Hi!
16-Mar-98 13:27 Steve Lawther wrote about "Re: Problem reading from LCD in 4 bit
 mode":

 >>      Brain,
 >>      Try the changes below - you read the data out of the LCD when E is
 >>      asserted
 >>
 >>      Steve Lawther
 >>
 >>
 >> Read_Data
 >>         call     Wait_Busy              ;wait for LCD=20
 >>         bsf     STATUS, RP0     ; Select Register page 1=20
 >>         movlw     0xF0          ; Set port to input
 >>         iorwf   LCD_DATA_TRIS,W ; Only set upper half of port=20
 >>         movwf     LCD_DATA_TRIS
 >>       bcf     STATUS, RP0     ; Select Register page 0
 >>         bsf     LCD_CNTL, RS    ; Set LCD for Data mode=20
 >>         bsf     LCD_CNTL, R_W   ; Setup to read
 >>         bsf     LCD_CNTL, E     ;toggle E for LCD=20
 >>         nop
 >>         movf    LCD_DATA, W     ;********
 >>         bcf     LCD_CNTL, E
 >>         ;not here movf    LCD_DATA, W     ; Read upper=20
 >>         nibble data andlw   0xF0            ; Mask out=20
 >>         lower nibble movwf   Temp
 >>         bsf     LCD_CNTL, E     ;toggle E for LCD=20
 >>         nop
 >>         swapf   LCD_DATA, W     ;**********
 >>         bcf     LCD_CNTL, E
    .....

    Yes, Brian, it was the capacitance. After "BCF LCD_CNTL,E"
the LCD's became hi-Z (and PIC's pins was hi-Z). So, you read
"the memory cells made of cable capacitance etc." :). And this
"cells" are discharged by LCD's pull-ups. This is not a specific
problem of 4-bit mode.

    Steve insert NOP's after BSF LCD_CNTL,E. These *may* be
unnecessary. Using short cable 4MHz PIC and LCD (I'd use 4-bit
mode too) work without these NOP's according to datasheet and
experiment. But when the cable length is about 30..50cm or
greater (I mean usual ribbon cable) LCD's outputs can't drive
long cable's capacitance and need extra time to recharge it.
NOP's provide this time.

    Of course, these effects occurs only while reading LCD.
AFAIK many people don't read LCD at all (including BUSY flag).

--

                                    Vladimir M. Klochko