trigger LED

I'm doing a small box that accepts a user trigger. I'd like to flash an LED every time it gets triggered. Something like 100 ms, on a blue LED.

The obvious circuit is a 74lvc1G123 one-shot, set to 100 ms pulse width. But how fast will that work? I'd like to go to 50 or maybe 100 MHz. Would it keep retriggering? I wouldn't want the LED to go out at high trigger rates.

There is a specified "retrigger time", but I'm not sure what that means. Has anyone really over-triggered these things?

I guess I could breadboard and try it, but asking is easier.

--
John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  
 Click to see the full signature
Reply to
John Larkin
Loading thread data ...

If you trigger a toggle flipflop, the Q and /Q make an AC signal; bridge rectifier into a resistor-parallel(LED, capacitor) will light the LED as long as events happen, and fade the LED out at an RC time... it looks like you want an 'activity' light to look.

There's lots of designs for logic probes (a common project forty years ago) that also would have clever ways to indicate pulses.

Reply to
whit3rd

Differential transistor pair with one base biased to half the nominal pulse amplitude, and the other gets the pulses through an RC lowpass to its base is pretty bug-free. Put the LED in the appropriate collector line for whether you want it to blink for positive or negative logic signal.

Reply to
bitrex

Or rather use PNP or NPN transistors depending.

Reply to
bitrex

I omitted the (kinda important) coupling capacitor from that AC signal into the bridge rectifier; the RC time that sets the pulse decay is from the coupling capacitor and the impedance of the LED being driven. I liked about a half-second decay, used a 10 uF nonpolar electrolytic.

Reply to
whit3rd

But what drives the LED?

--
John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  
 Click to see the full signature
Reply to
John Larkin

That would miss a fast (say, 5 ns wide) trigger pulse.

A retriggerable one-shot is ideal, if that would work at 100 MHz input rate.

--
John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  
 Click to see the full signature
Reply to
John Larkin

Section 6.6 gives timing. You're good with 5ns triggers at Vcc 2.5V or more . But retrigger, trr, times climb into us for Cext in the range you're talk ing about. And there's no mention of some kind of metastability if exceeded . Those are the pulse durations for predictable guaranteed operation. You w ill probably need to lockout the trigger input for reliable operation.

formatting link

Why do you want the LED to go out at high trigger rates? Seems an ambiguous indication confusing no activity with high activity.

Reply to
bloggs.fredbloggs.fred

How about using two monostables, 100ms led-on followed by an enforced

100ms led-off before the first trigger is re-enabled. Then the led will blink once for a single event and flash at 5Hz for sustained event bursts?

piglet

Reply to
piglet

Why not read the data sheet?

formatting link

The data you want is on page 11.

You haven't specified your supply voltage, but with a blue LED you will need more than 3V, and that would give minimum pulse widths of 3nsec, which is compatible with 100MHz operation.

Reading the data sheet would have been even easier, but that takes electronic design skills.

Building you own monostable with a couple of 5GHz transistors would have taken you even further.

Ghiggino, K.P., Phillips, D., and Sloman, A.W. "Nanosecond pulse stretcher",Journal of Physics E: Scientific Instruments, 12, 686-687 (1979).

I should have used an emitter-coupled monostable (as is mentioned in the letter), but the traditional circuit worked in the application we needed it for.

--
Bill Sloman, Sydney
Reply to
Bill Sloman

One of your patented 'Schottky + capacitor + no bleed resistor' level detectors running into a SC70 CMOS comparator might be the business, unless the trigger pulses are super short.

Same number of parts.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
 Click to see the full signature
Reply to
Phil Hobbs

How about turning the problem upside-down and generating a pulse train, and have the incoming user trigger gate it off momentarily. Then into a missing-pulse detector.

I'd expect detecting missing pulses in a train of them would be an easier thing to do reliably than detect isolated pulses of such short duration.

Reply to
bitrex

Min. trigger pulse for an 'AC123 is 5ns, but I don't think it'll re-trigger immediately, 10ns later...

formatting link
"If the input retrigger pulse follows the initial input pulse

is retriggered."

But does it matter? If you have a 10ns pulse triggering a 100ms one-shot, does it matter than the one-shot isn't re-triggered until 30ms into its timing cycle, instead of immediately? The LED will appear continuously lit, either way.

If you wanted to be sure without having to try things, a flip-flop in front of the 'AC123 could latch on the trigger pulse, and then be reset by the 'AC123 -- two small parts, minimal glue.

Cheers, James

Reply to
dagmargoodboat

Gate it off momentarily? With a one-shot!

Schematic?

--
John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  
 Click to see the full signature
Reply to
John Larkin

That's a pulse stretcher, but with a 20 million stretch ratio.

That's hard to do in one stage. Three or four cascaded stretchers would work, but that's a lot of parts.

Top posting is confusing on usenet.

--
John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  
 Click to see the full signature
Reply to
John Larkin

Say I'd want to stretch a 5 ns pulse to a 50 ms LED flash. 1e7:1 ratio. That's tricky to do analog.

I like the one-shot, if it will work. I guess I'll try it.

--
John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  
 Click to see the full signature
Reply to
John Larkin

As long as the LED stays lit, I don't care how confused the one-shot is, or how confused I am.

I previously built a fast one-shot from a 1 ns Tiny flop that reset itself, and use that to fire a second one-shot. But I want something simpler.

I was just wondering if anyone had done this.

It ought to work. I'll try it.

--
John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  
 Click to see the full signature
Reply to
John Larkin

I had a similar situation but with re-trigger rates closer to kHz, not MHz. I used an 'AC123 with a 100ms period.

But I can't really tell you if the thing was retriggering promptly -- the LED just stayed lit to the eye, which was all I cared about.

If extreme re-trigger rates befuddle the '123, I think a single flip-flop fixes it.

-+- | Rt +-Ct-. | | .----. .-------. .-|D | | | | | Q|------|> Q|----> LED trig >---|> | |'AC123 | | '----' | _| | | Q|--. | '-------' | '------------------------'

Cheers, James

Reply to
dagmargoodboat

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.