Would like to share an experience with you. Back in high school I built a relaxation oscillator; a neon bulb parallel to a capacitor. In series was a diode and resistor. This was connected to 120VAC. The diode blocked the reverse current while the resistor slowed the forward current. That charged the capacitor slowly. When the voltage was high enough to ignite the neon bulb it would flash and discharge the capacitor. The whole thing would repeat until unplugged.

HERE is the experience: While the bulb was flashing about twice per second I discovered that putting my finger near the bulb the flash rate would increase to about three times per second. It took me years to figure this out - it was the static charge I brought near to the neon bulb that caused it to flash at a lower voltage on the capacitor.

I DON'T KNOW if that will spark an idea but if you can bring a high voltage near to the switch it MIGHT cause it to switch. I'm thinking that the presence of electrons will influence the capacitor to think it has a higher charge, similar to when you put your finger near to the switch. KEEP IN MIND static charges cary extremely low current. If you experiment with high voltage MAKE SURE you are grossly limit the current.

Winter time when the air is dryer static charges are more common.

 

Could you use a varactor? Connect one terminal to ground and the other to something conductive attached to the touch pad, Then vary the voltage on the varactor.
No idea if it might work, but varactors seem to have about the right sort of capacitance variation.

 

Which one above ?
What about some foil on top of the capacitive sensor plate that will act-move with a compressed air nozzle ?

A remote controlled actuator touching the capacitive target ?

Oh haha, big typo back in post #17. It was supposed to read, "I cannot get inside the unit to modify anything or connect any wires, etc.", or something along those lines. Sorry about that

Yes, it looks like I will be going mechanical as I do not think there is a good way to do this. I already have actuators too.

 

Would like to share an experience with you. Back in high school I built a relaxation oscillator; a neon bulb parallel to a capacitor. In series was a diode and resistor. This was connected to 120VAC. The diode blocked the reverse current while the resistor slowed the forward current. That charged the capacitor slowly. When the voltage was high enough to ignite the neon bulb it would flash and discharge the capacitor. The whole thing would repeat until unplugged.

HERE is the experience: While the bulb was flashing about twice per second I discovered that putting my finger near the bulb the flash rate would increase to about three times per second. It took me years to figure this out - it was the static charge I brought near to the neon bulb that caused it to flash at a lower voltage on the capacitor.

I DON'T KNOW if that will spark an idea but if you can bring a high voltage near to the switch it MIGHT cause it to switch. I'm thinking that the presence of electrons will influence the capacitor to think it has a higher charge, similar to when you put your finger near to the switch. KEEP IN MIND static charges cary extremely low current. If you experiment with high voltage MAKE SURE you are grossly limit the current.

Winter time when the air is dryer static charges are more common.

Oh yes that sounds like an interesting idea. The only thing is, I would be afraid that a high voltage pulse would punch through the thin plastic and destroy the electronics. There's no way to experiment with it because once it blows out it stays blown out. Still an interesting idea though.

 

Could you use a varactor? Connect one terminal to ground and the other to something conductive attached to the touch pad, Then vary the voltage on the varactor.
No idea if it might work, but varactors seem to have about the right sort of capacitance variation.

Oh maybe, I haven't tried that. I guess I'd have to get a varactor I don't have any in stock.

 

This little circuit simulates a variable capacitance that can trigger touch switches with no moving parts.

When the input is high, the electrode becomes a floating node between two reverse biased diode junctions, a very low capacitance.
Set it low and it looks like a touch, a substantial capacitance to ground. (like a finger)
Stick this circuit as close as possible to the electrode, which can be a finger-sized piece of copper foil stuck over the button.
Depending on the type of sensor, you might need to fiddle with the electrode size and placement.

That looks promising.
So they are using the voltage dependency of the capacitance of a diode to simulate a varactor diode.

 

sort off... varactors are normally operated reverse biased and the capacitance is lowered by more negative voltage.
and here it looks like voltage divider is set to keep the diodes forward biased and right at the edge of conducting.

 

sort off... varactors are normally operated reverse biased and the capacitance is lowered by more negative voltage.
and here it looks like voltage divider is set to keep the diodes forward biased and right at the edge of conducting.

Yeah, and I am not sure about this yet. I would think that going from reverse biased to forward bias would cause a major change in capacitance.
Maybe using an array of diodes would work better, with them all in parallel, so the highest capacitance goes up.

 

Yeah, and I am not sure about this yet. I would think that going from reverse biased to forward bias would cause a major change in capacitance.
Maybe using an array of diodes would work better, with them all in parallel, so the highest capacitance goes up.

It's not just the capacitance in bulk, to get good sensitivity to touch emulation, the system must have the appropriate area coverage of the touch sensor target.

 

sort off... varactors are normally operated reverse biased and the capacitance is lowered by more negative voltage.
and here it looks like voltage divider is set to keep the diodes forward biased and right at the edge of conducting.

Not quite...

The diodes are acting as a switch, going from full reverse bias (where we have only the small junction capacitance) to conducting, where they are low impedance, effectively grounding the electrode. The electrode goes from floating to grounded, just like that coin demo video.

This represents a large capacitance change, which is detected by the sensor.

The sensors are always adaptive, they need to adjust to the nominal capacitance of the electrodes, a gross change in capacitance gets detected as a button press.

 

At least two touch switch schemes involve the touch changing the electric field at the button. So I still suggest experimenting with a ground-referenced higher frequency AC signal fed to a conductive foil piece stuck onto the button. Or maybe attached to a small coin stuck to it gently. Feeding in a capacitance coupled signal is the main path since getting inside is not available.

 

At least two touch switch schemes involve the touch changing the electric field at the button. So I still suggest experimenting with a ground-referenced higher frequency AC signal fed to a conductive foil piece stuck onto the button. Or maybe attached to a small coin stuck to it gently. Feeding in a capacitance coupled signal is the main path since getting inside is not available.

Your finger is not inside either but it works. What you need is an alternative path (earth/utility ground) for the electric field energy to trigger a response. The circuit for the alternative energy path is space.
The battery powered phone is well insulated, isolated and the clip is connected the case of a grounded appliance instead of a circuit common on the phone.
https://youtube.com/watch?v=ssftPIyXr4vt2ChZ

 

I decided to make a video about this:

Enjoy.

 

I decided to make a video about this:

Enjoy.

Hi,

That's a little "wordy" but cool. It illustrates the idea fully which is nice to see.

Now just a little question.
How is the reliability of the switch activation? That is, when you press one of the black buttons on the 'remote', what is the chance that one of the touch switches will not actually be activated until maybe you press the black button a second time?
This would be an important point for my use because if there was any chance at all of it not working even one time, I'd have to provide some sort of feedback to test if it was activated and if not 'press' it a second (or third) time.
Mine will eventually be controlled automatically, as if those black push buttons switches were actually an Arduino (or something like that) that activates the touch switches all by itself. That would sort of be equivalent to having the fan in that video turn on and off by itself at certain times of the day or something, controlled by a microcontroller. In fact, if you ever get around to adding that (Arduino control) to the video that would be informative and even cooler

 

The attached 20 year old patent explains how Apple's first touch switch worked. The first named inventor (Chris Krah) went on to design Apple's iPod and iPhone touch interfaces among other things.

The user's finger was sensed through a translucent plastic "plug" in the switch assembly.

Don't forget: Don Lancaster did it first, long before...

View attachment 59986

 

Most touch switches that are not optical work by sensing a field disturbance, either an increase or a decrease. Thus what will usually work to trigger them, will be to provide an alternative field disturbance. The simple way to do that is to provide an external electrical field increase. THAT can be done by having an oscillator provide an increased field by means of an electrode in the position that a fingertip would occupy, with a few volts of some kilohertz voltage, relative to whatever the common reference voltage may be. Either a sine or a square wave should work, some experimenting to find the best frequency may be required.

 

Most touch switches that are not optical work by sensing a field disturbance, either an increase or a decrease. Thus what will usually work to trigger them, will be to provide an alternative field disturbance. The simple way to do that is to provide an external electrical field increase. THAT can be done by having an oscillator provide an increased field by means of an electrode in the position that a fingertip would occupy, with a few volts of some kilohertz voltage, relative to whatever the common reference voltage may be. Either a sine or a square wave should work, some experimenting to find the best frequency may be required.

Maybe but there is zero need (as demonstrated here several times already) for a external sine or a square wave signal to make a 'field disturbance' capsense type system work reliably fingerless. These systems are designed to work on easily emulated conditions (monitoring internally generated e-field changes), with simple methods of capacitance switching.

 

Most touch switches that are not optical work by sensing a field disturbance, either an increase or a decrease. Thus what will usually work to trigger them, will be to provide an alternative field disturbance. The simple way to do that is to provide an external electrical field increase. THAT can be done by having an oscillator provide an increased field by means of an electrode in the position that a fingertip would occupy, with a few volts of some kilohertz voltage, relative to whatever the common reference voltage may be. Either a sine or a square wave should work, some experimenting to find the best frequency may be required.

Hi,

I'd have to check for that I guess. I guess that might also trigger a capacitive sensor depending on how they are measuring the change.

 

If you look at the waveforms on the sense electrodes, they are not doing simple stuff, it's a complex multiplexed dance of charge measuring cycles.
Any AC signal you hit it with would need to be synchronous and rather complex to work as a reliable trigger signal.

 

Capacitive touch sensors need to be carefully protected from external fields, lest they trigger at the wrong time, or more likely injection lock to the interfering signal and don't respond to the approaching digit.