T&L Publications

Hello all,<p>I'm new to electronics and am trying to build an event triggered stroboscope by piecing together bits of this and that from the internet. The idea is to use an IR LED and a phototransitor to detect a tab on a spinning wheel and trigger a timer, in this case a 556 in momostable mode (I didn't have a 555 handy), to produce a constant length pulse which turns on a super bright LED.<p><p>The problem I'm having is that the phototransistor circuit won't trigger the 556. If I manually connect the trigger to ground, the timer fires the LED. The transistor Q1 won't and I can't figure out why not. What am I missing?<p>Thanks for any help,
FlatTop

I think I would remove the differentiator (R1,C2) and try DC coupling.<p>Alternately, I would increase the size of the capacitor. Maybe 1 ufd - anything to guarantee that I could pull the trigger input down to zero. I'm not sure the existing circuit will do that.

Or try a higher value for R1 - say 50k~100k.<p>Also, I'm thinking if you used a CMOS 555 (or 556) you might could do away with Q1.

Hi,
The first thing you need to do is to turn Q1 off between the start pulses. Connect a 10k resistor from Q3 emitter to 0v(GND) This will rearm(discharge C2) the trigger and make a new sequence possible.<p>The R3 is way too high in resistance, I suspect that you should use 470 ohm not kohm. 470 ohm gives you a current through the LED(D2) of about 6mA, a normal value. R6 should be the same or maybe 330 ohms, for a proper light. This is depending on the LEDs you are using. If your circuit fires the LED, I suspect a typo for the 'k'.<p>If so, see my first note and check the result! If it doesn't work, reduce the pulldown resistor to 4,7 kohm.<p>TOK <p>[ July 02, 2005: Message edited by: Gorgon ]<p>[ July 02, 2005: Message edited by: Gorgon ]</p>

That 470k is definitely wrong! I didn't even look at that part.<p>You can disconnect Q3 from R5 and then connect R5 to +5V to see if the rest of the circuit functions. There's certainly nothing wrong with pulling Q3 base to ground with another 10k resistor. It will guarantee turn-off, assuming the opto device doesn't leak too much.

Your phototransistor circuit looks wrong to me. Here's how I have done it in several projects
<p>You should just feed OUTPUT directly into the 555 trigger. The trigger pulls less than 1 uA (per LM555 datasheet from National). It certainly has the right shape for it and a PT is basically just a transistor with base current coming from light.<p>So get rid of R1, R4, R5, C2 and Q1. With Q3, use a collector pullup (10K) to +V, emmitter to ground and collector to 555 trigger. You might need to play with the pullup resistor value depending on your PT.<p>As noted previously, both your LED dropping resistors are too high 470 vs 470K but I suspect the "k" is a typo since you said the superbright LED works.<p>Also, if I do the math correctly, your monostable 555 will trigger for 110mS seconds. Is this what you want? seems kind of short.<p>[ July 02, 2005: Message edited by: philba ]</p>

I like philba's simplification. But if you're concerned about the waveshape from the PT output you could insert a schmitt trigger between output and 555 trigger input to make it "snappy". and I think there is a way to make a crude one shot with a diode & R/C combo using the extra gates of a 74LS132. If so, that would get you back to one IC.<p>And who was Schmitt anyway?

Using a fairly low pull-up, like 10K should give the wave a reasonably crisp shape. The voltage won't drop until the PT is reasonably saturated though check the PT datasheet on that one. I don't think one needs hysteresis for this circuit as the 555 won't retrigger for at least the on period. <p>By the way, a very useful tool for experimenting with circuits is a SPICE emulator. You can build these circuits and play with them. You can look at voltage and current at any place in the circuit. It makes learning this stuff really easy and allows me to prototype a circuit with out ordering parts first. There is nothing more maddening that discovering that you don't have the right parts to finish your project. I use switcher cad from Linear. its free and is relatively easy to figure out. they have a 555 model that I have used.<p>Phil

In re-reading the topic starter from flattop I'm a little puzzled on the intended result of the project. If a spinning wheel triggers the illuminating LED and he is looking at the spinning wheel with light reflected from from the strobe, the wheel's motion will always appear to be suspended, regardless of how fast the wheel is spinning. Maybe it's a demonstration project of some sort.<p>And wouldn't it be true that when strobe lights are used to obsverve motion, the shorter the illuminating pulse, the better the perceived resolution of the object?

<blockquote><font size="1" face="Verdana, Helvetica, sans-serif">quote:</font><hr>Originally posted by jimandy:

And who was Schmitt anyway?<hr></blockquote><p>After some googling, here is the answer to your question. Otto H. Schmitt. http://www.thebakken.org/research/Schmitt/Otto.htm <p>Dig in <p>FlatTop,
How is the circuit adapting?<p>TOK

Thank you Gorgon, for the Schmitt link! Fascinating bio, plus he was born 10 miles from my birthplace of Buffalo, NY!

Oops! Historical correction- Schmitts' father Franz was born near Buffalo. "The" Otto Schmitt was born in St.Louis, MO.

I 'Otto' not say this but I am 'Schmitten' by the perserverance of members of this forum to find an answer to every question.

I would like to give a hearty thanks to all for the quick responses. I had no idea I would get that many answers that quickly. Unfortunatly I haven't had time to try any of them, but as soon as I do I'll post the results here.
I can answer a couple of questions, tho.
Gorgon, the 470K was in fact a typo. On the breadboard its 4.7k.
rstofer, I would like to keep the differentiator to ensure that the trigger signal stays shorter than the output, otherwise the output will stay high all the time. I think. Once I get it triggering correctly, then I can drop the output time to something like .001 seconds to give a nice, crisp view of the spinning wheel. Which brings me to...
jimandy. You nailed it right on the head. The idea is to make the spinning wheel appear stationary. And the shorter the pulse, the fewer degrees of revolution per pulse yielding a better view. I'm not at that stage yet, but I'll burn those LEDs when I come to it.<p>Thanks again, guys. I'll let you know how it goes.<p>FlatTop

<blockquote><font size="1" face="Verdana, Helvetica, sans-serif">quote:</font><hr>Originally posted by flattop:
Gorgon, the 470K was in fact a typo. On the breadboard its 4.7k.
<hr></blockquote><p>Then you are using some fantastic bright LEDs! If both the 470k is 4k7, you will get around 0.6 - 0.7 mA through the LEDs. I can possibly understand it for your disk reader, but the strobe LED. Amazing! <p>TOK