A quick guide to Hysteresis

[Shadowheaven]

Since some people asked me how my set-reset works, here's a quick guide to hysteresis:

Huh? Hysteresis? What's that?
Hysteresis is the phenomenon for which the response of a system to an external signal depends on the response previously given to the same signal; in simpler words, the relationship between input and output depends on the current status of the device (may sound confusing, but keep reading and it will become clearer).


Erm... What does it have to do with LBP? How does it work?
Hysteresis is integral and substantive part of magnetic switches and sensor switches, and carries out an important role in their mechanics, since it grants a very high degree of stability; basically, we should consider this

being something very close to this

The mag switch turns ON when a key gets in the yellow zone, turns OFF when it gets out of the blue zone and keeps its status when the key is in the middle


How can it be used?
Hysteresis is one of the most efficient and less thermo consuming status memorization; here are a few examples:

Rtm223's resettable permanent switch

Basic and very efficient: the piston contracts, activating the mag key; when it extends, the key stands in the hysteresis zone (between the yellow and blue areas), and the mag switch is still ON - the piston can be set to flipper (in) or no flipping, and to reset the device you simply have to select and deselect the magnetic key (when a key is selected with the popit, the switches ignore it)

My set-reset

The SET input contracts the winch, while the RESET input extends the piston. Pretty simple, huh?

Permanent OR/AND
Instead of linking an OR or an AND device to a permanent switch, you can include it in the OR/AND like this:

Talk about saving thermo! Of course, you can use the same method for other devices.

Feel free to correct eventual grammatical errors

[Sehven•]

Wow, no responses? This is actually pretty informative. I can't imagine I'll ever use it myself (I'm totally fine with dissolve based p-switches) but I would think it'd get a bit more attention than this.

Quick question. I've always assumed that the blue zone in your picture is the width of the circle on the mag key, but I've never really tested that theory. Do you know if that's correct? Based on this assumption, when I want switches with very fine-tuned triggering, I shrink the mag key to tiny size (make the object with the key much bigger than you actually need it, place the key as small as you can, then shrink the object: it'll shrink the key with it). In particular I use it on tilt sensors because I want very precise responses. If that is true, then you could use the opposite to your advantage as well: make the key big and there will it will have to tilt quite a bit to trigger but then it'll have to untilt quite a bit to de-trigger. That could be especially useful in stabilizing flying vehicles since the stabilizers tend to cut off before the vehicle has actually reached a stable orientation with normal sized keys.

Why am I asking about this? Off to test it!

[rtm223]

I'm pretty sure the key acts as a single point for the sake of activation. As far as I can tell it does and would make more sense for it to be that way, from a programming POV.

The hysteresis zone seems to be appoximately the blurry white band on the radius graphic, and does increase with radius. Can't say for sure if it's proportional to the radius or not, but if I had to hazard a guess I'd say yes.

[ladylyn1]

Is the method secure? By this i realise the hysterisis zone is pretty small for relatively small radii and so is it possible for the key to slip out of the zone? I use the sliding mag key method atm (given by the logic pack) but this looks a much more thermo efficient method

[rtm223]

I have no issues with it, I've used (and had to work around) hysteresis many times and it seems completely stable and predictable.

The methods above are pretty efficient in an "individual switch" scenario, but they lock you out of a lot of techniques that would increase efficiency further, so it's very much swings and roundabouts on the efficiency issue.

[Sehven•]

I'm pretty sure the key acts as a single point for the sake of activation.

Yeah. I feel stupid now. I'm always going to the trouble of shrinking keys ridiculously small and it was for nothing. I guess when I did it before, the whole contraption, including the radius of the switch was smaller: hence the perception that it was working. Still, knowing this I can configure things like tilt sensors to use switches with small radii. Does anybody happen to know if the hysteresis zone is the same when coming at an angled mag switch from the side, rather than the rim of its detection area?

[Shadowheaven]

Does anybody happen to know if the hysteresis zone is the same when coming at an angled mag switch from the side, rather than the rim of its detection area?

You mean if hysteresis is applied to all of the perimeter of the detection area of a switch? If so, the answer is yes - I used this principle in my spring based XOR (please don't laugh at it xD)

[srgt_poptart]

Correct me if I'm wrong, but isn't the hysteresis indicated by the small white area on the outside edge of the switch? If so, that's actually a pretty small area. I'm also gonna guess the hysteresis area gets larger when the proximity of the mag switch gets bigger.

[rtm223]

Correct me if I'm wrong, but isn't the hysteresis indicated by the small white area on the outside edge of the switch? If so, that's actually a pretty small area. I'm also gonna guess the hysteresis area gets larger when the proximity of the mag switch gets bigger.

*looks further up page*

The hysteresis zone seems to be appoximately the blurry white band on the radius graphic, and does increase with radius. Can't say for sure if it's proportional to the radius or not, but if I had to hazard a guess I'd say yes.

It is quite a small area, but certainly big enough to cause issues / be of use.