Capacitive Touch Sensors and Lab Gear

[madshaman]

I'm curious as to whether there are any technical reasons why using capacitive sensors on test gear might be a bad idea.

Has anyone experienced any trouble with these in the wild?

[DutchGert]

Gloves, water, physical feel of a button. All factors that are affecting Capacitive Touch

[madshaman]

Thank-you for the reply.  I should be more specific, I'm not looking as much for how a capacitive sensor works or it's limitations; more for issues specific to an electronics lab environment.

Fwiw, normal push buttons don't like water either.

[SeanB]

You might want to have some haptic feedback with them, or a LED to show function, as otherwise you might keep pressing in vain.

[codeboy2k]

yes, haptic would be a good idea.  Do they make a little surface mount mems vibrator?  I'm actually thinking of building my next home-brew test gear with some capacitive buttons laid out on a PCB for the front panel, and overlaid with the Papilo.com self adhesive photo film like robrenz used on his micro-ohmmeter.  Adding haptic would make it more usable, indeed.

https://www.eevblog.com/forum/testgear/ie/msg129618/#msg129618

But  it just seems like laying out a PCB with a dozen or so capacitive touch points and then putting a sticker over it would make an easy DIY front panel with a  professional look.   Could also use holes and reverse mounted LEDs as indicators, even in the middle of the buttons, (or the bottom corner of a square button)

[Harvs]

A couple of years ago I designed a display for my little 4WD.  Just a 2x20 display with a circuit board as the front panel (with route out to view the display) and cap sense buttons using the Atmel Qtouch library.  It worked very well in theory and on the bench.

The reality is though, to be able to use touch buttons you need to be able to support you hand/arm on something.  With phones, you're usually holding it with one hand or supporting your hand on a table.  It can be amazing how uncoordinated you become if you prop a phone up and try using it with one hand without resting your arm on anything.

So you've got about a 25% chance of hitting the wrong button.  Because with a normal button, you move your finger into place, then push. 

It works where you've got a small number of buttons with large distances (like a cooktop.)  But if I look at the 20 buttons crammed on the front of my bench DMM, this would be a complete nightmare as a touch panel, you'd constantly be hitting wrong buttons.

In summary, if you can work with only a couple of buttons, any reasonable number and you need physical feedback.

[codeboy2k]

yeah, good points there.  I think it would take some tweaking to get it right and usable, as you said.  I have a Samsung S3 phone, it has capacitive menu and back buttons on the bezel, if you don't know already.  And whenever I turn the phone into landscape to watch a video, my finger near the phone edge triggers the back button and stops the video.. grrrr.

with a benchtop device, perhaps combined with haptic feedback, two UI parameters could be set and adjustable by the end user: ignore time.. any button press less than this would be ignored (it can default to 500ms) and accept time : any button press longer than this is accepted, default to 750ms.  So you can swipe your fingers over the keyboard or lazily slide over the #4 key while hunting for the #5 key.  If you hit #4 for less than 500ms, no haptic, no beep, nothing happens.. then pause at #5 long enough for the haptic feedback and move to the next key press, if you touch another key while moving, it's not a problem.  The fact that a user can tweak it means that they can adjust it for their own style, because you can never please everyone with one setting. It has to be user configurable. and that means haptic on, off, duration and intensity, as well as beep on, off, duration and frequency I hate when the vendor chooses my beep for me lol

[madshaman]

Thanks guys, all good thoughts and feedback.  I hadn't thought too deeply about button crowding.  I also hadn't consider the tiny amount of hand support that physical buttons give.

Thanks for sharing your personal experience; this is exactly what I was looking for.

One of the main advantages of, and why I like this kind of sensor, is that they're immune to mechanical failure, cheap, simple, and as mentioned, can be put right on the PCB.

The lack of physical support could maybe be solved by putting the sensor areas on a raised besel or something.

The lack of haptic feedback is an annoyance and methods to rectify this all kill the simplicity aspect of the sensor and reintroduce mechanical failure :-(

[codeboy2k]

Gloves, water, physical feel of a button. All factors that are affecting Capacitive Touch

outside gloves , for cold weather, or construction work, definitely won't work.  However, capacitive touch does work for me with latex lab gloves on I've tried it on my phone, no problem.

Also, microchip has an app note for sensing capacitive touch through water, I read the note long ago. Apparently, water actually improves the sensitivity, and you get a stronger reading, which can sometimes lead to false readings, so there is an algorithm in that appnote they describe to help "recalibrate" the baseline, especially when it's wet, and determine finger touches vs water noise.  So water is not a problem.

A lot of industrial designs use membrane switches, these are several layers of mylar or poly sheet sandwiches that you press them to make contact.. the top layer is the printed layer and exposed to the user.  The only problem I find with membrane switches is they require a lot of force , and they wear out/break because of that force. and people just press them harder when they get dirty, and break them more.   Capacitive touch has the ability to replace membrane switches in some cases, but not all. Membranes (and real buttons) have the advantage that they still work while wearing your work/outdoor/protective/construction gloves.

[nctnico]

You can also use audible feedback. I recently designed a 4x4 capacitive touchpad for an outdoor SIP telephone. If you press a key correctly it will beep. The 'buttons' are about 15x20mm.

[codeboy2k]

You can also use audible feedback. I recently designed a 4x4 capacitive touchpad for an outdoor SIP telephone. If you press a key correctly it will beep. The 'buttons' are about 15x20mm.

yes, that's what I was saying in my previous post too.

Also, there was a thread long ago about instrument beeps, and everyone had their favorite tone.  So I think if I ever design a lab instrument for me or for sale, I'd like it to have an adjustable tone so you can pick your own favorite   I also did some beep tests in Audacity in that thread, you can have a look too and see which beep you like. Here's the posting and that previous thread I mentioned:

( Found in one shot via google search: site:eevblog.com inurl:buzzer audacity )

https://www.eevblog.com/forum/projects/what-buzzer-frequency-%28tone%29-would-you-consider-%27unannoying%27-if-at-all/msg221698/#msg221698

[VintageTekFan]

I also believe that capacitive sensors work with a 'high' frequency signal, and I wonder about the amount of transmission.