Weight sensors, yes, the infinite subject that I keep trying to search for more information on. But now, I take a break to try to think through things myself and explain how this can all work.
The easiest way to build a weight sensor is to modify a spring-based scale. A spring-based scale uses the displacement caused from compression of a spring, due to an applied force, to generate a rotational displacement on the display dial. The spring is constructed to be as close to linear as possible, then the display dial simply has a printing of linear numeric values that correspond with the applied force. The electronic version simply applies a variable resistor in place of the display dial, which then generates a resistance that linearly varies with the applied force. The primary disadvantage of this weighing system is that the spring requires a lot of space, adding excess thickness to your weight sensor and making it less attractive to use.
A thinner, but potentially less accurate, method to build a weight sensor is via capacitive means. Two large, thin metal plates sandwich a thin piece of insulator that is designed to compress under load. Effectively, this behaves as a capacitor. When a force is applied, it results in compression of the sandwiched insulator: the thickness of the insulator decreases, which causes the capacitance to increase. In this case, the primary issue in accuracy is getting a thin insulator material that has a linear compression response.
Wow, you can do a lot with simple resistors and capacitors, come to think of it. What are all the physical quantities you can measure with them?
- Temperature
- Humidity
- Pressure, force, weight
- Light intensity
- Analog joystick location
Note that I’ve included slightly more specialized such devices in my list too, i.e. photoresistor.
Now I go searching on the Internet, and wow, this is a really interesting article, a larger project has been built around making cheap capacitors using paper and aluminum foil. Yeah, it really is that simple if you use letter-sized paper and aluminum foil.
20191204/DuckDuckGo capacitive weight sensor
20191204/https://blog.yavilevich.com/2017/10/40-cent-diy-pressure-sensor-based-on-a-capacitive-principle/
Also interesting is mention of many other weight sensor technologies, force-sensitive resistors. Well, I guess these can be simple to make and intuitive to construct too. Somehow the decrease in distance in a resistor from an applied force results in a decrease in resistance, I guess that could be one way.
Indeed, the capacitance of a parallel plate capacitor is inversely proportional to the distance between the plates. So, if you have an insulator that compresses linearly by the applied force, i.e. twice as much force doubles the decrease in thickness, the capacitance is proportional to the applied force. This is essentially to say that you want an insulator that behaves as a Hooke’s Law spring, so we circle almost right back to where we were with the old fashioned spring-based scale.
Another consideration when building your own parallel plate capacitor is to choose an insulator that has a good permittivity, the ability to “polarize” in charge when an electric field is applied.
20191204/DuckDuckGo capacitor distance between plates and capacitance
20191204/https://electronics.stackexchange.com/questions/16457/why-does-the-distance-between-the-plates-of-a-capacitor-affect-its-capacitance
20191204/https://en.wikipedia.org/wiki/Capacitor
20191204/https://en.wikipedia.org/wiki/Permittivity
20191204/https://en.wikipedia.org/wiki/Hooke%27s_law
Now, another idea that I have. Often times when building cheap aluminum foil and paper tactile switches, you may have trouble where you build device that doesn’t quite activate conductivity when you want it to, yet the foil plates are very close. Wouldn’t it be nice if you could have a hybrid between a conductivity and capacitive sensor? This seems easy enough if you only have on switch. Simply have both a tactile switch circuit and resistor-capacitor PWM circuit wired up to the same device. Alas, matrix keyboards are more of a challenge. Here, you cannot use regular diode isolation since diodes have voltage regulation effect. But maybe you could substitute Schottky diodes instead of thte regular cheap silicon diodes and be able to multiplex a capacitive sensing circuit. No, that doesn’t seem to work, seems more like you’d need a transistor at every matrix point.
Another good realization, you can use a pressure sensor to measure fluid flow in a pipe.
20191204/DuckDuckGo capacitor pressure sensor
20191204/https://instrumentationtools.com/capacitive-pressure-sensor/
Piezoelectric sensors can also be used to measure pressure.
20191204/https://www.avnet.com/wps/portal/abacus/solutions/technologies/sensors/pressure-sensors/core-technologies/capacitive-vs-piezoresistive-vs-piezoelectric/
Something went wrong… www.avnet.com
We truly apologize that this error has occurred. We take these matters very seriously and ask for your help in notifying us of the problem. Please email us at DIGITAL-SECURITY@avnet.com, using Reference Number:
18.eaafedcc.1575446291.3dd676c0
Thank you in advance for your assistance in resolving this issue.
Avnet Customer Care