Post by thread:Keypad interfacing

OK, I got bored last night and got to thinking if I could interface a 3x4 matrix keypad using less than 7 lines without shift registers etc. I suspect it must have been done, but I figured out a (possible) solution which uses only 4 lines and some diodes. I haven't got any hardware to try this out on at present, but I'd be interested in any comments. I started writing the code in Hitech C but figured that I had to get to grips with PIC assembler some time so this is my first attempt, and is therefore probably full of standard mistakes and is most definately in need of optimisation. Check it out on: http://www.mikerj.clara.net/keypad.html and let me know why it won't work :o) Regards Mike Rigby-Jones spam_OUTmrjonesTakeThisOuTnortelnetworks.com

Check the Microchip website. As I recall, a Russian did a fantastic general-purpose keypad which significantly reduces I/O rqts using diodes. Andy At 01:13 PM 12/2/98 -0000, you wrote: {Quote hidden}>OK, I got bored last night and got to thinking if I could interface a 3x4 >matrix keypad using less than 7 lines without shift registers etc. I >suspect it must have been done, but I figured out a (possible) solution >which uses only 4 lines and some diodes. > >I haven't got any hardware to try this out on at present, but I'd be >interested in any comments. I started writing the code in Hitech C but >figured that I had to get to grips with PIC assembler some time so this is >my first attempt, and is therefore probably full of standard mistakes and is >most definately in need of optimisation. > >Check it out on: >http://www.mikerj.clara.net/keypad.html > >and let me know why it won't work :o) > >Regards > >Mike Rigby-Jones >.....mrjonesKILLspam@spam@nortelnetworks.com > ================================================================== Andy Kunz - Statistical Research, Inc. - Westfield, New Jersey USA ==================================================================

Yup, you're right, as always! This guy manages to connect bucket loads of switches with just 4 lines, however they have to be separate switches, not a matrixed array. For those who may be interested it is app note DD4012 at: http://www.microchip.com/Download/Appnote/Category/12CXXX/Logic/index.htm Regards Mike Rigby-Jones mrjonesKILLspamnortelnetworks.com > Check the Microchip website. As I recall, a Russian did a fantastic > general-purpose keypad which significantly reduces I/O rqts using diodes. > > Andy Kunz - Statistical Research, Inc. - Westfield, New Jersey USA >

Please ignore my previous ramblings regarding the microchip web site.. The correct url is appnote D4012 at: http://www.microchip.com/10/Appnote/Category/12CXXX/Logic/index.htm Mike Rigby-Jones .....mrjonesKILLspam.....nortelnetworks.com

On Wed, 2 Dec 1998, Rigby-Jones, Michael [PAI01:4837:EXCH] wrote: > OK, I got bored last night and got to thinking if I could interface a 3x4 > matrix keypad using less than 7 lines without shift registers etc. I > suspect it must have been done, but I figured out a (possible) solution > which uses only 4 lines and some diodes. > > I haven't got any hardware to try this out on at present, but I'd be > interested in any comments. I started writing the code in Hitech C but > figured that I had to get to grips with PIC assembler some time so this is > my first attempt, and is therefore probably full of standard mistakes and is > most definately in need of optimisation. I haven't seen your code yet but I have to say that AN512 allows one to connect buttons through a SIPP R ladder with good results ;) 8 keys per ladder is the usual limit, you can use 3 ladders with 4 pins to interface 24 keys to the PIC. hope this helps, Peter

Connect a 4 x 4 keyboard using only one pin. There was an article in 'Electronic Design' May 1998 issue that connected a 4 x 4 key matrix to a single I/O pin. By using resistors between the columns and rows and connecting them to a cap. The cap is then discharged (by making the I/O and output and driving it low), then change the I/O to an input and measure the length of time it takes to charge the cap. michael You may leave the list at any time by writing "SIGNOFF PICLIST" in the body of a message to EraseMELISTSERVspam_OUTTakeThisOuTMITVMA.MIT.EDU.

On Wed, 2 Dec 1998, Michael Hagberg wrote: > Connect a 4 x 4 keyboard using only one pin. > > There was an article in 'Electronic Design' May 1998 issue that connected a > 4 x 4 key matrix to a single I/O pin. By using resistors between the columns > and rows and connecting them to a cap. The cap is then discharged (by > making the I/O and output and driving it low), then change the I/O to an > input and measure the length of time it takes to charge the cap. I haven't seen this article but I'd like to say that solutions involving R chains longer than 8 keys are seldomly used in practice, for noise and repeatability reasons. The solution that I had suggested in another message could be used with 16 keys per chain because it uses an extra IO pin for calibration. These considerations apply equally well when a spare 8-bit ADC is used to read the chain. hope this helps, Peter

> Connect a 4 x 4 keyboard using only one pin. > > There was an article in 'Electronic Design' May 1998 issue that > connected a 4 x 4 key matrix to a single I/O pin. The article is online: http://www.penton.com/ed/Pages/magpages/may0198/ifd/0501ifd.htm#5 Adriano