Hello everyone,
I'm working on a prototpe for a project and I have several questions. I'm a complete beginner in electronics even if I have acquired some basis in my scolarity.


Here's the blueprint of the project.
The orange surface is the one that should be touch sensitive.
It needs to be able to detect swipe and touch as well as at least 2 simultaneous touches.
It should also be transparent (because this area will be glass) and working with gloves (for example) and in wet conditions (can work in wet conditions like bathroom but also in any rooms of a house).
If possible, Id prefer to have nothing visible on this sides:

I have done a lot of research, I think the technology that should be used to make this is the projected capacitive. But, I cannot find anything I can buy that respect those conditions or that is not a display (I don't need a display I just need this surface to be tactile). If you know something I'll take it.

The other solution is to build my own sensor. I think this is suitable, but again, I'm a complete beginner, I know the theory on how it should work but I have no idea on what I have to buy and how I could actually build it.

Last question, all the electronics should be in the grey part with the IO on the back (the screenshot is a front view). For this project I'm using an ESP32 Pico D4 as a microcontroller (since I also want to have those controls to be possible on mobile for example). But, since I have never done such a thing before, I have no Idea on how to drive this. In fact, I am an advance Python programmer, I'm learning the C++ for this project, but I don't know what output I'll have and how to translate them into an X and Y pos variable. I don't even know if I'll need an other microcontroller for this.

Thank you for your help,
I'll be very please to answer any question you may ask.
Have a nice day,
Louis Veran

 

If I may offer some advice, get to www.uspto.gov (U.S. Patent Office) and do a patent search for inventions that use the word "touch" in the patent name or body or the text and I think you will probably find multi-touch information if you include the names "Krah" and "Apple" in the search.

What you are proposing is a very complex task that could probably consume all of your waking hours for years to come.

Microchip has some ready-made solutions such as the one below. Simply search their site for "touch".
https://www.microchip.com/en-us/solutions/machine-learning/smart-human-machine-interface

 

You can get glass overlay touchscreens: not monitors, just transparent touch pads meant to go over a monitor.

https://www.elotouch.com/touchscreen-components.html

As far as getting one with those specific and non-rectangular dimensions, that's obviously a custom order deal, and probably nobody is going to even talk to you unless you're ordering 10k units per month.

My recommendation is that you lose the wacky 6-sided shape and pick a new shape (4 sided) whose dimensions fit something you can actually get.

 

Here is all you need to know. Get a "capacitance-to-ground" panel to handle gloved hands and read the "flat front surface - industrial design" section in the left column on page 6 to understand how to achieve the visual effect you are after.
You can get the shape you want by printing a graphic film (PET or polycarbonate) with UV-cure acrylic inks with your black bezel and laminate to glass with optically clear adhesive to get the shape you want and just ignore the touch area you don't want in software.

https://multimedia.3m.com/mws/media/788463O/tech-brief-projected-capacitive-technology.pdf

 

If I may offer some advice, get to www.uspto.gov (U.S. Patent Office) and do a patent search for inventions that use the word "touch" in the patent name or body or the text and I think you will probably find multi-touch information if you include the names "Krah" and "Apple" in the search.

What you are proposing is a very complex task that could probably consume all of your waking hours for years to come.

Microchip has some ready-made solutions such as the one below. Simply search their site for "touch".
https://www.microchip.com/en-us/solutions/machine-learning/smart-human-machine-interface

Hi, thank you for your answer,
I don't understand why it is hard, I'm just looking for the technology that is used in phones for example, this already exist, has multitouch, are waterproof and some are even working with gloves. That's why I don't understand why it is so hard to find it.
And thank you for the link to microchip, I'll see if I find anything useful!
Thank you for your help

 

You can get glass overlay touchscreens: not monitors, just transparent touch pads meant to go over a monitor.

https://www.elotouch.com/touchscreen-components.html

As far as getting one with those specific and non-rectangular dimensions, that's obviously a custom order deal, and probably nobody is going to even talk to you unless you're ordering 10k units per month.

My recommendation is that you lose the wacky 6-sided shape and pick a new shape (4 sided) whose dimensions fit something you can actually get.

Thank you for your answer,
You're right, I'm working on the prototype I don't need to have the exact shape and size I need at this point. Thank you for the link, I'll try to buy one from them and see if it works as I want it to. Afterwards I might be able to buy enough to make a custom command we'll see!
Thank you for your help

 

Here is all you need to know. Get a "capacitance-to-ground" panel to handle gloved hands and read the "flat front surface - industrial design" section in the left column on page 6 to understand how to achieve the visual effect you are after.
You can get the shape you want by printing a graphic film (PET or polycarbonate) with UV-cure acrylic inks with your black bezel and laminate to glass with optically clear adhesive to get the shape you want and just ignore the touch area you don't want in software.

https://multimedia.3m.com/mws/media/788463O/tech-brief-projected-capacitive-technology.pdf

Wow that's actually a really helpful answer, if it works it might be what I need! I'll have a look and try to buy what I need to do this.
I hope there is some information about the software side in your article because as I said, I have no idea on how to drive such a thing.
Thank you for your answer!

 

There are many little things that can sneak up on you and wreck havoc with a touch sensor, especially when you get into more than one button! It is a long and painful learning experience (have the T-shirt). Those who point your toward ready-made solutions are giving you good advice.

 

There are many little things that can sneak up on you and wreck havoc with a touch sensor, especially when you get into more than one button! It is a long and painful learning experience (have the T-shirt). Those who point your toward ready-made solutions are giving you good advice.

I understand, what you were saying is that making one myself would be extremely hard. You're right, I should look for something prebuilt as other have said. I think that as I said in a previous answer, since I'm working on a prototype I should just order something like a 7inch rectangular pannel even if it's not what I was initially looking for. And then when I have something 100% working maybe see with a company if I can order custom ones.
Thank you for your answer!

 

The easiest thing would be using a raspberry pi display (the official display of raspberry pi. It looks a lot like your design with the black bezel. It has a built-in capacitive touch sensor, frameless and overall a nice look.
You can use it as-is or try placing a printed graphic on it and simply using the display as a backlight for the printed film.

If you use it as-is, you will create and display a .png image to create your 6-sided graphic.

if you want to print a PET film (FedEx Office has great printing service to do this. Print the mirror image of your design on the back side of your film so you'll have a beautiful gloss image on the front). Bonding the printed film is a challenge - I don't recommend this bonded film option because of the challenge of hand-bonding the film onto the display. Any air bubbles are very obvious and air gaps hinder the performance of the cap touch display

https://www.amazon.com/Raspberry-Pi-7-Touchscreen-Display/dp/B0153R2A9I

 

The easiest thing would be using a raspberry pi display (the official display of raspberry pi. It looks a lot like your design with the black bezel. It has a built-in capacitive touch sensor, frameless and overall a nice look.
You can use it as-is or try placing a printed graphic on it and simply using the display as a backlight for the printed film.

If you use it as-is, you will create and display a .png image to create your 6-sided graphic.

if you want to print a PET film (FedEx Office has great printing service to do this. Print the mirror image of your design on the back side of your film so you'll have a beautiful gloss image on the front). Bonding the printed film is a challenge - I don't recommend this bonded film option because of the challenge of hand-bonding the film onto the display. Any air bubbles are very obvious and air gaps hinder the performance of the cap touch display

https://www.amazon.com/Raspberry-Pi-7-Touchscreen-Display/dp/B0153R2A9I

View attachment 264873

Hi,
If I understand correctly, what you're sending me to is a display. And actually, I'm just trying to build a smart controller, with very easy, intuitive gestures. The thing I'm actually looking for is the kind of tech that is below the screen in smartphone that acquires the X an Y pos. The same thing used (I guess) in trackpads on laptops etc... To be honest, what I'm building is closer to a transparent trackpad than a smartphone it could be a better example.
The screenshot in the original post is supposed to be the entire product, orange part being the "transparent trackpad" and grey part would be where the μc, the antenna, PCB, and all the electronics would be, covered by a plastic protection (linked image of the product on an other angle, everything is orange but you can see the two parts).

I'm not sure if that clarifies it. In your first answer I thought you told me that it was possible to buy/build a tactile tape or something to just apply on the surface and then that's it. It was a bit too good to be true sadly. I'll still read the entire document just to see if I can find anything helpful.
If you have an idea or any other questions don't hesitate.
Thank you very much for your help

 

You are being deluded into believing that because something is in common usage that it must be easy to implement.

Take voice recognition on your smart phone for example. It took billions of dollars and years of research and development to get to the stage it is.

Guesture recognition on a touchscreen is not child’s play.

 

Hi,
If I understand correctly, what you're sending me to is a display. And actually, I'm just trying to build a smart controller, with very easy, intuitive gestures. The thing I'm actually looking for is the kind of tech that is below the screen in smartphone that acquires the X an Y pos. The same thing used (I guess) in trackpads on laptops etc... To be honest, what I'm building is closer to a transparent trackpad than a smartphone it could be a better example.
The screenshot in the original post is supposed to be the entire product, orange part being the "transparent trackpad" and grey part would be where the μc, the antenna, PCB, and all the electronics would be, covered by a plastic protection (linked image of the product on an other angle, everything is orange but you can see the two parts).

I'm not sure if that clarifies it. In your first answer I thought you told me that it was possible to buy/build a tactile tape or something to just apply on the surface and then that's it. It was a bit too good to be true sadly. I'll still read the entire document just to see if I can find anything helpful.
If you have an idea or any other questions don't hesitate.
Thank you very much for your help

What I'm saying is, all the touch libraries for the touch layer of thst pad already exist in Python so you can quickly create a mock-up of what you want to do with a touch feature and, as a bonus, you can mock-up various graphic effects by using the display as a selective backlight or as the graphic generator itself - all on a seamless, flat, glass surface - for $100 plus a $35 Python-enabled, networkable, computer (Raspberry pi 4) to assist with development.

after you get the aesthetic look you want, the swipe/touch interface you want, the customer/user feedback you want - you can start investing real money into a low-cost touch element only for large volume manufacture.

 

You are being deluded into believing that because something is in common usage that it must be easy to implement.

Take voice recognition on your smart phone for example. It took billions of dollars and years of research and development to get to the stage it is.

Guesture recognition on a touchscreen is not child’s play.

Hi,
I'm not saying it's easy to make. That's for this reason that I am here. But what makes me think that it is not impossible is that I remember a few years ago in highschool we built something to get X and Y position using only two foils of aluminium a plastic layer and a few wires. So I was guessing that a more precise and modern solution could be a little harder but not by much. And I don't even need something as precise as a smartphone but like you said, buttonless, tactile areas are everywhere now.
Thank you for your answer tho, it was very relevant

 

What I'm saying is, all the touch libraries for the touch layer of thst pad already exist in Python so you can quickly create a mock-up of what you want to do with a touch feature and, as a bonus, you can mock-up various graphic effects by using the display as a selective backlight or as the graphic generator itself - all on a seamless, flat, glass surface - for $100 plus a $35 Python-enabled, networkable, computer (Raspberry pi 4) to assist with development.

after you get the aesthetic look you want, the swipe/touch interface you want, the customer/user feedback you want - you can start investing real money into a low-cost touch element only for large volume manufacture.

Oh, I understand, you're right this is also a solution. But shouldn't I for the other parts except for the tactile technology, use the parts I already want to use in the final product?
Because if I code something for the Raspy4 with some kind of prebuilt kit that won't teach me how to wire or to drive the real circuit. I just mean that for example the ESP32 maybe doesn't have the same I/O or isn't maybe as easy as the Raspy4.
This morning I thought about something. I'm going to change my smartphone soon. Do you think I could use the old screen with the tactile sensor to make a prototype?
That's just an idea but this could help me figure out the wiring and save some money for the wiring.
Thank you very much for your help, you are very invested I appreciate it a lot

 

The touchscreen is only the sensor.
All the fancy and difficult stuff is in the software.

 

But what makes me think that it is not impossible is that I remember a few years ago in highschool we built something to get X and Y position using only two foils of aluminium a plastic layer and a few wires. So I was guessing that a more precise and modern solution could be a little harder but not by much.

I am curious why you think there is little difference between a transparent, water tolerant touchscreen and your aluminum foil demo. This seems like saying, “I made a lemon battery, so making a modern car battery can’t be that much harder”.

It just doesn’t seem sensible.

 

I am curious why you think there is little difference between a transparent, water tolerant touchscreen and your aluminum foil demo. This seems like saying, “I made a lemon battery, so making a modern car battery can’t be that much harder”.

It just doesn’t seem sensible.

Tbh, the water tolerent part isn't the most important I wouldn't mind if it wasn't.. and you're right there's a big gap. I might have not been precise enough, what I tried to say is that I don't need a lot more than this aluminium demo. As I said, I don't need something as performant as a smartphone screen. But I'm using this example because it is very frustrating seeing that it exist everywhere and it is possible. And I'm counsious I would not be able to build something like this myself that's why I was asking for prebuilt solutions. I don't really understand the difference between what I want to build and a smartphone or trackpad touch sensor. It should even be easier..
Thank you for your answer that does brighten your point and your first answer was also very interesting in the way that it makes me see it from a different angle.

When I first started this project I didn't think that the tactile part was going to be the hardest thing to find

 

Tbh, the water tolerent part isn't the most important I wouldn't mind if it wasn't.. and you're right there's a big gap. I might have not been precise enough, what I tried to say is that I don't need a lot more than this aluminium demo. As I said, I don't need something as performant as a smartphone screen. But I'm using this example because it is very frustrating seeing that it exist everywhere and it is possible. And I'm counsious I would not be able to build something like this myself that's why I was asking for prebuilt solutions. I don't really understand the difference between what I want to build and a smartphone or trackpad touch sensor. It should even be easier..
Thank you for your answer that does brighten your point and your first answer was also very interesting in the way that it makes me see it from a different angle.

When I first started this project I didn't think that the tactile part was going to be the hardest thing to find

There is an old system for touch that used IR emitters and detectors along the edge. The X and Y emitters that were obscured indicated the touch point. It might be possible to adapt that to your application.

 

The touchscreen is only the sensor.
All the fancy and difficult stuff is in the software.

I can believe it, it was what I first expecting at first: hardware? Not a big deal. Software will be the hardest part but in the end as you can see in previous answers it is not as easy as I expected...