I'm new here, so this is also my little intro. I don't know much about circuits, and until recently, my only experience with circuits was ripping apart old RC cars and amping the voltage to make them faster and building new ones from old pieces. I started doing some reading on the site's textbook and actually started doing somethings that are practical with circuits.

What I would like to do, is create a high speed LED flash that I can use for some high speed photography. I'm shooting for a sub 20 μs flash. At the moment I'm merely trying to make a proof of concept circuit with 1 LED. Eventually the problem is going to be not enough light, but for now I want the timing right. Once I get the timing circuit I can add more more power and LEDs to the circuit.

Please bare with me, as I am just beginning to learn most of this.

The .pdf below is the circuit that I have come up with for my purposes so far. My circuit is derived from this example I saw, with R2 and C1 as the time controlling components for the 555 IC.

I have a very rudimentary understanding of many of the components I used, which should become obvious as I explain.

The 3.5mm jack supplies a 5v current from the middle of the jack when I want to activate the circuit.

The LEDs I plan to use at the moment require 350mA, hence the transistor that controls the power supply to the LED.

Please ask as many questions about my circuit, I made it fairly quick and I feel like there are many mistakes, so point them out if you see them.

The question I really need answered is will this work and if not how can I make it.

 

http://www.vwlowen.co.uk/java/timer.htm
use the above link to calculate the rc values, make sure you use a 7555 not a NE555

 

This is incorrect. A 7555 is a CMOS IC, and has radically reduced drive characteristics. If the OP wants 350ma for the LED then it gets even worse, since it is unlikely the 7555 can drive a BJT enough to drive the LED.

At 1µs, which is what you specified I doubt it will be usable for high speed photography. 1 millionth of a second is a very short period of time.

Having said that I know nothing about photography in general, though I do take a lot of pictures using my digital camera for this forum. What I would suggest is to double check this number (1µs) to see if it is what you really want.

I have a series of 555 articles I've written for the text book whose links you see above. I also have a large collection of 555 circuits. I can design what you need, if required, and am willing to do so.

First step is you need to define your power supply voltage. The higher it is the easier your project gets, and the fewer parts it needs. I would recomment 12V, but it is your call.

My Cookbook

The 555 Projects

By the way, Welcome to AAC to both of you!

 

This is incorrect. A 7555 is a CMOS IC, and has radically reduced drive characteristics. If the OP wants 350ma for the LED then it gets even worse, since it is unlikely the 7555 can drive a BJT enough to drive the LED.

At 1µs, which is what you specified I doubt it will be usable for high speed photography. 1 millionth of a second is a very short period of time.

Having said that I know nothing about photography in general, though I do take a lot of pictures using my digital camera for this forum. What I would suggest is to double check this number (1µs) to see if it is what you really want.

I have a series of 555 articles I've written for the text book whose links you see above. I also have a large collection of 555 circuits. I can design what you need, if required, and am willing to do so.

First step is you need to define your power supply voltage. The higher it is the easier your project gets, and the fewer parts it needs. I would recomment 12V, but it is your call.

My Cookbook

The 555 Projects

By the way, Welcome to AAC to both of you!

Trust me, the 1 µs is what I need to do high speed photography. If possible I would like it to be even faster, but for now I'll settle for that. Basically how it works, since the shutter in a camera is insanely slow compared to light, I much use a fast light and a dark room to expose extremely fast objects, such as bullets.

I obviously misunderstand the concept of a transistor. I thought that by adding any amount of current to the base, it would allow electrons to flow from the collector to the emitter. My thought had been that, since the IC can only allow a current of 200mA, I would use it to control the larger 9V battery and that would provide the necessary amperage to the LED through the transistor.

I did label my battery as a 9V in my diagram if you even looked for it.

Is there a problem with my choice of IC? Or is Jazz wrong in his statement? If he is right, which 555 should I use?

Is it necessary for the timing capacitor to be electrolytic? Currently it is not because I can't find a .01 microfarad electrolytic capacitor. I can use a .1 electrolytic and a 10k ohm resistor if I need the polarity.

 

Transistors have a set amount of gain, and when you want to use a BJT as a switch (as in this case) it becomes 35ma to output 350ma. A CMOS IC can not do this at low voltage, while a conventional 555 can. I suspect if you used 12V your concept would work just fine.

If you provide a schematic we'll (the whole forum) will take a gander at it, and if you need a schematic I will be glad to draw one up.

In general electrolytics are inferior caps, they are meant for low frequencies and DC. You need either a ceramic or similar type cap.

 

From tolerances of resistors and capacitors, you may not always get a flash every microsecond.

Something else to consider would be a counter clocked by a crystal. If you start with a 16 Mhz crystal, and take the output of the counter at divide by 16, it would be a more stable microsecond pulse train.

 

I had a brain fart and post the wrong schematic. I think he wants a monostable.

555 Monostable

 

I had a brain fart and post the wrong schematic. I think he wants a monostable.

555 Monostable

Indeed. That is why my example circuit is a monostable. Also, there is a .pdf of my circuit that has been attached to the thread since I made it. That is the schematic I've been working on and the one I referred to in the original post.

Here is my circuit again.

 

If you get a chance reread my article, you probably need a signal conditioner. Something that will reduce the input pulse to under 1µs. To test if it works use a longer output pulse to something you can see.

The other thing is you may want to use a color like blue.

A white LED uses a blue LED with a phosphor coating. The phosphor is similar to what is used in a fluorescent light. I suspect there will be a delay between the blue LED underneath lighting and the phosphor.

 

If you get a chance reread my article, you probably need a signal conditioner. Something that will reduce the input pulse to under 1µs. To test if it works use a longer output pulse to something you can see.

The other thing is you may want to use a color like blue.

A white LED uses a blue LED with a phosphor coating. The phosphor is similar to what is used in a fluorescent light. I suspect there will be a delay between the blue LED underneath lighting and the phosphor.

Won't the timing capacitor and the resistor work to make it a 1 µs pulse?

I'm aware of the phosphor in white LEDs, I plan on using a green LEDs at the moment.

Any suggestions on what would be the best LED to use? I have no idea and right now I plan on using these. However, I would love a better alternative for my cash.

http://www.mouser.com/Search/Produc...virtualkey62510000virtualkey720-LVW5AMJYKY25Z

 

Read the 555 monostable article I wrote, it explain it in detail. A 555 will hang in the high state while the input is low, way past the time it is designed for.

 

I'm having a bit of difficulty understanding the signal conditioner. I've attempted to add it to my circuit, but I still don't understand why it works and I don't know if that is how I'm supposed to use it.

 

I would suggest you run the experiment I have in the article. It would be much clearer after that.

You need a resistor between pin 2 and Vcc in this schematic.

Your current schematic only has 21ma for the LED, I assume it is a test schematic.

 

I'm having a bit of difficulty understanding the signal conditioner. I've attempted to add it to my circuit, but I still don't understand why it works and I don't know if that is how I'm supposed to use it.

I would like to make a couple of comments about this schematic:-

1. The transistor base is connected directly to the 555 output. Depending on the exact IC type, this may provide too much base current for the transistor, and/or the 555 may be overloaded. Normally a series resistor is added to setb Ib = Ic/10, or thereabouts.
2. Conventionally, this kind of circuit is drawn with +Vcc at the top, common at the bottom. This is by no means necessary, but easier to read for those used to this way of drawing things.

 

Well, since I apparently don't understand this even more than I thought, could anyone be so kind as to make me a circuit diagram that would accomplish what I am after with my design?

 

Here is a circuit that I built it on a breadboard a week or so ago and it worked flawlessly, however I had C1 sized for much longer pulses. The 1pF value should produce 1 μS pulses. Try it and see if it works as is.

If so, you will need to add a circuit to drive the LED's that you plan to use. In addition, you will need to invert the trigger signal that you showed on your schematic because the 555 requires a logic low to trigger.

Good luck.

 

Couple of questions:

1) How will you sync the 555 to the camera? Is this video where a stream of 1uS pulses would be used?
2) Do you have a DMM with frequency counter built in that will go to 1Mhz or a cheap osilloscpe that will go to 100Mhz? Without verifying the output, a human eye won't be able to detect if the pulse is 1/1,000,000 seconds or 1/100,000 seconds long. IOW, maybe build something to verify the output of this to be sure it is what you want.

Using the crystal/counter method, you could make the time on of the LED variable from 1uS to 20uS by combining outputs of the counter.

 

Couple of questions:

1) How will you sync the 555 to the camera? Is this video where a stream of 1uS pulses would be used?
2) Do you have a DMM with frequency counter built in that will go to 1Mhz or a cheap osilloscpe that will go to 100Mhz? Without verifying the output, a human eye won't be able to detect if the pulse is 1/1,000,000 seconds or 1/100,000 seconds long. IOW, maybe build something to verify the output of this to be sure it is what you want.

Using the crystal/counter method, you could make the time on of the LED variable from 1uS to 20uS by combining outputs of the counter.

I don't have an oscilloscope sadly. I plan to use this to photograph things at high speeds. If they are blurred, I will know that it is not working properly. I also may be able to find a oscilloscope at my school if I'm lucky.

Is there any other way to test the output other than an oscilloscope?

No, this is not video, it is a single still frame image. I plan to use a 3.5mm jack to connect the circuit and my camera. The 3.5mm jack will send a 5v signal to trigger the 555.

Here is a circuit that I built it on a breadboard a week or so ago and it worked flawlessly, however I had C1 sized for much longer pulses. The 1pF value should produce 1 μS pulses. Try it and see if it works as is.

If so, you will need to add a circuit to drive the LED's that you plan to use. In addition, you will need to invert the trigger signal that you showed on your schematic because the 555 requires a logic low to trigger.

Good luck.

By invert it, do you mean that I need to send a constant 5v and then drop the 5v current to trigger it? Or something else? Also, why 1 pF? Can't I just use my previous capacitor and resistor combination? I already have them laying around and wouldn't have buy them.

The NE555 is also only rated for 200mA output, how am I supposed to power a 350mA-2000mA LED?

 

A timing capacitor as low as 1pF is likely to be too much affected by parasitic capacitances in the IC and the connections to it. This 555 datasheet does not appear to recommend capacitors below 1nF (1000pF) - see pages 7 & 8. http://www.ti.com/lit/ds/symlink/lm555.pdf
Somewhat lower values could be tried (100pF?), but 1pF seems improbably low for this class of circuit.

The timing resistor will need to be altered to give the correct interval - see the nomograph Fig.3 in the datasheet - but note that operation below 10μs does not seem to be specified.

A driver transistor of some kind, either bipolar or MOSFET, will be needed for currents above 200mA.

 

Here is a circuit that I built it on a breadboard a week or so ago and it worked flawlessly, however I had C1 sized for much longer pulses. The 1pF value should produce 1 μS pulses. Try it and see if it works as is.

If so, you will need to add a circuit to drive the LED's that you plan to use. In addition, you will need to invert the trigger signal that you showed on your schematic because the 555 requires a logic low to trigger.

Good luck.

You have the same problem with yours as the OPs. You show a toggle, what do you think will happen if you leave the switch on? Answer, the 555 will be stuck in the on state, this is an illegal condition. A 555 requires a pulse shorter than the duration of the timer on pin 2.