I can't produce a bounced switch signal. What am I doing wrong?

[billbyrd1945]

Bought oscope to see difference between bouncing and debounced switch. Began with cheap breadboard buttons to get my "before" tracing. Problem is-- I can't seem to find just the right combination of settings to see any bounce. No need to wire in a debounce circuit until I have a nasty button to compare it to. Every one I try looks clean on my scope. I'm assuming that I'm doing something wrong. Any suggestions would be appreciated. Scope is Siglent SDS 1052DL+, 50MHz, 500MSa/s, with settings of 1.0V, M25.0us. Thank you

[taydin]

To see the signal fluctuations you are expecting, you need to use a switch that is REALLY bouncing. Switches with spring loaded contacts are particularly bouncy and will give you what you want.

[mcinque]

To see the signal fluctuations you are expecting, you need to use a switch that is REALLY bouncing.

AFAIK, any switch bounces due to its mechanical nature.
Even with 1MHz bandwidth "toy" oscilloscope, you MUST see the bounces.
Are you using one shot trigger? Is averaging disabled? How is set the "Acquire" menu?
And just to test the oscilloscope settings, try to emulate the bounces by lightly rub the probe tip over the positive terminal of an AA battery or a psu (with the ground clip connected to GND), you should definitely see them.

[T3sl4co1l]

Two things:

1. Tactile pushbuttons tend to be surprisingly well behaved, and I suppose probably slide switches too (if they aren't dirty?).  Maybe try a power switch, or a toggle kind?

2. One edge tends to be bouncier than the other, usually closing I think.

Also, the time constant arises from the, well, contacts bouncing -- whatever their mechanical resonances are.  Usually somewhere in the milliseconds, so look on a similar time scale.

Tim

[ogden]

I am afraid that you are not using resistor in your test circuit and pull-down resistor is 1Meg of your scope input. Most likely that's why you do not see bounce.

Try circuit below, with 470 ohm resistor. Note that low pull-up value is for bounce testing only - to get fast signals. Usually you would use >= 10k pullups.

[Mechatrommer]

Try increase timescale to the order of ms /div

[billbyrd1945]

Ah ha! Thanks to everyone. I apologize for not including a pic of my circuit. For the record-- I was using 10k for pull down. The 470r made sense to me and the quick brush across the AA battery was a great solution as well. Did them both, got my bounce. Thanks much. EEVblog is the best!