Rising and Falling Edge triggered one-shot

How long is long, how short is short, and what supply voltage do you plan to use?

Cheers

Phil Hobbs

How short is much shorter? XOR against a delayed copy of itself?

That's the classic. An RC, or a string of gates, can do the delay.

Or even a PCB trace!

Hi and thanks for responding.

I have a touch sensor. Goes on when touched and off when not. I want my u C to do a service operation at each edge of that (possibly long) 'pulse'. My uC needs about a second or so from power up to the beginning of program start. The uC is energized by that sensor signal being ON, at which point it immediately asserts another bit that controls whether to continue power to itself and other circuits. So the one-shot duration is just long enough to wake up the uC, which then services the event and then goes back to sle ep.

I plan to use a four cell (or possibly six) penlight pack for power. I'm h oping to tolerate battery decay to just above 3.3 volts from a max of about 6 or 9.

So the one-shots are about a second or so in duration and the spacing betwe en the sensor signal edges is typically several seconds. Most one-shots I' ve found assume a trigger pulse much shorter than the one-shot itself and I need the reverse. An input that initiates an identical pulse at either ed ge would be ideal.

Price is a consideration - a cheap timer and few external components is a p lus.

I want to implement a sleep mode whose quiescent current only powers the to uch sensor (at ~ 5nA) to maximize battery life with all else being off.

Thanks, Ted

ulse at its rising and falling edges. Ideally, only a single input should be required (i.e. the one long pulse) for minimal part count. I've looked over several timer ics and none seem to fit the bill or are too pricey like the 6993s. Can anyone suggest an appropriate ic? I would think it a comm on enough function to warrant a dedicated chip. Thanks, Ted

Hi and thanks for responding.

I have a touch sensor. Goes on when touched and off when not. I want my u C to do a service operation at each edge of that (possibly long) 'pulse'. My uC needs about a second or so from power up to the beginning of program start. The uC is energized by that sensor signal being ON. At program sta rt it immediately asserts another bit that controls whether to continue pow er to itself and other circuits. So the one-shot duration is just long eno ugh to wake up the uC, which then services the event and then goes back to sleep.

I plan to use a four cell (or possibly six) penlight pack for power. I'm h oping to tolerate battery decay to just above 3.3 volts from a max of about 6 or 9.

So the one-shots are about a second or so in duration and the spacing betwe en the sensor signal edges is typically several seconds. Most one-shots I' ve found assume a trigger pulse much shorter than the one-shot itself and I need the reverse. An input that initiates an identical pulse at either ed ge would be ideal.

Price is a consideration - a cheap timer and few external components is a p lus.

I want to implement a sleep mode whose quiescent current only powers the to uch sensor (at ~ 5nA) to maximize battery life with all else being off.

Thanks, Ted

ulse at its rising and falling edges. Ideally, only a single input should be required (i.e. the one long pulse) for minimal part count. I've looked over several timer ics and none seem to fit the bill or are too pricey like the 6993s. Can anyone suggest an appropriate ic? I would think it a comm on enough function to warrant a dedicated chip. Thanks, Ted

Chips like:

CD4538 -

seem promising, but I'd like to spice test whether any can do the job but can find no models. Hoping I might happen upon a guru.

Could only find: HC4538_JT.zip -

Chips like:

CD4538 -

seem promising, but I'd like to spice test whether any can do the job but can find no models. Hoping I might happen upon a guru.

Could only find: HC4538_JT.zip -

interrupt on pin change?

Hi Mike,

I think that would require my Bluetooth Bee Standalone v1 (the uC) to remai n energized to catch the interrupt, so that would not work. I could simply keep the uC on while the touch sensor is on, but I'd rather not consume cu rrent when the uC is not needed for service (i.e. the bulk of the long puls e).

Well, that sort of lets out the XOR plus delay approach. You could use half of an MC14538A or CD14538A with the long pulse capacitively coupled into both trigger inputs (separately). That's one chip and three RCs, and will work over your supply range.

The CD4047 is sort of a monostable/astable toolkit, with a separate pin to control retriggerability. Looks like you'd still need two RCs though.

But before somebody else says it, you can do all this and more with a very small microcontroller such as an ATtiny5 (20 cents) and a micropower voltage regulator.

Cheers

Phil Hobbs

I only spent a few minutes with google, but going from bluetooth bee to the chip specs

I'm a PIC person, so have no direct experience here.

I'm not sure I understand the problem, but my first thought was an inverter and some RC high passes.

George H.

want my uC to do a service operation at each edge of that (possibly long) 'pulse'. My uC needs about a second or so from power up to the beginning of program start. The uC is energized by that sensor signal being ON. At program start it immediately asserts another bit that controls whether to continue power to itself and other circuits. So the one-shot duration is just long enough to wake up the uC, which then services the event and then goes back to sleep.

how much of that interconection is non-negotiable?

why do you need the one-shot when the sensor turns on?

I still dont know what you want.

so, how does ths work when the sensor turns off again - what powers the one-shot?

Doesn't the CPU support a hibernate mode with wake on change for certain pins? Even if it was only for one edge transition then a pair of those pins and a single inverter would get you what you want.

The sleep and hibernate modes are worth studying (as is figuring out the minimum clock rate you can operate at and still be just fast enough).

It seems a bit brute force and crude to start up from cold every time.

The batteries will leak and destroy their battery compartment if you push them that far. Duracells these days seem particularly prone to leaking if used in low voltage tolerant low current drain LCD devices.

I notice in passing that Everyready are advertising their latest alkaline offering as "leak free" - that *is* a brave claim.

It probably isn't worth getting average quiescent current much below the self discharge rate of typical alkaline batteries (ie

For microseconds, I agree, but the OP wants seconds.

Cheers

Phil Hobbs

The "standard" approach...

...Jim Thompson

microfarads,megaohms and a CD4070BE will achieve that, seems an unpopular part, but TI is still making them.

I guess that's always good for a pulse wide enough to clock a flip-flop.

If you're squeamish add an even number of inverters between pin-5 of the 'HC74 and pin-2 of the'HC86. ...Jim Thompson