switch debouncing

At AC you have this- I don't see good and uncomplicated reason why the oscillation should not sustain itself: View in a fixed-width font such as Courier.

. . . . . . 5 | . | . +-------------+ | | | | . | | VT+| /\\ /\\ / \\ . | | | | / \\ | / \\ / \\ . | |\\ |\\ | VT-| \\ / \\ / \\ / \\ / . +-| o---| o---+-> | \\/ \\/ \\/ \\ / . |/ |/ | | | . | | | | . schmitts 0 +---+----+---+-------------------- . 1x 2x 3x etc... . . Tpd->

The main advantage of the 555 would be extremely precise thresholds compared to that logic Schmitt inverter and you can still use the Ganssle circuit. He wants 0.7R2C>=20ms so for C=0.47u R2=56K then R1 can be anything nominal compared to R2 like 10K. This makes the worst case delay 30ms at 1.1RC. The draw back is one sw/555 whereas you get three sw/HC14, or possibly six. View in a fixed-width font such as Courier.

. . . +-------------------+-------------+------< . | | | . | | | . | ------------- | . | | V+/RST | | . [10K] | | | . | | | | . | |THR OUT|------------->

. +--[56K]--+--+TRIG/ | | . | | | DIS| | . o === | |N/C | . -| 0.47U| | CON| === . o | | GND |0.1U . | | ---------- 555 | . | | | | . +---------+---------+-------------+------< . . . . . .

Well, that wasn't shown.

low pass input filter with R2. The low output R of G2 is vital to circuit operation.

Neither end of C is connected to ground; both ends are connected to semiconductors, and semiconductors are what's killed by esd zaps. 50 ohms won't impress a good arc, or G2 can be the IC that's killed. Of course, the transzorbs fix that problem.

One of my pretty firm rules is to never have semiconductors connected to the outside world, arcs included.

input and the C,R2 node.

This schematic seems to have all sorts of invisible parts. Good, that resistor definitely shifts the damage path toward G2.

not needed with large values >of R2

a miss trigger to propergate from G1 to G2.

of G2's output 50R and R2=1e5 we only need a 2000V input change to cross the 1V hytersis threshold.

Again, assuming R2 doesn't arc.

Well, if you analyze it carefully enough it should be OK. But simpler circuits have less hazards. Without the transzorbs, this one would be scairy.

John

You want debounce. See....

Newsgroups: alt.binaries.schematics.electronic Subject: Switch Debounce (from S.E.D) - NoiseBlank-555-RC.pdf Message-ID:

...Jim Thompson

-- | James E.Thompson, P.E. | mens | | Analog Innovations, Inc. | et | | Analog/Mixed-Signal ASIC's and Discrete Systems | manus | | Phoenix, Arizona Voice:(480)460-2350 | | | E-mail Address at Website Fax:(480)460-2142 | Brass Rat | |

I've seen switches with bounces 50 to 500x as long as yours.

Sheeesh, Win, It's just an example, dredged from a noisy voltage signal application... change the RC already!

I have some EXTREME "debouncers" if anyone is interested.

...Jim Thompson

=_NextPart_000_0077_01C5D4AF.2B09E040 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable

"John Larkin" wrote in = message news: snipped-for-privacy@4ax.com...

message=20

On Thu, 20 Oct 2005 03:50:39 +0000, Fred Bloggs wrote: [snip]

Nope. ;-)

I'm not sure that could be called a debouncer- maybe more of a noise blanker. You have that known V1 clock and corrupt it with much higher frequency (1-V2V3/25). A switch debouncer does not have that luxury.

low pass input filter with R2. The low output R of G2 is vital to circuit operation.

input and the C,R2 node.

not needed with large values >of R2

a miss trigger to propergate from G1 to G2.

of G2's output 50R and R2=1e5 we only need a 2000V input change to cross the 1V hytersis threshold.

Did the OP say he was designing a wheel-well switch for the AirBus?

The sources are for illustration ;-) NOT "known V1 clock"... it's NOT a clocked system... look carefully before leaping for a change.

...Jim Thompson

I had a temperature switch that bounced forever. It was mounted in a tugboat, on the side of a diesel engine.

John

See "NoiseBlank.pdf" on the S.E.D/Schematics page of my website for several different variations.

...Jim Thompson