Russell,

When Row1 is 0 and Row2,3 is 1 the only buttons will
work is 1,2,3.
Row2 and Row3 are logic 1 and col1, col2, col3 have pull-up (logic 1)
no thing will happen when they pressed. 
I am looking at the schematic I am sure about that.

Andre




_________________


In your example:

With no diodes -

Row 1 low.
Switches 1,6,7 are possibly legal (not 1,2,3)

BUT
If you simultaneously press 1,2,5
then key 6 APPEARS to be pressed due to the current path Row 1 - SW1 - 
SW2 - SW5 - column 2
As Row1-column2 is the legal result of SW6 it looks like SW6 is 
pressed.

As this will be decoded when SW2 and Sw5 are NOT being decoded (as 
they are on ROW2 which is high) you will not "notice" this possible 
error if you do not examine the resultant key scan table after a 
complete keyboard scan.

The best you can do without diodes is decide whether a key activation 
MAY be spurious.
eg if SW1 SW2 SW5 are pressed then you WILL see SW6 pressed whether it 
is or isn't.

This sort of problem is usually only an issue when you have users who 
depress multiple keys simultaneously. This is most likely to occur 
during eg high speed typing or data entry and far less likely to occur 
with Joe Average keyboard entry.

Adding the diodes prevents this path.
Diode protection would be mandatory when using keyboards whose design 
encourages multiple adjacent key depressions and then sorts out the 
intended key in software such as the very small "thumb entry" 
keyboards on some PDAs.


Without the diodes you can correctly decode any key sequence that does 
not lead to such spurious paths. Generally you will be OK with up to 
any 2 keys pressed. Some combinations of 3 or even more keys may be 
OK.
eg here Sw1, Sw2 Sw9 should be OK as their is no "orthogonally 
connected" path between the 3 keys.



        Russell McMahon



-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist