TV Remotes

[#1]

Just curious as I need to make a remote for something...

How on earth do TV remote batteries last so long? From my limited understanding they must have a processor that is active at all times scanning for button presses however they seem to last for an eternity.

If I were to try and replicate that with an arduino the best I could hope for is a few days.

What gives?

[#2]

I don't know... but I do like that my remotes are a decent retirement home for batteries that have drained in childrens' toys... and remain so for years rather than months. 

[#3]

@7up  Hmm don`t think that there is a "processor" running "all the time"... it would be activated by the "on" condition of the button pressed, surely,   whatever button was pressed would send a voltage/current to the processor to select the necessary command ? ? ? ? 

If my understanding is correct... then that would explain the longevity of the batteries, and the ability of the "second user" of @APP

[#4]

I generally agree with the battery life on remotes but my remote for my Amazon Fire Stick with Alexa gobbles batteries!?

---

[#5]

Because they use Bluetooth and the voice recognition, and even the non Alexa enabled remotes eat batteries

[#6]

---

@shutter wrote:

@7up  Hmm don`t think that there is a "processor" running "all the time"... it would be activated by the "on" condition of the button pressed, surely,   whatever button was pressed would send a voltage/current to the processor to select the necessary command ? ? ? ? 

If my understanding is correct... then that would explain the longevity of the batteries, and the ability of the "second user" of @APP

---

What you're saying is kind of correct in a way - any chip that accepts digital inputs will react to an "on" signal however I've never heard of a chip that has multiple Vin pins to turn on the chip.

The chip is usually powered via a Vin pin and a GND pin and the rest of the inputs are read while the chip is running.

I suppose the other way to do it might be for the button to trigger a supply to the chip through a thyristor and also setting a flip flop so that the chip can read which button was pressed?

[#7]

You don't need multiple Vin pins, all you need is a passive diode OR gate that switches the power on from any key press. Alternatively of course you could have a proceesor with a very low power sleep mode. I suspect that different manufacturers have different solutions.

[#8]

So once it's turned on via that OR gate, how would you then recognise which button was pressed? Or would I be right in thinking that the chip would execute its code so quickly that the button would still be making contact?

Moderator's note by Mike (Mav): Full quote of preceding post removed as per Forum rules.

[#9]

It would not need to fully execute the code, only latch in to its memory which button had been pressed, which should take milliseconds. Then having locked in to memory which button was pressed, regardless of the condition of the button, execute the code. After a preset time it could then clear down the button press memory ready for the next button press or switch off if there's no consecutive button presses.

I'm not saying that this is the way these things work, it's my first thoughts on how I would design a remote control.

[#10]

---

@Baldrick1 wrote:

I'm not saying that this is the way these things work, it's my first thoughts on how I would design a remote control.

---

Ah.. but i'm asking how it works lol. What you describe isn't really possible AFAIK - the processor needs to power up and then *somehow* know which button was pressed.

It can only do that by reading the value of the pins, just pressing a button can't store it in memory before the processor has even started to execute code - especially if it wasn't even powered up. Well thats my take on it anyway.

[#11]

I agree that the button has to be pressed for long enough to boot the processor and read the button. The question is, how long does this take? The point I was trying to make was that you don't need to hold the button until the code has finished running. Alternatively of course if the processor is slow to boot you could latch the button data using cmos chip(s) external to the processor.

[#12]

Well having a small stash of parts such as arduino nanos and other misc bits I decided i needed to set about trying to achieve this. Although i have a large collection of integrated circuits someone gave me moons ago I had no idea what they were or if they were suitable so set about this in the easiest way possible - using thyristors... or rather self made thyristors made from pnp and npn.

So the button is pressed, the thyristor activates and keeps the power flowing through a diode to another npn transistor that sucks blood from another pnp transistrr (minimal V drop) to power up the arduino nano. To turn itself off the nano can then set an output pin high and power up another npn transistor to ground the gate of the thyristor(s) and put it out of action.

That's great.. but more than one button? The arduino can also read which thyristor circuit is turned on by reading the cathode voltage on the thyristor (using a pull down resistor between ground). The output from each thyristor via a diode ensures that the arduino does not read a voltage on all of the thyristor cathodes at the same time. Here is a schematic:

Here is a working sample of the circuit: http://tinyurl.com/yalp64s5

And the breadboarded / working version: