I think the cheapest option so far is the 2 transistors per button solution

I think the circuit in post #9 using a CD4044 is comparable

The active electronics cost per channel is about $0.14 if using a CD4044 and $0.10 for the diode.
By comparison, two transistors cost about $0.40-0.50.

Of course, if you need to drive an LED with more than a couple mA, then you will need to add a transistor to my circuit.

 

You seem to be looking for problems.

Not knocking your design Just trying to understand why debouncing isn’t needed.

What is you guess problem that multiple buttons pushed could cause?

It depends on the application.
I understand how the priority works.
But there is an assumption that only one button is pressed at a time. If more than one button is pressed and bouncing occurs, which input will be selected?

 

Depends on the design.

If push-buttons generate clock pulses then it depends on if the D-type flip-flop changes state on the rising edge or the falling edge. Let us make the assumption that the flip-flop latches on the rising edge.
If there is no bounce on the trailing edge (when the button is released) then the switch with the late rising edge transition wins. If there is bounce on the trailing edge, then the switch with the latest rising transition again wins. So it depends on which switch bounces and when it happened.

In the two designs offered at this point, the transistor circuit and the IC circuit, both of these are level sensitive, not edge triggered. Hence the button that is last released wins. Let us assume that multiple buttons are released at the same time and all have bounce. Then the button with the last bounce wins.

Which ever way you cut it, what would you expect if multiple buttons are released at the same time?
This does not constitute a valid problem.

It is as if you are selecting one of two radio stations and you have fat thumbs. So if you got the wrong station whose fault is it?

 

Depends on the design.

If push-buttons generate clock pulses then it depends on if the D-type flip-flop changes state on the rising edge or the falling edge. Let us make the assumption that the flip-flop latches on the rising edge.
If there is no bounce on the trailing edge (when the button is released) then the switch with the late rising edge transition wins. If there is bounce on the trailing edge, then the switch with the latest rising transition again wins. So it depends on which switch bounces and when it happened.

In the two designs offered at this point, the transistor circuit and the IC circuit, both of these are level sensitive, not edge triggered. Hence the button that is last released wins. Let us assume that multiple buttons are released at the same time and all have bounce. Then the button with the last bounce wins.

Which ever way you cut it, what would you expect if multiple buttons are released at the same time?
This does not constitute a valid problem.

It is as if you are selecting one of two radio stations and you have fat thumbs. So if you got the wrong station whose fault is it?

I agree..it depends .
Wether or not it constitutes a problem also depends on the application.

 

I agree..it depends .
Wether or not it constitutes a problem also depends on the application.

It depends on the circuit. It is not a problem if it is a combinational circuit.
It is a problem if it is a sequential circuit, such as a clap activated lamp or a deli counter.

 

I agree..it depends .
Wether or not it constitutes a problem also depends on the application.

Find me an application for radio buttons where this would be a problem.

 

Find me an application for radio buttons where this would be a problem.

If there is no debouncing and bounce occurs, will there be a predictable input selection?
(In the case where multiple buttons have been pressed at the same time)

 

If there is no debouncing and bounce occurs, will there be a predictable input selection?
(In the case where multiple buttons have been pressed at the same time)

It's whichever input goes low last.

 

If there is no debouncing and bounce occurs, will there be a predictable input selection?
(In the case where multiple buttons have been pressed at the same time)

Can you demonstrate why you think switch bounce would create a problem?

 

Can you demonstrate why you think switch bounce would create a problem?

you haven’t answered my question.

 

you haven’t answered my question.

I believe all your answers can be found above and I am starting to think that you are a troll.

What was your question?

 

I believe all your answers can be found above and I am starting to think that you are a troll.

I don’t know why you would.

What was your question?

See post #13
I’ve lost interest now.
But I recommend to debounce the switches.

eT

 

See post #13
I’ve lost interest now.
But I recommend to debounce the switches.

Have you read the answers?
And the answer is still NO. You do not need to debounce the switches.
Where did you ever get that strange notion?

 

I don’t know why you would.

Because you keep posting the same question and act like you don't understand the answer.

But I recommend to debounce the switches.

Recommend if you like, but it's totally unnecessary, and a waste of electronic parts for the circuit.

It would appear you have some difficulty understanding how these circuits work so suggest, especially if you've lost interest as you state, that you troll elsewhere.

 

Because you keep posting the same question and act like you don't understand the answer.
Recommend if you like, but it's totally unnecessary, and a waste of electronic parts for the circuit.

It would appear you have some difficulty understanding how these circuits work so suggest, especially if you've lost interest as you state, that you troll elsewhere.

when did you become owner of this this forum!
Neither you or MC have given me an sensible answer to my question.

 

when did you become owner of this this forum!

I don't have to be the owner to express my opinion or give a suggestion.
You, of course, are free to do whatever you like with that suggestion.

Neither you or MC have given me an sensible answer to my question.

We have, but apparently not in words you understand.

If you look up the difference between combinational logic (which this logic is) and clocked sequential logic (which this is not) then you should see why multiple pulses to one or more inputs of combinational logic (as would be caused by switch bounce) has no affect on the final output state after the bounces stop.
I don't know if it can be stated more clearly (or sensibly) than that.

 

It depends on the application.
I understand how the priority works.
But there is an assumption that only one button is pressed at a time. If more than one button is pressed and bouncing occurs, which input will be selected?

It would appear that you are obsessed with two separate issues:

(1) You insist that push-button switches need to be debounced.
(2) You assume that a problem exists if two switches are pressed at the same time.

Let us assume that all the switches are debounced. This eliminates issue (1).
What would be your answer if two buttons are pressed simultaneously?

Both @crutschow and I have repeatedly given you the answer which you refuse to accept.
I will repeat the answer here.
Actions (1) and (2) are independent.
Both (1) and (2) have no effect on the outcome of either two circuits presented here.

Here is a thought experiment.
Take an R-S flip-flop. Apply clean debounced signals simultaneously to the SET and RESET inputs of the flip-flop. What is the outcome at the Q-output of the flip-flop?

Now can we end this discussion?