Incremental Rotary Encoders
Incremental rotary encoders are either something you know about, or you don’t. This is, of course, true of everything, but more-so in the case of these little scamps.
Most people come into contact with these rotary encoding rascals many times each day without even realizing it or thinking about it. Not-so-long ago, for example, the volume control on your car radio was implemented using a potentiometer. This had a smooth motion and a limited rotational range of say 270 degrees. Today, this function will almost invariably have been replaced with an incremental rotary encoder that has a tactile “click-click-click” feel as it turns, and that has no limit on how far you can rotate it clockwise or widdershins.
The cheap-and-cheerful versions of these devices, which are proliferating everywhere, are mechanical in nature. Although conceptually simple, their internal mechanisms vary wildly, while their switch bounce characteristics can make your eyes water.
In this MicroTalk, we will take a short, sharp plunge into the world of incremental rotary encoders, answering all the questions you never even thought to ask.
Why does a single detent (one click) on a cheap mechanical incremental encoder sometimes produce multiple counts or even a reverse step?
I will let the conference organizers know -- in the meantime, you can access them here:
https://vimeo.com/821402357/ca62284d57
https://vimeo.com/818121190/d936047e1f
Don't call me "Shortly" :-)
I look forward to hearing what you think about them :-)
As long as you don't mind me calling you "Shirley". ;) Leslie Neilsen would be proud. ;)
The scary think is that there are many young folks today who wouldn't get the reference :-(
Thanks for the quick reply! I'll be watching them shortly :)
I live to serve LOL :-)
Great microtalk with an excellent point about the switch bounce. One interesting thing that occurs to me is that if only 1 side bounces (or if they both bounce, but not at the same time as seen by the oscope), that it may be somewhat self-correcting. The positive edge of the bounce looks like it counts "up" (arbitrarily chosen) one, followed by the negative edge counting "down" one. This could happen a number of times depending on the mechanical bounce before settling out at the "correct" value. But if both phases are bouncing simultaneously, then you can get false counts depending on the respective phase and width of each channel's bounce waveform. This may be why we don't see it occurring most of time when we cycle the encoder by one detent. Also, this assumes that the edges are detected/decoded in hardware versus just having the signal be sampled by firmware. Of course, firmware sampling is probably the more likely implementation since it is cheaper negating the "benefits" I posited above :(
Hi Max, just letting you know .... I haven't forgotten outstanding! I'll feedback via another channel as opposed to this semi-public one. All went well I thought, I've sent congrats to Jacob & Stephane already even though I've endless videos & Q&As and Q&Cs to go through. PS: I'm getting touchpad bounce currently ... any suggestions LOL !!
Hi Bob -- my email is max@clivemaxfield.com -- cheers -- Max (1/2 man, 1/2 beast, 1/2 wit)
By my calculation .... or are you talking spins LOL
Hi Max, I'll feedback in full later. Re: microtalks Q&A - Intend to ... trying to lubricate myself through the Rust workshop so I'm a bit stuck right now !
Hi Bob -- I could do with a bit of lubrication myself -- I'm 100% loaded at the moment (and not in a good way with alcohol). It will be great if you do get to join the Q&A, but no pressure :-) Also, if you get a spare minute, it would be wonderful if you could take a look at the high-voltage switch debounce module survey we just created https://freeonlinesurveys.com/s/nSFeF4nJ -- cheers, Max
Have I got space for even more data-sheets ... this time for non-contact sensors!? PS: Above 'these other types'.
There's always room for more data sheets LOL I'm not sure to what what your "PS: Above 'these other types'." comment refers (said Max, sadly)
Oh, one other thing, room for data-sheets ... not if they're in hardcopy form !!!
The problem is that you know in your heart of hearts that if you throw any away to make room, those are the ones you will need the very next day LOL
Will you be joining us for the Q&A on these micro-talks (starting at 11:30am Central Time)?
In the original text, ignoring the grammatical errors, I had written 'these are types' ... my mistake! What confused me briefly was that I expected this amendment to appear below the original comment ... Nop! PS: Difficult to completely move beyond electro-mechanical devices especially for high power stuff, hence your valid business case ... Logi... ;<)
We're contemplating offering a module with a high-voltage side (say 24V) to be connected to remotely mounted switches and relays optically isolated from a low-voltage side (2.5 to 5.5V) driving the logic. In addition to removing switch bounce, the module would also protect against over and under voltage spikes and current surges on the high-voltage side. Do you think there would be a market for this?
Electro-mechanical - switch bounce most definitely (progressively getting worse with age) however, for other methods (optical, capacitive. magnetic) the word 'bounce' I think is inappropriate. Interesting what the characteristic behaviour for these are types would be, hysteresis certainly comes to mind hence threshold detection. As I would agree, interesting! The things we miss but ultimately, must overcome!
Hi Bob -- I don't think the other optical, capacitive, and magnetic encoders will exhibit switch bounce, but I haven't experimented with them -- I stuck to mechanical devices because these are cheap and common and they do have bounce problems (which is one of the things we focus on at LogiSwitch.com).
When attempting to watch this video (6/27/23), I get a "Sorry This video does not exist." error. :( Same with Clive's other video on BCD.