Buffer as switch debounce

Dave Jones crafted a schematic that is much simpler than that. Of course you need to sit through a 45 minute video on youtube to get it, but you can fast-forward until you see the schematic.

Link? ...Jim Thompson

-- | James E.Thompson | mens | | Analog Innovations | et | | Analog/Mixed-Signal ASIC's and Discrete Systems | manus | | San Tan Valley, AZ 85142 Skype: skypeanalog | | | Voice:(480)460-2350 Fax: Available upon request | Brass Rat | | E-mail Icon at

Power-on state?

RL

Found it. I wouldn't call that "simpler". ...Jim Thompson

Why waste the other half of that flop?

Picky! Picky! Go back to the SPDT approach... or add a POR to the D-FLOP. ...Jim Thompson

-- | James E.Thompson | mens | | Analog Innovations | et | | Analog/Mixed-Signal ASIC's and Discrete Systems | manus | | San Tan Valley, AZ 85142 Skype: skypeanalog | | | Voice:(480)460-2350 Fax: Available upon request | Brass Rat | | E-mail Icon at

I did this one (sorry not been following thread)

It has indeterminate startup I think (did not matter for this).

It takes a SPDT switch, which the OP opined he's like to be rid of. ...Jim Thompson

I have something right now that if I mess with it enough I can get a bad contact although most of time it's fine. I'm using an optoisolator with an npn output. The output has a pullup resistor on it and a .1uf cap to ground and that is connected to a pin on a raspberry pi.

My question is -- maybe I can find an opto that has a smoother output? Don't they make some that have more sophisticated logic outputs? You guys may know. That would be great because I could just plug in a better part.

Why not software debounce?

There are opto schmitts. Do you actually need the isolation? I see people use optos grounded on both sides, as if they have some magical filtering properties, which they don't.

Does anyone have a PWL or similar that you can plug into LTSpice to model the switch bouncing? I guess one could capture a trace on a digital scope or lab jack.

So add a POR...

...Jim Thompson

-- | James E.Thompson | mens | | Analog Innovations | et | | Analog/Mixed-Signal ASIC's and Discrete Systems | manus | | San Tan Valley, AZ 85142 Skype: skypeanalog | | | Voice:(480)460-2350 Fax: Available upon request | Brass Rat | | E-mail Icon at

Weird EMP protection? In the worst case the opto dies, not the MCU. A pair of depletion mode FETs would be better.

Best regards, Piotr

No, just bad design.

What kind of switch is it? Non-snap membranes are the worst; they can bounce (and ooze) for 100 milliseconds.

Here's the netlist from NoiseBlank.pdf on the S.E.D/Schematics Page of my website. The digital pulses may be difficult to decipher so the PDF has been updated to show the pulse combinations to make a noisy signal.

• Schematics Netlist *

R_R1 V2 0 1K R_R2 V3 0 1K U_DSTM3 STIM(1,1) $G_DPWR $G_DGND V1 IO_STM IO_LEVEL=0

• 0 0
• +10u 1

• +10u 0
• +10u 1 + ENDREPEAT R_R3 V1 0 1K U_DSTM1 STIM(1,1) $G_DPWR $G_DGND V2 IO_STM IO_LEVEL=0
• 0 1
• +0.5u 0

• +4u 1
• +1u 0 + ENDREPEAT R_R4 N_1 N_2 39K R_R5 N_2 N_3 39K R_R6 N_4 N_5 5.1K R_R7 N_4 N_6 5.1K V_V1 N_4 0 +5V X_U1A INPUT N_2 $G_DPWR $G_DGND 74HC04 PARAMS:
• IO_LEVEL=0 MNTYMXDLY=0 R_R13 N_7 0 1K R_R9 N_8 0 5.1K R_R8 N_9 0 5.1K R_R11 N_4 N_10 1K R_R12 N_10 N_7 3K X_U5A OUTPUT N_11 N_12 $G_DPWR $G_DGND 74HC00 PARAMS:
• IO_LEVEL=0 MNTYMXDLY=0 X_U4A N_13 N_12 OUTPUT $G_DPWR $G_DGND 74HC00 PARAMS:
• IO_LEVEL=0 MNTYMXDLY=0

• 0 0
• +0 1

• +150n 0
• +50n 1 + ENDREPEAT Q_Q3 N_3 N_3 N_9 Q2N3904 Q_Q4 PROCESS N_3 N_8 Q2N3904 Q_Q2 PROCESS N_1 N_6 Q2N3906 Q_Q1 N_1 N_1 N_5 Q2N3906 C_C2 PROCESS 0 3.3pF ...Jim Thompson

Well, 3 transistors, 4 resistors and 1 cap versus your 2 ICs, 2 resistors and 1 cap. And it can source some current at the output. (it is intended to be a power on/off switch)

opto schmitt is nice -- that would help a lot. using optos because this could have long cable lengths and maybe in industrial environment. hate to expose the I/O pins of a Raspberry PI to that. thanks.

Put the NPN of your opto in place of the switch in this drawing...

Depending on your application... if you don't need on/off toggling, just eliminate the D-Flop. ...Jim Thompson