My immediate problem is turning a long pulse into a short -ve pulse to trigger a 555. I'm using a FET to invert a slowly varying voltage,and plan to use a capacitor/resistor passive differentiator to turn the low going output signal into a low going spike. Two questions. 1 - What kind of values will give me a 10-100 ms spike? 2 - in primers on differentiators, they show a negative going spike - probably not good to take 555 pin 2 below ground - is this a problem and if so, how should I fix it.

 

How slow is the "slowly varying voltage"?

 

its variable - a voltage divider with a CdS in it. General application would be shadow passing over CdS - thanks!

 

You will not get a sharp pulse from a slowly moving signal with an integrator. A better way may be to use a comparator, such as the LM339, to generate a fast pulse when the voltage crosses the comparator selected trip voltage.

 

Thanks - I was hoping to find a simple way to clip the signal without using another chip.

 

Thanks - I was hoping to find a simple way to clip the signal without using another chip.

Well, you can try a RC differentiator and see how it works. It would be a simple experiment. You might use R*C time-constants of 10ms to 100ms to start.

 

your point about 'slowly changing' is taken. But in the case ie of a shadow passing across a CdS, there will be a fluctuation, one cycle of an asymmetric waveform

 

your point about 'slowly changing' is taken. But in the case ie of a shadow passing across a CdS, there will be a fluctuation, one cycle of an asymmetric waveform

I understand. But unless you can further quantify the nature of the wave frequency and amplitude, it's not possible to give an analytical answer better than I gave you.

 

My immediate problem is turning a long pulse into a short -ve pulse to trigger a 555. I'm using a FET to invert a slowly varying voltage,and plan to use a capacitor/resistor passive differentiator to turn the low going output signal into a low going spike. Two questions. 1 - What kind of values will give me a 10-100 ms spike? 2 - in primers on differentiators, they show a negative going spike - probably not good to take 555 pin 2 below ground - is this a problem and if so, how should I fix it.

I suggest you post the schematic, and a description of the overall function of the device.

The 555 has a trigger pin which already provides a comparator type function, and depending how you wire the 555 pins you can probably configure it to trigger on a slowly high-going signal (so can also elminate your FET).

There are also tricks you can do with the threshold pins etc on the 555 to trigger it under desired conditions.

 

Pin 5 - that's an interesting idea. I've messed with 555s for years but haven't done much with it.

The bones of my circuit couldn't be simpler - on the input side, a CdS (ldr) in a voltage divider, and a 555 monostable. Currently inverting the divider voltage with a FET to give a low going pulse to 2.

But back to pin 5 - my understanding is that a varying v on 5 messes with the 2/3 vcc value, and thus it varies the shape of the output pulse width. My goal is to trigger the monostable (ie with pin 2) using the CdS voltagedivider as input.

 

On the subject of the trigger pin (2) - RB said "depending how you wire the 555 pins you can probably configure it to trigger on a slowly high-going signal (so can also elminate your FET)." That is what I am trying to do But as Crutschow has argued, its hard to see how to do that without a comparator or schmidt trigger, given that the varying voltage is not a repeating waveform.

RB - "There are also tricks you can do with the threshold pins etc on the 555 to trigger it under desired conditions. " In in astable mode, the threshold retriggers pin2. I saw a circuit which applies a varying Vin on pin 6 (threshold), tied to pin 2 - trigger. (with no RC, pin 7 open). That's a good trick, but is essentially a 555 wired as a schmidt trigger.

 

Yeah you really need to specify if your sensor is going \ or / edge (or if you can select that) and also if the timed output period needs to be Hi or LO.

With that info it might be possible to do what you need with just the 555, there are a few 555 experts here and the 555 datasheet is quite informative too.

 

An R-C "differentiator" is actually a (crude) way of turning a rate of change into a voltage. What you're trying to detect is a rate of change in the light falling on the photocell. So, you can feed your light-dependent voltage through a C-R differentiator with the resistor biased to some voltage above the 555's trigger threshold voltage, and feed that into the trigger input.

When the light drops at a rate high enough to cause the differentiator output to fall below the trigger threshold, the 555 will trigger. You can adjust the difference between the differentiator bias voltage and the trigger voltage to set the sensitivity. For this application I would use an RC time constant of around a second.

If you'd like more detail, just ask.

 

Isn't there a 7400 series chip to lengthen pulse width? About a eon ago , I remember troubleshooting a bunch of bad boards that no one else could fix.

Long story short, this chip was bad. There was a tiny delay between when the input pulse started and when the output pulse started. It was enough to throw the timing of the whole circuit off. Replacing the chip did not fix it. We had a bad batch of chips. Our idiot purchaser ordered the same batch number even though I told him to pick another manufacturer. A trip to Radio Shack ended up getting a couple dozen boards off the repair shelf. But I must have paid any times per chip than I had to.

Wish I could remember the chip we used.


Ah afterr lla of that OP wants to shorten pulse width.

 

try this simple circuit......