Maker Pro
Maker Pro

Looking for a Micro Transducer

sixpak

Sep 25, 2013
7
Joined
Sep 25, 2013
Messages
7
I am looking for a micro transducer of some sort that is capable of sending at least 13 discrete or distinct signals through a thin, non-metallic material such as plastic, wood or fiberglass whereas the receiving unit must be able to decode or decipher each signal. The transducer can use RF, sound, capacitance, frequency or whatever will work.

Example: Let’s say I have thirteen pieces on one side of a game board and fifty sensors on the other side. I would like to know which piece is directly above each sensor.

There are a couple more requirements. The transmitters must be approximately ¾” or less and the decoders must approximately 1” or less. The most important thing is they must be inexpensive. After all, we are talking about a board game.:)

Thanks for your suggestions.

Drew
 

goldfist

Sep 18, 2013
31
Joined
Sep 18, 2013
Messages
31
I'd go with programming each one to be ID'd by their pulse signature so that the game board sends out a call to all 13 devices, and all 13 devices report back. Each device can have a unique signature in which when the signature is a match it reports back so that not all 13 are trying to report back at the same time.

So you could have a board with a matrix of coils which are used to transmit and receive the pulses to poll devices by location on X,Y and know of their placement. Without knowing your budget and skill level though I am unsure if you can achieve this or not.

But if the game board sent out a poll such as

10001 to unlock an output from device 1
10010 to unlock an output from device 2
10011 to unlock an output from device 3
10100 to unlock an output from device 4
10101 to unlock an output from device 5
10110 to unlock an output from device 6
10111 to unlock an output from device 7
11000 to unlock an output from device 8
11001 to unlock an output from device 9
11010 to unlock an output from device 10
11011 to unlock an output from device 11
11100 to unlock an output from device 12
11101 to unlock an output from device 13

And the game board waits for response and confirms receiving response and processes this data to your program or does not detect the response and so it calls out again to it until it gets the response and likely alerts to an error condition where a game piece is not responding. *If it reports back, then call to device 2 and 3 and so on and register on a matrix playing field as to where each device is.

The first bit is a wakeup and the following 4 bits are the address of the device for its unique ID to be called. To make it simple, lets say it responds with its Unique ID back after a short delay so the game board says Device 1, and 1/4 sec delay and Device 1 HERE basically at strongest reading at X=32, Y=24 as this is the coil that received the message that it is there so this is its location.

It processes this polling until all 13 devices report back. To make this work however and low cost, I would interface this with a PC, however this could also be done with a Raspberry PI I suppose. My choice would be with a PC to run the program and test it as for C++ is my programming and interface language of choice and I havent worked with Raspberry PI micro computer yet.

Not to scare you off on this, but your probably looking at an initial build cost far beyond what you have to play with in your budget. I have many many circuits drawn up that usually dont come to prototype build because I dont have the budget to make it. So it remains on paper as soemthing that would work and likely work out really well if I had funding to make it happen.

The board itself will be pretty easy to make with a group of coils mounted under the board under a thin enough material so that the magnetic transmission can pass through without too much loss. The hardest and most expensive part will be the circuits that are embedded in each of the game pieces that need to listen and respond with their name when their matching Device ID is called.

Below are methods that are not a match to your description but position detection methods.

Its too bad you dont have just 2 game pieces as for you could set one piece to South pole facing the game board and the other with North pole of magnet facing the game board and detect by pole ID.

The count of 13 in a board game I find interesting because most board games have an equal number of pieces between 2 players, so I am curious as to what type of game this is, if that info can be shared or not.

An extremely low low cost board game to keep track of pieces, now this is the absolute lowest tech board game and primitive, would be to have pieces that physically plugged into places on the game board or made electrical contact with contacts on the game board. Each of the 13 pieces can have a unique circuit and where they are placed on the X and Y is identified by their circuit signature. This could be performed for less than $100 given its so primitive and low tech. You still would want to interface this with a computer probably though to track game piece locations etc unless you can squeeze the logic into a low cost micro computer. If you already have a computer you can write a program that polls the game board for every square of the X,Y playing field and each square that has a game piece in it will report a different circuit config back to a game piece identifier/translator that is then reported back to the PC. You would need to program up the code for free on your own time to keep costs to bare minimum so that only the hardware that you need composes the costs of the build with time to build it not measured in monetary value.

The game pieces by the way to clarify how low tech they are are just bread board and traces with no components on them. Each one when in electrical contact with the game board will report back with a solid binary signature that is hard wired so if you have 5 contacts on each game piece you have your common and 4 other contacts which are the binary signature, so if contact 1 is common and its connected directly to contact 4 and 5, with 5 being the first binary digit 0 or 1 and contact 4 being the second binary digit of 0 or 2 and contact 3 being the third binary digit of 0 or 4 and contact 2 being the fourth binary digit of 0 or 8, ( contact 1 common tied directly with contact 4 and 5 reports back as hard wired Device ID # 3 )
 
Last edited:

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
One possible solution is to place an RFID tag on the bottom of all the pieces.

Below each square you can place a small coil. With a little ingenuity, you can connect these coils, one at a time to an RFID reader. That will give you a signal for each.

To save time scanning the entire board, you could also place a small magnet n each piece. Then, a hall effect sensor under the coil (I would have a small ferrite rod) could determine if a piece was on the square..

Smart software could see that a piece had been moved and scan just the positions that have changed.

I would use 150kHz RFID as it's a lot easier to switch these signals than for higher frequencies.
 

goldfist

Sep 18, 2013
31
Joined
Sep 18, 2013
Messages
31
To save time scanning the entire board, you could also place a small magnet n each piece. Then, a hall effect sensor under the coil (I would have a small ferrite rod) could determine if a piece was on the square..

Good Idea! Using Hall Effect Sensor to know which squares of the board have pieces to not waste time checking all.

My concern with RFID was that the size of the squares we are working with is unknown and I was thinking that multiple coils would pick up the same response or ID, but I suppose you could adjust the range of the RFID com by the field strength at the coil to avoid multiple squares as coming up as position positive, but with a Hall Effect Sensor that would definately pin point to that square only if there is any multiple positive read coils.
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
I've re-wound an RFID coil (it was about 40 turns with an air core) using a ferrite rod and I found it to be very directional and insensitive off axis.

You would have to do some experiments yourself though. I was using RFID tags that were about 20mm in diameter.

I used an inductance meter to determine when my coil had the right inductance. I can't remember how many turns it was.
 

sixpak

Sep 25, 2013
7
Joined
Sep 25, 2013
Messages
7
Follow up

goldfist:

Thank you for your very comprehensive response. Your understanding of my requirements is right on target. I will likely be programming in C or C++. I will be using a matrix very similar to what you describe. I intend on using a microprocessor contained within the board to scan the matrix. My only problem is finding an off-the-shelf solution for the coils that you describe. It seems to me that Steve may have hit on a solution, but it would interest me very much to know how you would make your transmit and receive coils, since that is my biggest knowledge gap. Links to information are just fine.

On the subject of low cost, low tech solutions, reed switches in the board and magnets in the pieces would work just fine. The only problem is the ID.

Thank you so much for your input.

Steve:

I had not considered RFIDs because of goldfist’s concern regarding field strength. But now you have peaked my interest by your rewinding experiment. I would really like to know more details - what materials: RFID, ferrite rod, tags, etc. Again, links would be fine.

Thank you so much.

Drew
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
I would really like to know more details - what materials: RFID, ferrite rod, tags, etc. Again, links would be fine.

I'm going away for a couple of days so I won't be able to take pictures and describe what I did until I get back.

PM me or post a reminder here in 4 days if I have not gotten back to you.

Basically I wound (using similar enameled wire) a winding around a ferrite rod I had purchased from China for the experiment.

I am quite likely to be able to point you to where I purchased them, how many turns, and the RFID reader I used (it was a $9 module)
 

sixpak

Sep 25, 2013
7
Joined
Sep 25, 2013
Messages
7
I'm going away for a couple of days so I won't be able to take pictures and describe what I did until I get back.

PM me or post a reminder here in 4 days if I have not gotten back to you.

Basically I wound (using similar enameled wire) a winding around a ferrite rod I had purchased from China for the experiment.

I am quite likely to be able to point you to where I purchased them, how many turns, and the RFID reader I used (it was a $9 module)

Thanks Steve. See you in 4 days.

Drew
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
As I've advised sixpak, I'm having a little trouble finding my modified module.

However, this is the module I'm using. Note that it has a USB interface, which may be a problem in an embedded microcontroller project. However, when I find the modified coil, the same technique should apply to pretty much any other 125kHz RFID reader.
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
Just as a bit of an update, I also have one of these that is better suited to interfacing to a microcontroller.

The coil on this looks identical, so I expect the same technique will work.
 

sixpak

Sep 25, 2013
7
Joined
Sep 25, 2013
Messages
7
Just as a bit of an update, I also have one of these that is better suited to interfacing to a microcontroller.

The coil on this looks identical, so I expect the same technique will work.

Thank you Steve. I'll give it a shot.

Drew
 
Top