Reverse engineering an membrane keypad

[jsw2000]

I have a machine which have 2 BLDC motors and I need to control it. I'm planning to control the motors by tapping the membrane keypad which already exists in the machine using microcontroller signal. I found the 2 pins that connected to certain button and when I keep pressing the button and checking the continuity, multimeter didn't give a beep sound. When I checked the voltage difference in each pin when I pressed the membrane button one was giving 14V while other pin having 4V. But these voltages didn't fluctuate when I pressed the button, but the button function keeps working which means the button membrane should be working fine. Is there a way to mimic the button press using a microcontroller or I need to know an alternative way to control 2 BLDC motors which are 12V and 15A rated current.

 

[Nigel Goodwin]

Membrane keypads are mostly multiplexed, so you need to ensure that you're using the correct two connections for that specific button - depending on type though it may not be low enough resistance to make your meter beep.

The ones we have made for work are low enough, as they include nice mechanical dome switches - most of what I would call 'membrane' switches are low resistance, but not all.

Are you absolutely sure you checking on the correct 2 pins - it's often not very obvious.

Assuming you find the correct pins?, and assuming they are multiplexed?, then a small relay is probably the best and safest way to 'press' the button from a micro-controller. A CMOS switch 'might' work, but isn't guaranteed to, where a relay is.

 

[jsw2000]

I took out the membrane keypad and observed the two connections that are connected to the button and the pins. Can I short circuit these 2 pins and see if the function works? Can I use a transistor instead of relay? Since these pins working at 15V to 4V range is it possible to short circuit the pins using jumper wires?

 

[Pommie]

Read up on how multiplexed keypads work Here.
As suggested, a relay is probably going to be the most reliable method.
I don't understand the 15V to 4V range and that would worry me.

Mike.

 

[jsw2000]

Thanks for the help, can I short circuit the 2 pins using jumper wires to check the functionality of the button, and should I use the relay in between these 2 pins or GND and the pin that makes 0 logic? This is for my university project btw.

 

[Nigel Goodwin]

Thanks for the help, can I short circuit the 2 pins using jumper wires to check the functionality of the button, and should I use the relay in between these 2 pins or GND and the pin that makes 0 logic? This is for my university project btw.

Yes, you can short the pins - if multiplexed there MUST be no connection to ground, or anywhere else.

I'm somewhat concerned though if it's for your University Project, and yet you don't know the simple basics of electronics?.

 

[Pommie]

I'm somewhat concerned though if it's for your University Project, and yet you don't know the simple basics of electronics?.

This really makes me wonder what kind of world my Grandkids will live in.
I've designed a PCB that allows Tic Tack Toe (Noughts and crosses) to be played with an Arduino Nano. The idea is to teach C at a fundamental level and Game Design as the most simple games need a start, test for complete (winner) and how to start again. Could even go as far as writing a library for the board as I believe this skill (writing libraries etc.) will be in huge demand in the future and they'll be no-one around who knows how to write it. Could be a huge earner for the right person.

Mike.

 

[Tony Stewart]

The keys are scanned with a matrix or rows and column switches, so the DMM readings you get are from scanning all the keys so the waveform is shared by all key signals in a time multiplexed sweep. If you share a common ground and supply it is possible to use a transistor an opto switch or a reed relay. An optoisolator is probably your best bet.
I did the same when i graduated for a friend at work who wanted to use the 1st Sears LED calculator as an exercise machine event count up/dn and calorie counter with have a 1" numeric LED display.

So this is a relatively trivial task, once you learn how the interface works. and avoid adding too much capacitance to the switch. (e.g. 100 pF) which adds memory to the open circuit matrix scan. (not good) Much like how dynamic memory is matrix scanned with dynamic charges in each cell.

 

[Tony Stewart]

This really makes me wonder what kind of world my Grandkids will live in.

AI will make most repetitive jobs easier, allowing more creative jobs but if you don't know the fundamentals to validate the assumptions, you can be easily fooled. Most of the results will come easier like search engines using AI, but not always better...

 

[Pommie]

AI will make most repetitive jobs easier, allowing more creative jobs but if you don't know the fundamentals to validate the assumptions, you can be easily fooled. Most of the results will come easier like search engines using AI, but not always better...

I'll assume that AI will write the AI as the kids I see coming out of school won't. And won't know if/when they are being fed gibberish.

Mike.