The Problem with Capacitive Screens

I thought it'd be worth getting a discussion going about this since it's probably something many of us have thought about at one point or another.
The reason you can't just go implanting a touch screen as seen in certain movies (such as Total Recall), is that a capacitive touch screen works by measuring minuscule changes in a small electrical field generated by the screen, this determines where over the screen a finger (or other similar object) is located, capacitive buttons (as are becoming more common on set top boxes, DVD players, and TVs) work in a similar way to how an individual pixel on a capacitive "digitizer" built into a smartphone does.

The problem comes in that, even if calibrated properly to account for interference generated by flesh between the screen and a finger that may touch the skin above the screen, it would likely be very inaccurate, not able to determine very precisely where you were touching. That is, if it worked at all.

My solution to the potential problem is to measure nerve impulses triggered by contact on the surface of the skin.

My question (mostly to drive some discussion) is if anyone happens to know what it would take to measure said nerve input, obviously a single device implanted in a more central location that can monitor as many nerves as possible all at once, is preferable. However, also more difficult. I admit that I've not done extensive research on technology to measure nerve activity, but I thought I'd bring up the idea before necessarily knowing everything.


  • you could also do a thin layer of silicone on top of the display to keep normal skin slightly away from the digitizer, and then when you push you are depressing the silicone making the skin now closer to the screen so it can read the location. This is a very popular way that DIY touch tables work as its accurate and simple to do. The number one issue I see is of course power. and the image would suck. The screen responding to touch is not as issue as I see it.
  • @Benbeezy, at that point why not just use a resistive touch screen instead? Maybe you could do an eink with a glowy background like one of the newest kindles. I know they sell various eink displays in adafruit which look pretty cool.
  • @ChrisBot Because you could still use the multi touch functions.
  • My thoughts were that a somewhat uniquely designed, monochrome, e-ink screen with an LED lighting panel would be best. The lighting source itself is he most likely part to fail in most screens so going I watch a dedicated one could prolong the potential lifespan, that said, OLED screens have an upper limit on practical thickness of like 4 mm. Most are much thinner, the ability to individually light up separate pixels is good too. OLED a screens have been known to last only a few years before they begin to experience issues, bad pixels, failing completely, etc... So that's less than optimal, just because I think that implants should be designed to stay around for at least ten years before NEEDING to be replaced/removed. And I do still think that some sort of nerve integration is best for detecting touch. Resistive touch screens are generally thick and inherently rigid (not flexible), and not great for implanting in a curved arm.
  • OLED can be put on a flexible substrate and still work even with touch, these are the only displays that you can do that with, they also glow themselves and need no extra light from behind. this is why they are so damn thin. The only issues with OLED is burn in and that hasn't been an issue in a very long time. I have a OLED tv that I have had for about 5 years with no issues and I have phones that are even older that still have no issues to speak of. another good thing about oled is that it is very low power and when you display black you just turn off that section completely. This is how phones get away with turning the screen on when you pick it up in with a black background, its so low power that it doesn't effect battery life.

    Not true that it adds thickness. You could make the entire thing work just fine with multiple touch points with a .5mm thick silicone layer, this also flexes just fine and will still work. So this idea still stands as it can be flexible with the screen and the touch sensor. 

    Nerve integration sounds super cool but do you have any fucking idea how to do that? yeah, neither does anyone else. Touch is still the most practical way to do this as its a proven technology. I just with a had a few hundred K to actually build some of this stuff out. I bet it could be done in under a years time if the resources were available.
  • It was just a concept. I admit to having no intention of attempting to produce this nerve-based touch system myself. Just thought it would be a good idea, but it's probably a few years of research, a few years of government regulations, and another few years of consumer product development away from being feasible.

    As a side note, I'd do a non-touch screen implanted and some other means of controlling it any time. If it had practical purpose of course. I suppose the main problem is still powering it.
  • To be considered, I'm using this nerve integration system may be adding an unnecessary step. While you could probably read mechanosensor-related nerve activity through some means, you would still need a means to interpret and use that information. I feel that adding all of these steps together makes it kind of impractical. However, developing or improving upon current ideas relating to signal coding, or even finding ways to implant the relevant sensors would probably be really beneficial. Plus, they may be able to eventually play into the idea of an integrated device. Hopefully I'm not too off base, I haven't done much research on this. Let me know if this doesn't make sense.
  • Well why do you want to start with a screen? I think a trackpad in my arm would be pretty impressive already, and it would be considerably easier to produce if you are making it in the form of a capacitative implant. We already have bluetooth implants that could be used to relay the input to a computer or bluetooth mouse friendly device.
Sign In or Register to comment.