 Notice. New forum software under development. It's going to miss a few functions and look a bit ugly for a while, but I'm working on it full time now as the old forum was too unstable. Couple days, all good. If you notice any issues, please contact me. |
Forum Index : Microcontroller and PC projects : 4 x 4 matrix keypad
Page 1 of 2   |
|||||
| Author | Message | ||||
| viscomjim Guru  Joined: 08/01/2014 Location: United StatesPosts: 925 |
Hi All, Did a bit of digging in the forum and could not find any info or code to read and effectively use a 4 x 4 matrix key pad like the typical one attached (picture). This is a 4 row and 4 column matrix, basically using 8 I/O pins. I did find the matrix keyboard to analog pin circuit and code, but does anyone know of some good code to scan and read one of these using the digital pins instead? I would like to interface one of these to micromite and would appreciate any info one may have. Maybe using a tick interrupt routine? Thanks to all!!! 
|
||||
Grogster Admin Group  Joined: 31/12/2012 Location: New ZealandPosts: 9308 |
Perhaps there is something in the MMBasic library available from Geoff's site? From what I understand, you connect the rows to outputs(or is that inputs?), and the columns to inputs(or is that outputs?), then the code pulses the output lines one after another, looking for one of the inputs to go high, indicating that that button on that row and column has been pressed. Should not be that hard to code - I might have a go at this as a MM experiment in the next few days. Others will no doubt chime in with their ideas.... EDIT: I've just had a sniff around in the latest MMBasic library, but can't find anything there. Would be useful. I will play around, and see if I can come up with something, and I am quite sure that other members will add their own ideas too. Smoke makes things work. When the smoke gets out, it stops! |
||||
| MOBI Guru Joined: 02/12/2012 Location: AustraliaPosts: 819 |
Some years ago, I did a I2C version of this same (or at least very similar) key pad. The first one I did was for a 12 key unit (4x3) but as there were spare pins on the pic, I did a 4x4 version. I thought I posted it on TBS but maybe not the code? Don't remember, but I am more than happy to re-hash it and provide the cct diag and source/hex code. It was for the PIC16F88/819 chips. The keypad in downwind's solar tracker currently being discussed uses the 4x3 version. David M. |
||||
| Grogster Admin Group Joined: 31/12/2012 Location: New ZealandPosts: 9308 |
That would be useful - go right ahead!  Smoke makes things work. When the smoke gets out, it stops! |
||||
| MOBI Guru Joined: 02/12/2012 Location: AustraliaPosts: 819 |
I found the 4x3 keypad on TBS with a zip file with code etc. http://www.thebackshed.com/forum/forum_posts.asp?TID=5379 Meanwhile, I'll see if I can scrounge up the data on the 4x4 keys David M. |
||||
| viscomjim Guru Joined: 08/01/2014 Location: United StatesPosts: 925 |
Hi Early Responders! I am trying to use just the micromite to read the keypad (using 8 pins) and run a 4 bit lcd (2004) (using 6 pins). This will still leave 7 I/O pins to use for other things. I am trying to keep the part count down to a very minimum so using another chip for a keypad would not work in this instance. As the micromite is relatively cheap for what it can do, this allows for a low cost yet very powerful little controller with minimum components. Thanks again for any and all help on this keypad routine. I did not see anything in the library unless I am going blind... |
||||
| MOBI Guru Joined: 02/12/2012 Location: AustraliaPosts: 819 |
I understand. Scanning the keypad wasn't difficult. All I did was cycle a "1" through the 4 outputs and read the inputs which gave me a 16 bit count. There were two options for key value. One was a key number from 1 to 16 and the other was to feed the key value to an array which contained the ascii key values or what ever you like. Personally, I think it is a lot of wiring which can be reduced to 2 wires using I2C which the MM and microMite (uMite still in beta) do very well. The other thing about I2C is that the coding is very simple, besides, you can "plug and play" additional modules. David M. |
||||
| akashh Senior Member  Joined: 19/01/2014 Location: IndiaPosts: 115 |
You would set up a matrix like this: R1 ----- x -- x -- x -- x | | | | R2 ----- x -- x -- x -- x | | | | R3 ----- x -- x -- x -- x | | | | R4 ----- x -- x -- x -- x | | | | C1 C2 C3 C4 Connect each button to the row and column at the x. If you have a ready made keypad the chances are it will be wired like this, you can check with a multimeter. Set pins R1 to R4 as outputs, and C1 to C4 as inputs. Now periodically set each Row pin to high and see whether any column inputs are high. I am not sure whether mmbasic supports this but I would try this: create an interrupt routine when any pin c1 to c4 goes high. Then make all row outputs high. Now whenever any key gets pressed the interrupt will get called, and you can quickly change the row outputs to determine which row was pressed. This would eliminate the need for regular polling. Edit: sorry the matrix didn't come out as well as I would like it to |
||||
| Grogster Admin Group Joined: 31/12/2012 Location: New ZealandPosts: 9308 |
Have to agree with MOBI here. If I was doing any kind of keypad and LCD idea, I would be inclined to use I2C keypad, and an I2C LCD module arrangement too, both on the I2C bus with the MicroMite(or other) as the master. My 2c only. Nothing wrong with the idea of squeezing it all into the one chip though. I suppose if you mount the keypad above the PCB, and do the same with the LCD, then the one PCB can have the processor, the LCD and the keypad, and include all of the connections on the one board. Each to their own. Smoke makes things work. When the smoke gets out, it stops! |
||||
| viscomjim Guru Joined: 08/01/2014 Location: United StatesPosts: 925 |
That interrupt thing sounds like a good start. I believe any pin on the micromite can be used to have an interrupt routine occur. I will approach this that way. That way, I won't be polling a lot. I do agree with the I2C thing, but in this particular instance, hopefully just one chip will be used (micromite). I will keep trying to get this going and am searching the web on this subject. Keep the great ideas coming! Maybe this solution can be a library item soon. |
||||
TassyJim Guru Joined: 07/08/2011 Location: AustraliaPosts: 6101 |
You might need pulldown resistors to stop unpressed lines from floating high. I would make all lines inputs and only change one at a time to a high output. If you leave some lines as low outputs and two buttons are pressed at the same time, you may have a short circuit between the high and low outputs. All this is easy to do with the Maximite. I do have one of the keypads shown but have yet to do any code for it. Jim VK7JH MMedit MMBasic Help |
||||
| MOBI Guru Joined: 02/12/2012 Location: AustraliaPosts: 819 |
Resistors (10k) were required in my i2c model. As well, I set all pins as inputs and only enabled 1 row at a time (except when looking for "keypress" where all rows were activated). Also, seeing as there were 12 pins available, the keyboard could accommodate 32 keys. So, two of the membrane pads could be connected giving a more powerful keypad. David M. |
||||
| isochronic Guru Joined: 21/01/2012 Location: AustraliaPosts: 689 |
There was a scanned keypad + lcd arrangement for the bare-bones MM in the ideas section of "Silicon Chip" a while ago, I think the basic code was put on their website (?) |
||||
MicroBlocks Guru Joined: 12/05/2012 Location: ThailandPosts: 2209 |
You also have to take care of 'key bounce', especially when using an interrupt. Instead of registering a single keydown you might have tens. To debounce it you wait a little (5-10ms) and then read the same column and row again to see if it is still pressed. Then next scan should run about 50ms later, this will give you about a 20 per second repeat rate or you could wait until the key is released before starting a new scan. Also the release can have bounce. Some procedure. Microblocks. Build with logic. |
||||
elproducts Senior Member Joined: 19/06/2011 Location: United StatesPosts: 282 |
If your open to a free compiler and 8-bit part I have a keypad and LCD code examples for Great Cow Basic at this site: http://www.greatcowbasic.com/sample-projects.html GCB has a keypad and an LCD command built into its library. www.elproducts.com |
||||
| MOBI Guru Joined: 02/12/2012 Location: AustraliaPosts: 819 |
The I2c module takes care of all the debounce etc and having registered a key pressed, will not accept another key until the master has read the key and reset the keypad. It is easy to check if a key has been pressed by testing a control byte. The control byte is used to reset the keypad and to test for a key press. If a key is held down, it will be recognised pretty quickly and can be read and the programme redirected as needed other wise the programme drops through and only uses a few milliseconds. Hope that makes sense. David M. |
||||
| paceman Guru Joined: 07/10/2011 Location: AustraliaPosts: 1329 |
If you're open to using a different technique with the 4x4 and 4x3 keypads, do a forum search on "4x4 Matrix board with one analogue input" using "Topics". I posted a method (17 July 2012 - with photos and code) that uses a two-wire approach, rather than the eight needed for the scanning approach. It finds the pressed key by reading one analog input and works pretty well. There are some drawbacks: 1. Extra components needed, but they're very cheap - resistors, one cap, one LM334Z plastic TO9 package. 2. "Double" keypresses, i.e. Ctrl C etc. can't be used. Greg |
||||
| MOBI Guru Joined: 02/12/2012 Location: AustraliaPosts: 819 |
Yes, that was pretty nifty. Just goes to show that there are lots of options and innovations. An analogue pin on a PIC i2c slave can also be used to select up to 8 possible addresses using a voltage divider and a bit of firmware. David M. |
||||
| viscomjim Guru Joined: 08/01/2014 Location: United StatesPosts: 925 |
Looks like the I2C and analog read are pretty popular. Wouldn't you have to scan the keyboard with the I2C the same way you would with the micromite, just using an additional pic? So I'm thinking if this is so, we could just use the micromite to do this and eliminate one chip from the circuit. I will keep digging into the scan method. I found a couple of lines of code doing this with the picaxe, so I will try and port to micromite, shouldn't be too hard. Also, there is an article in the silicon chip mag archives, so I will try to check that out also. I still like AKASHH's idea of implementing micromites ability to be interupted on any pin to start the "which key press" routine without having to constantly poll the keypads lines. Thanks for all your help. |
||||
| MOBI Guru Joined: 02/12/2012 Location: AustraliaPosts: 819 |
Yes and no. The pic attached to the key pad does all the work i.e. scanning and code production. It is an I2C slave and to access the key pressed information (get a key value), one only needs to issue a single i2c instruction to the i2c bus from say, a MaxiMite and the key pad will respond with the key data. e.g. i2c read kpad,0,3,0,key,control Where kpad is the i2c address of the key pad, key is the key value and control is a handshake byte David M. |
||||
| Page 1 of 2 |
|||||
 Print this page |
