I've just gotten my hands on a brand spanking shiny new Rigol DS1054Z oscilloscope (yay!) and I just had to fiddle around with it a bit.
My simple tests included watching switches and buttons bounce and during these thest I found something rather surprising.
These are couple of screenshots of some micro and panel mount switches.
Both are behaving as expected. The number of bounces and the time they take vary, but without fail they are there, jumping merrily.
However, the tactile button I tested next showed this:
As you can see, it's perfect!
Adjusting time division and countless attempts with other buttons of the same design produced this as the worst case from a bunch of different buttons:
Keep in mind that these are not some super expensive mercury-wetted military grade buttons. They were ordered from ebay for less than peanuts, including the free shipping.
As for the testing method, imagine the boat shaped thingies are the places where I hooked the probe. The buttons and switches were on the breadobard.
So, does anybody see what I'm doing wrong? I doubt the buttons are this good.
These buttons commonly have some conductive lubricant on the "contacts", making them perform somewhat like mercury-wetted contacts.
The problem is twofold - over time the lubricant can dissipate, so that they will actually start to "bounce". Secondly, it can attract moisture and cause leakage resulting in "ciphers", particularly when these buttons are used with resistor ladders to encode to a single analog input.
You have shown only the "make" result. What about the "break"?
Not much inside. The disk is concave and appears to be made of brass. I haven't found much evidence of lubricant. It appears to be dry inside.
Yeah. The sealing is nonexistent. Top is covered with a metal plate and the inside is not protected very well. I can easily see moisture being a problem.
I don't understand the last part. Are you saying I should make a rig to repeatedly press the button to test the reliability of it?
Shpaget:
I can easily see moisture being a problem.
Food. Dust. Oil. Lint.
Sealed ones are nice. In my experience, well worth the extra cost.
Ok, let's take a look!
Excellent work!
The brass disk is designed to invert when a specific force is applied. And then flip back when a lower specific force is applied.
If the disk is nice and springy, it will invert very quickly; it will "slam" into the other contact.
As the button is released, there will be a moment when the force is just a bit too small to hold the disk in an inverted position. The disk will snap back to its idle position very quickly breaking the connection.
This should leave just one source for contact bounce: when the disk inverts does it make a single solid contact with the other side or does it bounce off. For your buttons the disk does appear to make a single solid contact.
I suspect that small ripple in the last Rigol image is the disk moving a bit after making contact. A contact shuffle instead of a contact bounce.
Oh, sorry. I got the impression it had to do something with the "make or break" expression, which made no sense to me.
Anyway, I did some contact break testing and still the button is relatively behaving nicely. I had to do about 15 tests before I got this:
Coding, yeah it's a simple design. Effective, though.
Mark, I wanted to see what is going on so I could devise optimal debouncing methods for these particular buttons and switched I have around and to get a better picture of what happens and when. It's really just about learning and gaining experience (and having fun doing so). I started this thread because the result I got was not what I expected, which is always nice way to learn something new.
Raschemmel, as per your prognosis, a cap does work.
Anyway, I've gotten myself a new toy and I can already see many future hours spent by its side.
Around $430. A bit more than your red tag.
Used oscilloscope market is almost nonexistant here in Croatia, and shipping from anywhere other than China would be more expensive than the scope itself. Add the import taxes and you get 2/3 of the way to this price.
This way I have the warranty and the fuzzy feeling of supporting a local bussiness.
I guess I'll just eat less for the next year or so. :o
