Freescale's Xtrinsic 3.0 Software Lets Touchscreens Overcome Water
The TWRPI Plug-In Starter Kit comes with two touch panels for testing of the Xtrinsic 3.0 software algorithms for touch interfaces. Configurable rotary, slider, and keypad decoders, and pre-exiiting application code and HMI functionality, comes free.
If this technology were to become more widespread, this would be a super-helpful breakthrough to the use of touchscreens. Who hasn't spilled a bit of beverage on their device and experienced the pain of trying to use it afterwards? Interesting development to keep an eye on.
Sounds like a different way to achieve one measure of ruggedization in systems traditionally built with various methods for ruggedizing, i.e., medical and industrial ones. I agree with Cabe, that consumer systems could also benefit from this. In fact, I keep wondering why consumer systems don't take advantage of more ruggedization strategies that already exist for industrial electronics. I know it costs more--but I also know my older computers were a lot hardier and expected to last longer.
Good point, Elizabeth. Virtually every computer user has spilled something on a keyboard. Until recently, I always kept a small netbook computer in the kitchen to check morning news stories, but the motherboard is now dead, thanks to an unknown source of moisture.
I once had to replace the keyboard, trackpad and logic board of my Mac just because of a minor water spill! So I speak from experience about this. Obviously, this wasn't a touchscreen issue, but still, perhaps this could help make other types of computer interfaces more waterproof as well. It mystified me at the time that a bit of water could kill a computer so thoroughly! I thought they were sturdier than that. Then again, I had this happen a second time (duh!) and after letting my Mac dry out, it started working OK again (except the trackpad...that is toast).
Cabe, this is an interesting update. I have used the previous versions of Xtrinsic sensors. This seems to be a very important upgrade. Freescale, and many other companies, have been building more intelligence into their sensors, thus making it easier to bring rich functionality to many types of products more quickly.
The moisture problem is one more reason to avoid using a touch screen in missionn critical systems. Of course, that only applies to systems where reliability and availability are far more important than low cost. Of course good pushbuttons cost a bit more, and they are not nearly so flexible, but it is often the case that there is only one chance to get the input command right, and getting the command right is way more critical than saving three cents on a button, or having some "neat" feature that does not add any benefit except for being "cool". Possibly I am being way too serious, but from where I stand most of the touch screen applications look a whole lot like toys. Not childrens toys, for sure, but they look like toys all the same. That is why none of the machine control systems that I have created use touch screens. None of those machines were toys.
I did see an invisible automotive keyless entry touch screen that worked very well even in a simulated heavy rainstorm. It may be that there was some heavy software applied, but I did not think so, and nobody claimed that there was software involved. I think it was in early or mid 2008.
Everything is touch screen these days. If the touch applications look toy like, that is because touch interfaces are still an immature area. Although Apple has kept its same look, the way it operates has changed. Android continuously adds features and navigation options. Now, most prior non-desktop navigations buttons are now on screen. Windows Phone 8 brought a radical departure from the icon based HMI. Even more players will come and add new gestures and HMIs, Blackberry and Ubuntu come to mind first.
I remember a story I heard about an engineer commenting on expanded ram. He said, "why would you need more than 640k?" Evolve, embrace, or fall to the wayside.
Cabe, that is exactly my point, which is that in many applications the touch function is just sort of stuck in, not well thought out ast all. Those applications would be far better served with actual buttons. The 17 cent touchscreen is not reliable and it does not last long, either. My point is that it would be much better to do the job right, instead of using some new technology that does not work very well.
Have you ever made a HMI before? I have made a few in the past, and I was faced with the challenge of making a simple and intuitive system. I have to tell you, it is difficult. What made sense to me, was not clear to others. After the attempts to make a complete system, I switch to the one button approach. "Hit it, and it goes" methodology.
An interesting HMI that I created a while back used an alph-numeric display with several buttons below. The function of each button depended on the message displayed. What made the interface unique is that when the system was functioning correctly there were several messages that went by so fast that they were not noticed, but when mechanical parts of the system would hang up or get stuck, there woulkd be a message asking them to wait for that function. A form of diagnosics that did not need any branching logic, it wound up being very reliable.
The buttons below the screen served the similar purpose as tough screen buttons, but they were waterproof and did have a nice click feel when operated. Sot of "haptics for free"
Do you think you could design something better than what is currently available in the touch market? If so, perhaps you should work on that. Produce a concept, flow charts, etc. If patented, you may be able to get ahead of the big contenders. Then, sell the design. Just a thought.
I really don't see that anything better needs to be designed. For systems that are mission critical, and those where reliability trumps following the current fad, non-touch systems have been much better for quite a few years. One more thing is that the off-screen buttons use less power to operate, and incur less circuit complexity, making them more reliable in a very fundamental way. Of course, that atttitude would not add anything to the profits that are obtained from selling touch screen systems.
Cabe, For starters the interface to a rugged pushbutton is more robust than the interface to a touchscreen would normally be. Next, buttons in a row below a screen, used as smart buttons, which change function depending on their screen label, are often very durable. The one button that I have seen fail had been accidentally filled with oil and metal particles after it was installed, when holes were drilled in the panel abobe it without any precautions being taken. But that same model of button has survived mud and floods and being hosed off to clean it. But those buttons do cost a few dollars each. The seven cent buttons are not so robust and durable.
You are probably right. The Optimus Maximus (and Popularis) keyboard(s) featured OLED screens in each key. The function label of each key could them be changed as the application demanded. It does look useful. Not sure how popular the $1000 USD keyboard was. I am sure it wasn't a happy time for buyers when they dumped a Diet Pepsi on the keys... you know it happened.
Also see the Optimus Mini-six, a 6 spot keypad for $700 USD.
For industrial control applications, or even a simple assembly line, that machine can go almost 24/7 without a break. But what happens when the task is a little more complex? That’s where the “smart” machine would come in. The smart machine is one that has some simple (or complex in some cases) processing capability to be able to adapt to changing conditions. Such machines are suited for a host of applications, including automotive, aerospace, defense, medical, computers and electronics, telecommunications, consumer goods, and so on. This discussion will examine what’s possible with smart machines, and what tradeoffs need to be made to implement such a solution.