AVR-Chat

Thanls for the feedback.  I have the datasheet for the display.  It 
suggests a pot of 10k or 20k.  I used a simple voltage divider with 
2 10k resistors.  After experimentation, I found that the optimal 
voltage divider delivers  pure GROUND to the contrast pin.  So that 
is how I have it set now.

In the end, I had the physical wiring correct, but despint making 
corrections to the include file, I corrected the WRONG file.  So, 
the s/w was trying to use the wrong pins!  

-Tony


--- In AVR-Chat@yahoogroups.com, "Ivan Vernot" <ivernot@o...> wrote:
> Another thing to keep in mind is that a voltage is required to set 
the
> contrast.
> The voltage however varies for different LCDs (Some are happy with 
a 0-5V
> range others need a -Ve supply. You will need to check the data 
sheet
> 
> One simple thing to try is to wire a potentiometer (the value is 
not
> critical , 1K - 10K is fine) between GND and 5V with the wiper 
going to the
> contrast pin (sometimes labelled VLCD).
> As you vary the voltage (by turning the pot) you should see the 
LCD 'dots'
> change intensity.
> 
> It is easy to get caught out thinking the LCD is dead/not 
initialised and
> its just the contrast setting.
> Although, to he honest, in this case it is MOST likely the init 
sequence as
> David points out. :-)
> 
> Have a look at AVRfreaks.net -  I am sure that there would be some 
sample
> code for a text based LCD there
> 
> HTH
> Ivan
> 
> 
> ----- Original Message ----- 
> From: "Dave VanHorn" <dvanhorn@d...>
> To: <AVR-Chat@yahoogroups.com>; <AVR-Chat@yahoogroups.com>
> Sent: Saturday, April 23, 2005 11:27 AM
> Subject: Re: [AVR-Chat] New LCD will not work.
> 
> <snip>
> >
> > LCD displays generally don't display anything
> > useful on powerup, but you should be able to vary
> > the contrast voltage, and get black squares.
> > If you can't get that, then even if you are
> > talking to it right, you may never know it.
> >
> > Other than that, it's TIMING, TIMING, TIMING.
> > Do ALL the init steps, and insure that all your delays meet spec 
plus a
> bit.
> > You can always trim the delays back to spec+1 after you get it 
talking.
> >
> > It can be tedious, since you don't usually get
> > anything viewable until you get all of it right.
> >
> > I can sympathize, I just went through this with a
> > small 16 bit color LCD on another project, with a
> > VERY long init sequence, where the parts that
> > were correct, were poorly documented. :)
> >
> >
> <snip>