Hi guys, I am planning and prototyping a case which would require a touch sensitive switch like you find on many modern monitors and cell phones. I want to have the switch hidden behind smoke acrylic, and activated by a touch on the acrylic. I googled many different names for this, and found this guide, on the very forum I frequent: http://forums.bit-tech.net/showthread.php?t=100148 The only problem is that all of the pictures are broke, the part codes are dated, and I don't know enough about circuits to understand the design just from the guide. If anybody could give me a link, or a quick explanation to how I can make this work, it would be greatly appreciated.
The first problem I see is acrylic is an insulator and all touch sensitive buttons work on a change in capacitance. So you'd need your material to conduct.
You're wrong there Pedro, in some capacitative touch switches (certainly in the case of the qprox kit) the electrode is mounted behind a dielectric surface, there's no need to touch the electrode directly. It works like this, the electrode forms one plate of a capacitor, when your finger isn't present the capacitor has no second plate and capacitance is very low. When you put your finger to the button your body forms the other side of the capacitor and capacitance is increased, it's this change in capacitance that the sensor detects. If you touched the electrode directly you would essentially be shorting the two plates of the cap and the capacitance would stay low. In reality it's a bit more complicated than that because the human body can itself act as a capacitor. Capacitative touchscreens work in a similar way which is why they wont work with a stylus unless it's been specially designed to be used with a capacitative touchscreen hence why people took to using sausages as styli for their iPhones. You should be able to find a replacement for the chip used in the guide, Qprox kit has become quite popular so I'm sure there are other, more up to date, guides on the 'net. Moriquendi
You learn something new every day. I was thinking about a design with two or four thin pins of metal embedded in the acrylic connected to a Darlington pair, but I guess it's kinda redundant now.
If you were looking for a design that you could construct from discrete components then you're probably on the right track and using resistance rather than capacitance would probably be much easier. The Qprox stuff is designed so you can mount it behind an unbroken front panel for easy cleaning and so on. Moriquendi
There is another way, but it's not exactly easy. I have no formal electronics training (unless you count a degree from the University of Life and 11 GCSEs from the School of Hard Knocks), so I learn by doing. Hence the odd method and my previous confusion (this would need a conductive button). You set up two resistor-capacitor timer circuits and an oscillator to periodically charge them. One of the capacitors you link to the button. You use an op-amp, transistor and a diode so that when the voltage on the button capacitor is higher than the reference (i.e. the capacitance has increased and the time period of discharge has increased), something is switched. Like I said, really kinda crazy, but you can get it to work. It's easier to do on a microcontroller using the same method. On second thoughts. just use the QT113!
I wasn't having a dig at you just pointing out that capacitance wouldn't require a conductive surface. I suspect that any analogue system would be very difficult to tune and would probably drift pretty quickly too, especially when you look at the tolerances on typical components. The Qprox chip is the way to go if you want a reliable system. Moriquendi
I didn't take it that way at all. If I didn't get anything wrong, I wouldn't have learned anything at all. You're bang on when you say QT would be a better implementation (I've had a good track record of getting free samples from Atmel, so it could be done for next to no outlay, too!) I just wanted to share the way I set it up as a weekend project. I didn't have a dedicated IC, so that's the way I did it with what I had to hand. Tuning wasn't too bad, as I used a couple of trimmer caps and a Schmitt trigger style affair. I can't tell you about reliability; I only had the circuit breadboarded for a day or two.
Thanks for the replies, I do want an unbroken design, but the electrical competence and knowledge of parts is the key. Do you know how to wire the qprox? I can solder fine, but soldering the right part in the right place is key. Ive been thinking about this, or any monitor with a similar switch for that matter, but I would bet a shiny nickle (or whatever the heck you English folk use ) that the switch and controller is integrated into another pcb, not a standalone part. Thanks for the replies, if anyone has any more help, ideas, or guidance, please feel free. EDIT: Or where to purchase a qprox controller for that matter. They seem more oriented towards large scale manufacturing. SPNKR
I did have a touch sensitive switch working through 1/8" acrylic but I was not very good at the electronics so it did not work completely flawless. It did work tho. But just not as good as I planned. If you could understand the electronic part of it im sure you could do it. It was done by using chips by qtouch I think. http://www.atmel.com/products/overview_touch.asp
certain versions are hard to come by. it was 2 or 3 years ago when i bought mine from: http://www.saelig.com/ digikey or newark maybe also have some. they are QTXXX . Im not sure of the model number. There are so many different versions depending on your application. THey are pretty cheap tho. Mine were only a few bucks a piece.
Here you go. Touchsensitive switches for mounting behind glass/plexi/ceramic... http://www.rwcomponents.com/Touch Sensitive Switches.htm
POW!!! Thanks JRS, that seems to be exactly what I'm looking for, you get a cookie. Thanks for the help, this issue is resolved. Now on to the acrylic... EDIT: Oh yeah, plus rep for y'all
