Lpc2000

Hi,

Im using the onboard ADC for a project I am working on. However I 
seem to be getting some strange values and it isnt acting as I 
expect it to.

Two things mainly:

One, when I take a voltage input of a constant voltage (0.5 * 3.3V) 
I get fluctuating values from the ADC which I output to 
hyperterminal. This is very strange as they should be all the same 
value.

Two, when I take the value from ADDR I place it into an int. However 
I am saving these values into an 8-bit value array so I have to 
shift right twice (due to the ADC being 10-bit). Or so I thought. 
The values I get from this array seem to be about right, but the 
resolution is in steps of 8 or 16 not the expected 1. This is no 
good for my application. I have also tried dividing by 4, but this 
produces even stranger results! 

Any help would be appriciated.

Thanks,

FG

At 09:28 PM 11/22/04 +0000, you wrote:


>Hi,
>
>Im using the onboard ADC for a project I am working on. However I
>seem to be getting some strange values and it isnt acting as I
>expect it to.
>
>Two things mainly:
>
>One, when I take a voltage input of a constant voltage (0.5 * 3.3V)
>I get fluctuating values from the ADC which I output to
>hyperterminal. This is very strange as they should be all the same
>value.

How much are they fluctuating?  I'd expect some fluctuation.  Also what 
does your signal look line on an oscilloscope (AC coupled at about 5mV/div)?


>Two, when I take the value from ADDR I place it into an int. However
>I am saving these values into an 8-bit value array so I have to
>shift right twice (due to the ADC being 10-bit). Or so I thought.
>The values I get from this array seem to be about right, but the
>resolution is in steps of 8 or 16 not the expected 1. This is no
>good for my application. I have also tried dividing by 4, but this
>produces even stranger results!

Have you compared before and after values?  If the division is not tracking 
the input values it's probably just a simple arithmetic error.

Robert

" 'Freedom' has no meaning of itself.  There are always restrictions,
be they legal, genetic, or physical.  If you don't believe me, try to
chew a radio signal. "

                         Kelvin Throop, III

At 04:43 PM 11/22/04 -0500, you wrote:
> >Two, when I take the value from ADDR I place it into an int. However
> >I am saving these values into an 8-bit value array so I have to
> >shift right twice (due to the ADC being 10-bit). Or so I thought.
> >The values I get from this array seem to be about right, but the
> >resolution is in steps of 8 or 16 not the expected 1. This is no
> >good for my application. I have also tried dividing by 4, but this
> >produces even stranger results!
>
>Have you compared before and after values?  If the division is not tracking
>the input values it's probably just a simple arithmetic error.

Also note that the A/D result register contains more than the result. The 
top 16 bits are various status fields and the low 6 bits are zero (allowing 
compatible expansion to 16 bits of resolution).


" 'Freedom' has no meaning of itself.  There are always restrictions,
be they legal, genetic, or physical.  If you don't believe me, try to
chew a radio signal. "

                         Kelvin Throop, III

--- In lpc2000@yahoogroups.com, Robert Adsett <subscriptions@a...> 
wrote:
> At 04:43 PM 11/22/04 -0500, you wrote:
> > >Two, when I take the value from ADDR I place it into an int. 
However
> > >I am saving these values into an 8-bit value array so I have to
> > >shift right twice (due to the ADC being 10-bit). Or so I 
thought.
> > >The values I get from this array seem to be about right, but the
> > >resolution is in steps of 8 or 16 not the expected 1. This is no
> > >good for my application. I have also tried dividing by 4, but 
this
> > >produces even stranger results!
> >
> >Have you compared before and after values?  If the division is 
not tracking
> >the input values it's probably just a simple arithmetic error.
> 
> Also note that the A/D result register contains more than the 
result. The 
> top 16 bits are various status fields and the low 6 bits are zero 
(allowing 
> compatible expansion to 16 bits of resolution).
> 
> 
> " 'Freedom' has no meaning of itself.  There are always 
restrictions,
> be they legal, genetic, or physical.  If you don't believe me, try 
to
> chew a radio signal. "
> 
>                          Kelvin Throop, III

Thanks for that. I actually noticed that as I was reading through 
the user manual for inspiration, I then found an example with a 6 
times right shift and an arithmetical and worked on to the output of 
the ADC. I always thought that was to extract the information for 
that particular users' program!

Stll adding that in and a few other slight modifications and my 
project is working fine.

Thanks,

FG