I was the other day fixing a phone (land), and saw the rubber pad with all the numbers and buttons, and for the first time, after having seen these rubber pads in my portable consoles, TV remotes and basically everywhere, I asked myself...
Wait a second, how in heavens these rubber (RUBBER!!!) switches work?
How can they close the circuit?

So I checked continuity with my DMM and BOOM, no continuity, no BEEP, no threshold trespassed. That automatically means, wow, these SWITCHES either don't work or are pretty much broken. However these things still worked, flawlessly, but how, I was thinking.

So I measured resistance and I was getting so very bad values, between 5000 ohm and 300 ohm, depending how much I pressured them. Every time I've seen a switch it's designed to, well, be a switch, no resistance at all.
So my question is... How these work with such high values for a switch?

Yeah, I've read there's a thin layer of carbon at the bottom black rubber part, but still, why so much resistance and how do they work in all the appliances so well, having 300 ohm?

How do you check if these are worn or not then?
Starting with their value, which is quite varying.
Even more importantly, why a switch with such high resistance was ever designed or accepted?

 

The answer is that the conductive elastomer pads depend on the circuit sensing the difference between a very high resistance, (open circuit) and a much lower resistance, (Less than 100K ohms). And so they use typically a carbon-filled elastomer, and the better quality ones use plated PCB pads.
The motivation for acceptance is simple: It was not possible to produce anything cheaper that worked. And as the general public has no clue at all, the lowest cost that delivers an adequate production yield wins. There is no concern about if the remote or phone fails after the 90 day warranty expires.

 

The conductive element in a molded keypad is called a carbon pill and on the low end can have a resistance of <10Ω (there are "low resistance" versions that offer this sort of conductivity) and more typically about 100Ω. Because the resistance across the contacts of such a keypad with the key up is effectively infinite at the voltages involved, there is no problem distinguishing the key down state which actually conducts.

The pills are carbon or graphite in a silicone elastomer. They can get dirty enough to stop working reliably. There are after market self-adhesive versions to repair keypads in that case.

 

I was the other day fixing a phone (land), and saw the rubber pad with all the numbers and buttons, and for the first time, after having seen these rubber pads in my portable consoles, TV remotes and basically everywhere, I asked myself...
Wait a second, how in heavens these rubber (RUBBER!!!) switches work?
How can they close the circuit?

So I checked continuity with my DMM and BOOM, no continuity, no BEEP, no threshold trespassed. That automatically means, wow, these SWITCHES either don't work or are pretty much broken. However these things still worked, flawlessly, but how, I was thinking.

So I measured resistance and I was getting so very bad values, between 5000 ohm and 300 ohm, depending how much I pressured them. Every time I've seen a switch it's designed to, well, be a switch, no resistance at all.
So my question is... How these work with such high values for a switch?

Yeah, I've read there's a thin layer of carbon at the bottom black rubber part, but still, why so much resistance and how do they work in all the appliances so well, having 300 ohm?

How do you check if these are worn or not then?
Starting with their value, which is quite varying.
Even more importantly, why a switch with such high resistance was ever designed or accepted?

There are actually a couple of varitions on this.

Normally, the button has a conductive layer on the bottom of the rubber, painted on (conductive to some degree). This is then pressed down onto pcb traces that are not connected, but interleave like two hair-combs with their teeth intertwined. The conductive layer on the rubber button makes a path between the pcb-traces, lowering the resistence.

 

I had a number of remotes and cordless phones fail because the contact resistance of that carbon path become too high.
DirecTV alone has provided with about three new remotes (on their dime) after the old ones failed.
I guess that's cheaper for them than paying for remotes with longer lasting keys.

 

I guess they still made many, many devices with such a poor designed switches because it is insanely cheaper than a regular switch like these:


Plus no soldering... so easier.

 

I guess they still made many, many devices with such a poor designed switches because it is insanely cheaper than a regular switch like these:

Correct about the cost, but incorrect about the "poor design". CR keypads are very well designed for what they are intended to do. Disagreeing with the design goals does not mean the goals were not exceptionally well met. A tact switch like you show is very low cost, but a conductive rubber switch is at most 10% of that cost, and even less when you add in the cost of inventory and assembly. Times a few hundred *million* switches - per year - and there is a strong business case for the reliability tradeoffs.

ak

Note to Wally: They last longer if you don't use them as a doorstop. Or a tack hammer. Or . . .

 

I guess they still made many, many devices with such a poor designed switches because it is insanely cheaper than a regular switch like these:
View attachment 257976

Plus no soldering... so easier.

That's the idea. Save every penny possible to make the next quarter's earnings statement look better than the last!

 

When a 1k ohms rubber switch closes on a circuit input that is 1M ohms then the closed voltage loss is only one thousandth.

 

When a 1k ohms rubber switch closes on a circuit input that is 1M ohms then the closed voltage loss is only one thousandth.

What? Mind explain it more detailed?

 

The resistance of the switch contact and the circuit input impedance or pull-up resistance or whatever the switch is connected to form a 2-resistor voltage divider. If the pullup resistor is 9 x the max switch contact resistance, then the voltage *change* seen by the downstream circuit will be at least 90%. In the example by AG, the voltage change would be from Vcc to Vcc / 1001.

ak