We are using a matrix to detect depressions of a membrane keypad.
From a perspective of human factors what is suggested debounce time?
I typically use 10milliseconds and have not had any problems.
Usually a scan rate of something like 10 ms will do.
I'm not sure what you mean by "human factors". If you don't debounce the key long enough to suit the mechanicals, the human user will be annoyed. And most switches can be debounced faster than most humans can type, so that shouldn't usually be an issue.
(A good keyboardist is usually around 3-5 characters per second, or 40-60 words per minute; the world record is 170 words per minute. Human reaction time to see something and respond averages on the order of 100 ms. Visual response time -- the length of time the stimulus has to be present -- is about 30 ms. All the numbers are rough, of course.)
I think the membrane keyboard takes 10-20 seconds.
the main "human factor" is DO NOT take the easy way and react to the key on release.
anyhow, my 'grab bag' debounce is 10ms, but, of course, there is no guarantee that is enough for yours.
one more re "human factors" since the touchscreens have trained every teenager to 'tap' I would make the debounce as short as safely possible
View all questions in Keil forum