Valentine's Day project

[jwny_]

Hey folks.

So, I'm thinking about a possible Valentine's Day DIY geek gift for my wife (early, I know).  I was considering something small, heart shaped PCB, possibly CR2032 powered, such that when two PCB's get close to one another, they would do -something-; perhaps light up a few LED's or something equally dopey. 

However, how to detect proximity between the two devices?  Something RF related (which is, to be honest, probably beyond my skills)?  IR perhaps?  Although IR LED's seem to be rather power hungry.  Any other ideas that I could consider?

If this post is in the wrong area, please let me know.

[jpanhalt]

And your purpose for such a gift is to satisfy your wife or yourself?  Consider something your wife might actually like.  Save the geek stuff for another occasion, like Halloween.

[Kim Christensen]

It depends on how close to your wife you want to get before it goes off!
Maybe some sort of inductive "coupling"...   

[jwny_]

<post edited for the ridiculousness of the reply>

Hi, @jpanhalt.

I suppose the civilized answer to your post would be:

It was just an idea for a goofy side gift type of thing - perhaps I should rethink it.  Ty.

But what I really want to say?  I was just asking for a suggestion on how to detect proximity between two PCB's, and somehow you divine that I should get her something she might actually like; somehow insinuating that she wouldn't like something I had considered making?  What a presumptuous prick.  Jesus, lol.

Ban me from the board if need be - I really don't care.

Thanks,
John

[themadhippy]

reed switch and magnet?

[SiliconWizard]

You thought it was an EE forum, but you get some occasional marital counseling.

Anyway, making something like this low-power (so you can leave it permanently powered with no mechanical on/off switch) and not too expensive isn't actually that simple if you want to make sure the devices don't trigger from anything else (otherwise the "value" of this gadget drops to near zero IMHO).

What would be the maximum distance at which you'd want the detection to occur?

[jwny_]

What would be the maximum distance at which you'd want the detection to occur?

I think this is where my main point of contention comes in -- I was considering something along the lines of a small wearable item with a detection range of a few feet, thus the reason I was thinking RF or IR (at least the sensitivity on the receiving side could be played with to get something close to what I'd like), but as you had brought up, there are not only power issues, but also the requirement of only tracking a single source.  I also considered time of flight and those new mmWave sensors but they also fail the single target problem.

I suppose there may not be a palpable solution to this; at least while keeping it sane.

Thanks for the reply -- appreciated.

[jwny_]

It depends on how close to your wife you want to get before it goes off!
Maybe some sort of inductive "coupling"...   

Haha, like the pun. 

Honestly speak tho, such as with RFID, something like this could potentially work; maybe simply generating/receiving a pulse every so often and determining the strength of the signal for distance estimation.

This is definitely a bit out of my comfort zone, but it's quite interesting.  I'm definitely going to add this to my (rather long) list of items to look into.

Thanks!

[Manul]

Depends how simple you want it to be and what are your preferences.

IR is doable, but you probably need some small, low power MCU. So you wake up, burst, sleep. That way it should be pretty power efficient. The problem is, it's easy to shield IR, so at some specific orientation the hearts might be blind to each other. Also you need to transmit both directions and only one direction might be shielded - one sided love - that is arguably even worse.

I had similar idea years ago, but did not proceed. I remember I was thinking along the lines of putting an RF oscillator into each heart with slightly different frequencies and then both sides detect beat frequency when they are close. It should not be very difficult to make, I don't know, haven't tried.

If you are good at programming, the simplest and quite robust option might be a small MCU with integrated Bluetooth, like STM32WB. You don't necessarily need to design a proper PCB antenna or anything, cause it will still work at small distance. Then you write code to sleep, wake up, exchange a few packets and go to sleep. Can work on CR2032 if done correctly. Based on RSSI you can judge distance and increase the LED heartbeat rate when it comes closer.

If you want a complete overkill, you can use UWB tranceiver IC/module and measure distance by time of flight. Accurate up to centimeters. Might still work with CR2032 at low duty cycle.

[jwny_]

reed switch and magnet?

Thanks for the reply, appreciated!

I was, unfortunately, not very clear in my initial post regarding the distance I had in mind, which was about 3'ish feet or so.  My fault for not being clear, but thank you for the suggestion!

[themadhippy]

not very clear in my initial post regarding the distance I had in mind, which was about 3'ish feet or so

use a bigger magnet

[Kim Christensen]

not very clear in my initial post regarding the distance I had in mind, which was about 3'ish feet or so

use a bigger magnet

Him: "Well, hello there..."
Her: "Is that a large magnet in your pocket or are you really excited to see me?"

[jwny_]

[...]
I had similar idea years ago, but did not proceed. I remember I was thinking along the lines of putting an RF oscillator into each heart with slightly different frequencies and then both sides detect beat frequency when they are close. It should not be very difficult to make, I don't know, haven't tried.

If you are good at programming, the simplest and quite robust option might be a small MCU with integrated Bluetooth, like STM32WB. You don't necessarily need to design a proper PCB antenna or anything, cause it will still work at small distance. Then you write code to sleep, wake up, exchange a few packets and go to sleep. Can work on CR2032 if done correctly. Based on RSSI you can judge distance and increase the LED heartbeat rate when it comes closer.
[...]

Thank you very much for your reply.

This may be the way to go - and although I've not messed around with the BT capable STM32's yet, I've used their STM32G and F series MCU's in the past, so it wouldn't be that far of a stretch...

Guess it's time for a new Nucleo Dev board to play with; glad I have a few months to hash this out!

Thank you again and have a good one!

[SiliconWizard]

What would be the maximum distance at which you'd want the detection to occur?

I think this is where my main point of contention comes in -- I was considering something along the lines of a small wearable item with a detection range of a few feet, thus the reason I was thinking RF or IR (at least the sensitivity on the receiving side could be played with to get something close to what I'd like), but as you had brought up, there are not only power issues, but also the requirement of only tracking a single source.  I also considered time of flight and those new mmWave sensors but they also fail the single target problem.

The "few feet" range rules out a number of approaches, like NFC or some magnetic detection. NFC would give you identification, but it's too short-range and also, the emitting side requires significant power.

Would it be acceptable if both devices were slightly different? (ie. one would have a "beacon" role and the other would just receive said beacons.)

[themadhippy]

hows about rfid tags,some of those can read a good few meters.