I am trying to connect my mkr pins to a remote control that sends rf signals to a set of devices.
I already placed a opticoupled relay im between the mkr pin and the remote control button to be replaced, but i cannot figure out this last part.
It seems the remote control pcb has the two terminals of the button printed on its surface as conductive carbon areas separated by an insulated strip in between. The original button had a piece of conductive plastic that, when the plastic button was pushed, shorted the two carbon areas.
I have been told i cannot solder my wires to the carbon surface. It would jusr burn and won't adhere. Conductive paint or epoxy were the only options apparently, but i doubt they would hold the wires properly and be able to remove if the remote is ever needed again.
Some type of clamps with the wire soldered could do the trick.
Any suggestions/ideas? None i came up with feel right.
That is a quite common method, for years ago! Follow the traces back to where real solderable connections are on the board and connect your wires there. Be aware that you will need to have one connection that is ground on your mkr.
Afterthought: What are the voltages on the switch you are replacing? There must be some voltage difference for your opto coupler to work with and you must have the polarity connected properly.
The only available metal pins i can find are the legs of the IC. For now. I will have a more in depth look when i get a lense (my eyesight is very poor).
The remote control uses a CR2032 battery so I assumed 3v would be the operating voltage at the button pins, but it is true i have not measured yet. I guess a simple voltage measure between the pins while unpressed the button would suffice, right? I wonder if my metal probes would work or damage the carbon surfaces.
Any other way to do it? Any experience on conductive paint or epoxy?
Also, why sharing the ground with the mkr is relevant? I thought the purpose of the optocoupled relay was to make both sides independent of each other in every (electrical) possible way.
Yes, access the pins from the under side of the circuit board, if possible. Tried conductive epoxy to repair a 100kHz crystal suspended by conductive quartz fibers. Didn't work.
Your reply seems to indicate the button switch is connected to TWO pins on the IC. What is the IC? Did you research the IC to see what the pins do?
I did not know you were also keeping the power separate.
Either solder directly to the IC's pins that you traced the contacts back to.
Alternatively, scrape away a bit of the solder mask from the traces that connect to the carbonized contact pads and solder to the traces. Be careful not to exert too much pull force on the traces after you've soldered on your own wires. Also be careful not to solder at too high a temperature. Both of these mistakes can result in the traces lifting off of the PCB. When you've soldered on your own wires, glue or tape them to the original PCB so that any pull force is handled by the tape/glue.
This should work. There are other ways, but relays are intuitive to use, so go ahead.
It's an opto-coupled relay, probably a relay module with an opto at the input. The optocoupler doesn't play a role in the actual button shorting process; this is handled by the relay part, at least the way I read OP's question.
There is nothing written at the top of the IC. The back if the board is empty, not even strips there since it is single sided. There is just a side naming code reading U1.
After studying the pictures, scraping away some of the solder mask on the copper traces seem to be the easiest modification to the circuit board. This is a common practice when modification to a circuit board is necessary. Use fine sized stranded, tinned copper wire to make the connection. Make the wire extra long during your testing. After soldering, use a gob of electronic type RTV to hold the wire tight to the board, but not over your solder joint. Then make your test using the wires to simulate the push button.
Thanks. I never scraped the epoxy over the strips before, so I'll look for a dead board to practice first. It is very likely I won't do it right the first time. I assume it should be better than the scraping of the carbon, right?
I noticed many of the eight buttons share the same wire. I understand one side must be Vcc, but it wouldn't make sense for the other pin to be connected to any other button unless the IC was some kind of three/four digit binary encoder (eight buttons require 3 pins, but the IC would need to be oddly specific to match exactly the need). Perhaps digging into the purpose of the IC as was suggested would provide alternatives. Not that it would be easy since some strips are bridged with more carbon on top of rubber and there are hidden strips under the IC, but since it is single sided it should be doable.
This is donr for pin economy; this way they don't need 8 GPIO's on the microcontroller (that's what the chip likely is), but much less. Each button has a unique pattern of connections made when pressed. The drawback of this is that when the user manages to press several buttons at a time, this may result in unanticipated behavior. However, most of the time this approach works fine and prevents the need for a bigger and more expensive controller. Every cent counts when you're making these in the thousands.
It enables you to keep using the remote as originally intended, i.e. the buttons will still work. That's an advantage.
Makes a lot of sense. But, wouldn't that enable me to wire just three pins to my mkr for full control using my digital outputs? No relay pretending to be a button. And since mkr works at 3.3volts and the remote control works on a 3v battery, wouldn't that allow for a direct interconnect of 3 pulled up digital outputs at the mkr and the 3 pins at the remote?
I also noticed that the resistance of the carbon strips is around 20 to 60 ohms. If I use the relay, wouldn't i need a resistor in series too?
I see five traces connecting the buttons to the controller IC. So no, 3 wires won't cut it to control 8 buttons.
Having said that, it's possible that you could spoof the buttons in a different way than using a relay. JFETs or CMOS switches come to mind. The latter would be relatively easy to implement without having to figure out much about how the controller on the remote actually reads out the buttons. Just use an array of 8 CMOS switches that you control with your own circuit, one for each button.
You could also try to emulate the entire keypad using an interface of only the 5 button traces/wires (plus common GND), but you'd have to start by drawing the schematic of how the buttons are wired and then recreate that logic on your own board.
The way it was must have conducted very little current, designed that way. If you short those pads with a piece of antistatic bag, will it work? Because if so. it might be possible to make a flexible bridge button and glue it right over the carbon pads.
Simpler with the relay then, though way more expensive and pin intensive on the mkr. I managed to bind all receptors with just one of the buttons (D on/offf). So not so bad for my specific purpose.
Maybe. If hot glue could work, it can be undone and redone with heat. Alternate is a frame for all the buttons and mechanical means to hold it in place and any of that would require close work and possibly drilling some holes (shudder).
If there's a cover, something might mount on that to be in place when the cover goes on.
I can sympathize with the vision trouble. I was using a magnifier visor to do wiring and got to the point of not knowing the inability to focus due to hardening eyeballs from cataracts growing but they tell me that upcoming surgery will give me clear vision! I won't be able to squint-focus but with different lenses I should see clear and sharp again. That's where I tell them that I was born with lens defects, if it works I;ll be able to see clear for the first tume ever! Booyah! Life's a beotch!
Great news on that eye fixer/upper. Too bad it cannot be DIY too.
Yes, there is a cover holding the button pad down. So I was thinking, could it work if i purchase some cooper plated tape, fold a third of it onto itself to make it double sided conductor and stick the last third to the pcb surface in a way that the other two lie on top of the carbon coated area. That way i can avoid scratching the traces and presolder my wires to the cooper tape. And at the same time, the buttons would still work. I remove the battery for external power 12vdc to 3vdc and all the wires can come out through the empty battery socket.
That was the concept behind the copper tape. Adhesive next to the button pins but rely on something else to keep it touching the pins. Just didn't figure out what or how yet. Tiny rubber cilinders maybe. The coils may be an option if not conductive unless i could fit two with something in between.
Maybe keeping it simple would be best. Folding the copper tape so its adhesive on three sides keeps it in contact with the carbon surface without the adhesive there. Not so sure about soldering the wires to it since you don't feel inclined to it. Worth the try though. Not a fan of soldered connections, but sometimes they are the only way.
Too bad it cannot be DIY too. 
