 

Putting a scope on the gate of the IRL520 should provide some answers. Consider providing 10V, and up to 300-500ma, to the gate using a 2N7000 as a switch.

How would this differ from powering the FET directly as both the FET and the 2N7000 have similar gate threshold voltages?

 

How would this differ from powering the FET directly as both the FET and the 2N7000 have similar gate threshold voltages?

16ma ttl gate drive may not overcome the gate capacitance in 1us. So checking the mosfet gate with a scope would put that angle to rest one way or the other.

 

Human vision sees light pulses normally when the duration of the pulses is 30ms or more.
But when the duration of the pulses is less than 30ms then the light appears to be dimmed.

 

How would this differ from powering the FET directly as both the FET and the 2N7000 have similar gate threshold voltages?

Don't look at the threshold voltage that is when the Mosfet is barely turned on (it is almost turned off) with a current of only 0.25mA.
Instead, look at the GS voltage when the on-resistance is guaranteed to be a stated low resistance.