super simple pulse stretcher

I have a need to take a square pulse of 0V/+5V and about 5-10ms duration, and stretch it out long enough to flash a LED brightly enough to be visible indoors under "ordinary" lighting conditions.

I currently only have op-amps on the board and would rather not add any logic ICs to do this. So using low-speed op amps and discretes would be fine. The pulse source into the stretcher would be low impedance.

An optional feature would be to have the "stretch" duration decrease the more rapidly pulses are coming into the circuit. But that is not a requirement if it needs significant complexity.

Reply to
bitrex
Loading thread data ...

What's wrong with a 5-10ms flash? I'm flashing some at 400us and they are easily visible in my shop with florescent lighting.

Reply to
John S

NPN emitter follower driving R-C to ground, then a comparator. Use an opamp as the comparator.

--
John Larkin         Highland Technology, Inc 
picosecond timing   laser drivers and controllers 
 Click to see the full signature
Reply to
John Larkin

Supplies? ...Jim Thompson

--
| James E.Thompson                                 |    mens     | 
| Analog Innovations                               |     et      | 
 Click to see the full signature
Reply to
Jim Thompson

I'm probably going to need to be using something like an LM324A. Will they work OK in that role?

Reply to
bitrex

Inject the pulse via a diode to the (+) input of a op-amp. The (+) input will have a small cap on it going to common for the delay off with a R also going to common as part of the RC calculations. Put a voltage divider on the (-) input of the op-amp of lets say 2.0 volts, that will be your threshold for the comparator.

You may also want to have a low current feed back from the output to the (+) input to aid in a little snap, hysteresis, for the off state.

If for some reason your reference pulse appears to be very low current in nature, it could have issues driving the cap you select. If this is the case, buffer the pulse with another op-amp in unity mode or use a NPN trannie in emitter follower mode to drive it, in which case, you won't need the diode.

Jamie

Reply to
M Philbrook

Looks like it's going to be +5V virtual ground split into +2.5 and -2.5 at this point so they'll be low-voltage opamps.

Reply to
bitrex

That will work just fine. Just keep in mind that a 324 gets closer to the bottom rail than it does the top rail. Infact, it can become detached but I don't think that should cause you any issues.

Jamie

Reply to
M Philbrook

I'll explore that a bit in simulation, thanks.

Reply to
bitrex

So your input pulse is -2.5V to +2.5V _relative_ to your virtual ground? ...Jim Thompson

--
| James E.Thompson                                 |    mens     | 
| Analog Innovations                               |     et      | 
 Click to see the full signature
Reply to
Jim Thompson

Nosir, it will be 5 volts relative to the overall common, i.e. the -2.5 volt point wrt virtual ground.

Reply to
bitrex

Isn't that what I just said ?:-) ...Jim Thompson

--
| James E.Thompson                                 |    mens     | 
| Analog Innovations                               |     et      | 
 Click to see the full signature
Reply to
Jim Thompson

Sure. If you have two sections available, use one open drain to discharge the capacitor, and the other section as the threshold sensor. Quiescent high, RC pulsed low. Easy.

We sometimes have a "trigger" LED that is basically a stretched input. In some cases, we want to catch very short pulses - a couple ns - but also indicate a static high input. Laser drivers want that sort of indication. We use a fast one-shot (SN74LVC1G123) and OR its input into the output.

--
John Larkin         Highland Technology, Inc 
picosecond timing   laser drivers and controllers 
 Click to see the full signature
Reply to
John Larkin

Correct.

Reply to
bitrex

I'm not sure I follow what you mean by using one "open drain." If you have a moment could you show me in LTSpice? Also I only got 4 hours sleep last night and jesus I'm lazy today.

Reply to
bitrex

Oh, sorry, I was thinking LM339, the comparator. I'm starting to forget those antique part numbers.

If you use opamps, you'll need a diode or a transistor to discharge the cap. An active discharge part (transistor or mosfet) can give fast discharge and high stretch ratios.

formatting link

and here's an oldie

formatting link

--
John Larkin         Highland Technology, Inc 
picosecond timing   laser drivers and controllers 
 Click to see the full signature
Reply to
John Larkin

GND From /-\ V+ supply ___ V+ supply ( ) Threshold ref -|___|- \+/ ___ ___ | |--+|___|+-+++|___|-GND V -> + | R4 - | R3 + | short pulse + | | |/ | |\| ||-+ o--------o-| NPN +-+|-\ longer Pulse || | >-+---o-------+--------+||-+ + +++|+/ | + | | |/| | | +-----+-----+ ___ | | | +--+|___|+--+ | + .-. R2 | --- | | | C1 --- | |R1 | + '-' | | | | | | | === === + GND GND GND (created by AACircuit v1.28.6 beta 04/19/05

formatting link

C1 and R2 is the charge constant hold time before it goes below Threshold.

R2 is a + feed back to aid in a little snap action for the off cycle, make sure to select a value that does not hold + input higher than the reference input while output is on, otherwise it'll latch.

Jamie

Reply to
M Philbrook

Cool, thanks!

Reply to
bitrex

I think I may go with option #2 - should the op amp be configured as a comparator in that example?

Reply to
bitrex

The one with the 2N7002?

Opamps usually work OK as comparators, as long as you don't need super fast response or rise/fall times. They usually don't oscillate so generally don't need hysteresis.

So there's not really a "configure as a comparator" issue.

Oh, there's one gotcha. Some brands of LM324 do weird things when used as a comparator. When one section rails, it messes up the bias of the other three sections. That's because opamp sections use shared current sources. The original National part did that, maybe still does.

--
John Larkin         Highland Technology, Inc 
picosecond timing   laser drivers and controllers 
 Click to see the full signature
Reply to
John Larkin

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.