Hello Guys, yes I'm sorry, another newbie asking questions...
To be upfront, i'm retired and this is a home project. I thought it would be simple but has got me stumped.

Objective: Control my hot tub via home automation.
Details: It's a lazy spa hot tub and has touch sensor buttons, that I'm trying to operate via a smart switch relay modual.
Regarding the touch switches, I assume they are capacitive as they work with through a waterproof membrane. I have had the control unit apart and it is completely potted with luckily clear compound. The touch switches are springs that protrude out of the compound and each spring looks to have only one connection. The connections go via a series surface mount component (resistor?) to an IC with no number printed on it. To be fair, it looks very well made.
Am I correct in thinking the IC probably pumps the spring with AC and monitors the capacitance? Then it it is the change in capacitance that makes the switch activate.
The supply to the board is a standard 5V.

I have tried using various size capacitors with one side connecter to the 0v to simulate a touch operation, but so far, can't get the detect this as a valid touch request.

On the home automation side, I have wifi four relay unit all with isolated change over contacts.

Thanking you in advance. Any advise welcome.
Paul.

 

The touch terminals will probably go via the body to the ground terminal on the circuit board, try and test it by jumping it to ground.

 

The touch terminals will probably go via the body to the ground terminal on the circuit board, try and test it by jumping it to ground.

Thank you Dave,
Yep already tried this, direct 0v to spring. No joy. (I attached the 0v wire to a piece of wood, as I didn't want to influence the signal!)
It's like the chip needs to see a change in capacitance in a 'sensible' range. Just a shame the chip number has been removed...

Also done some experiments using screened cables as ultimately I need to wire to the relay controller.

Frustrating, it seems it should be possible, but maybe not...

 

Okay what about the positive rail , does it work then?

 

Okay what about the positive rail , does it work then?

Nope. I've also tried holding a cable with connected to an AC adaptor near the spring, still doesn't trigger.

 

So it must be picking up AC signal from the body..?
Any pictures of the circuit board?

 

Do the springs connect to larger conductive electrodes behind the front panel?
The circuit might not work correctly without the electrodes.

The circuit shown below is from a machine designed to test capacitive touch switches, the electrodes are placed where the finger is supposed to touch.
The end of D3 goes to a TTL 0-5V signal, when this is low, it simulates a "touch"

 

Welcome to AAC.

Try wrapping several turns of small gauge wire around a plugged in power cord and connecting that to the switch via the relay.

Keep the wire as short as practical and be sure all the insulation on both wire and cable are fully intact.

 

Usually it works by driving the touch pad via a resistance, then measuring the delay caused by the RC network formed by the resistance and the capacitance to ground.
Unfortunately, there may be enough capacitance between open relay contacts to trigger it, so I doubt your plan will work.
How about a photo of the touch pad arrangement?

 

Thank you for your useful suggestions, I'll try and work through them.

I've been doing a little more research and under magnification and strong lighting, discovered the mystery chip part number. Its FTC334. The only data sheet I could find is in Chinese, but here is the Google translation of part of it.

FTC334F is a professional capacitive touch button processing chip. It adopts the latest high-precision digital capacitance measurement technology and can prevent various interferences, prevent the influence of water droplets on the panel, and adapt to various power supplies. It can support up to 10 touch button functions. The output adopts scanning encoding output method. Only 3 wires are needed to read the button value. It can be connected with LED driver ICs such as 1628. It is very suitable for the application of various electrical products with digital screens.[/FONT]

Features:

Super anti-EMC interference, can prevent the antenna of transmitting equipment such as walkie-talkies with power up to 5W from being close to the touch point from interfering.

* Extremely simple peripheral circuit, the simplest application peripheral only requires a reference capacitor. (Depending on customer requirements, if you need to improve ESD and EMC, you need to connect a resistor to each button)

Water-proof and flood-resistant, patches of water droplets covering the touch panel will not affect the effective recognition of the keys.

The ultra-wide operating voltage range is 3.0V-5.5V and can be applied to the currently widely used 3.3V systems and 3.0v battery systems. *Power supply voltage change adaptation function, built-in voltage compensation circuit, automatic compensation when the power supply voltage changes within the working range, without affecting the normal operation of the chip.

* Automatically adapt to changes in ambient temperature and humidity, and the application of slow environmental adaptation technology allows the chip to work continuously for an unlimited time without sensitivity differences.

* Adjustable sensitivity, the sensitivity can be adjusted by external capacitor capacity to adapt to different designs. *

Provides a scan output interface to facilitate user system docking with the display driver chip 1628.

It is quickly initialized after power-on. Within about 300ms, the chip can detect the environmental parameters including automatic adaptation, and the button detection function starts to work.

*Sensitivity automatically adapts. If the parasitic capacitance of each button lead is different due to different lengths, it can automatically detect and adapt, and the sensitivity of different buttons can be consistent.

 

There's enough information in that datasheet for me to see how it works.
Your best bet would be to intercept the connections between controller and touch pad IC.
D0-D3 are a binary code for the switch that is pressed.
The voltage level on SCAN determines the data to be presented on the OUT pins.
Not a trivial job, but by no means impossible!

 

There's enough information in that datasheet for me to see how it works.
Your best bet would be to intercept the connections between controller and touch pad IC.
D0-D3 are a binary code for the switch that is pressed.
The voltage level on SCAN determines the data to be presented on the OUT pins.
Not a trivial job, but by no means impossible!

Thank you Ian. Trouble is the board being completely potted in clear compound adds to the complication

There may be another cheat option. After a some more extensive ~Goggling~ have found a website offering what is probably a wifi linked microcontroller that sits in-between the switch boards and the main PCB located in the bowls of the unit, It seems to intercept the signals and takes control when commanded with an phone app. Also links to 'Home Assistant' All this for £46.

https://hottubmanager.co.uk/

I feel slightly ashamed to admit, this is probably the way I will go, but I much appreciate everyone's time and input.

 

I feel slightly ashamed to admit, this is probably the way I will go, but I much appreciate everyone's time and input.

In that case, I won't take the trouble to work out exactly how to intercept it!
I wonder if the same IC is in my fridge, which keeps changing temperature at random. . .