On Fri, Sep 12, 2003 at 07:34:34AM -0500, Patrick Richards wrote:
> Byron, thank you for your reply.  You have given me several things to
> think about.  After seeing your code snippet, I have decided I should
> really try to work out this whole problem myself instead of hacking up a
> complete piece of code I found from Microchip.  I have started to sketch
> out the problem in pseudo code.

Good. You'll generally be able to better utilize the tools you build instead
of the ones you borrow.

> [Languages snipped for brevity]
>
> >         CBLOCK   0x2C   ; CBLOCK will auto the address.
> >                         ; 0x2C works for most 16F style pics.
> >         display         ; Make sure to init to 0
> >         ENDC
>
> > Update  incf     display,W  ; Next display
> >         andlw    0x03       ; Keep in range
> >         movwf    display    ; And stick it back in display
> >         clrf     PCLATH     ; Paranoia. Keep in 1st page
> >         call     disp2bit   ; Convert to a bit number
> >         movwf    PORTA      ; and write to port A.
> >         return              ; C-YA!
>
> > ; disp2bit presumes a couple of things. One is that W has the display
> > number.
> > ; Two is that PCLATH is cleared. Fix as necessary.
>
> > disp2bit
> >        andlw     0x03       ; Paranoia. Just in case W > 3
> >        addwf     PCL,F      ; Jump table add
> >        retlw     0x0E       ; Clear PORTA:0
> >        retlw     0x0D       ; Clear PORTA:1
> >        retlw     0x0B       ; Clear PORTA:2
> >        retlw     0x07       ; Clear PORTA:3
>
> Has someone on this PICList posted standards for coding and commenting?

Not that I know of. Everyone has a different style.

> I mean, how many tabs to indent for code and comments, what to
> capitalize, etc.

I usually use tabs in my code. Note I put spaces above because I have no clue
what effect a tab will have on the 1500 different E-mail programs that are
used to read this list.

>
> > You have another issue to think about. It's very likely that if you
> are
> > using a port pin to drive the cathode, that the display will be too
> dim to
> > be useful.
>
> You're definitely right about this.  I prefer using a transistor for
> each display.  It helps me to keep all the work in the PIC instead of
> depending on logic in another chip, but either method would work.  Have
> far along are you on segment multiplexing?

I'm in the contemplation phase. ;-)

>  That would solve the power
> problem for my situation, but I expect it would require quite a bit more
> code.

Not much more code at all. And definitely no additional hardware. I'd probably
do it the same way as above using a segment counter and a 8 element jump table
to convert the count to a bitmask. Then mask the bitmask against the current
value for the display. I believe you said you had common cathode, so you can
just write the masked value directly into PORTB.

There are quite a few advantages (theoretically) to segment multiplexing:

1) You may not need high powered hardware since only one segment is on at a
   time.
2) Depending on how you wire whole displays you may get brightness variation
   based on the number of segments that are lit. Since segment multiplexing
   only turns on one segment at a time, you should get equal brightness.

BAJ

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
email listserv@mitvma.mit.edu with SET PICList DIGEST in the body