[time-nuts] Happy Holidays

[Bruce Griffiths]

Bruce Griffiths wrote:
> Poul-Henning Kamp wrote:
>   
>> In message <4956D31B.8040809 at kasperkp.dk>, Kasper Pedersen writes:
>>   
>>     
>>> Bruce Griffiths wrote:
>>>     
>>>       
>>   
>>     
>>> The charge redistribution adc would have been good, true, but there 
>>> wasn't one in my junkbox.
>>>     
>>>       
>> I'm pretty sure the ADC in most, if not all, micrcontrollers are
>> charge redistribution types.
>>
>>   
>>     
> That is true for both the Microchip PIC16F917/6/4/3 series and the
> ATMega128 (however the low bandwidth internal amplifier should not be
> used for this application).
>
>
> Bruce
>
>   
The PIC16F688 used by Richard also has a charge redistribution ADC.
Consequently an external buffer amplifier is probably not necessary.
However a delay of about 5us after the termination of the interpolator
ramp is necessary to allow the charge on the ADC sampling capacitance
and the ADC internal amplifier to settle before beginning the ADC
conversion cycle.
When using the built in multiplexer to switch between analog inputs some
charge injection effects will occur.
A dedicated charge redistribution ADC per interpolator is preferable
especially if very high resolution is required.

The PIC16F688 ADC input pin leakage current is specified at 500nA,
however this may be due more to test time limitations than the actual
leakage.
The actual leakage current of the ADC input pin can be readily measured
by using a DVM to measure the voltage drop across a series resistor in
series with the analog input.

If the ramp capacitor charging current is sufficiently high the effect
of such leakage will be small.

Bruce