Note that going to GND in-between channel switching will not get rid of
spikes, it will just make them negative and independent of the values on
other channels. Instead of crosstalk you will have a gain error (albeit a
small one)

On Thu, May 5, 2016 at 10:44 AM, Van Horn, David <
david.vanhorn@backcountryaccess.com> wrote:

> Yup.  Just applied that to all analog inputs.    I'm also going to
> recommend that SW park the ADC on ground when not in use, and that they
> look hard at the settling time again.
>
> All gremlins must die.
>
>
> -----Original Message-----
> From: piclist-bounces@mit.edu [mailto:piclist-bounces@mit.edu] On Behalf
> Of alan.b.pearce@stfc.ac.uk
> Sent: Thursday, May 5, 2016 8:40 AM
> To: piclist@mit.edu
> Subject: RE: [PIC] Glitches on analog input
>
> IIRC one of the techniques to use is to have a capacitor on the ADC input
> pin that is large enough to swamp the ADC input capacitance, but with a l=
ow
> enough source impedance to be an LP at the highest frequency of interest,
> without going beyond the source impedance restriction of the ADC.
>
>
>
> > -----Original Message-----
> > From: piclist-bounces@mit.edu [mailto:piclist-bounces@mit.edu] On
> > Behalf Of Van Horn, David
> > Sent: 05 May 2016 14:32
> > To: Microcontroller discussion list - Public.
> > Subject: RE: [PIC] Glitches on analog input
> >
> > Well, I was sort of hoping that part of the "simplification" would be
> > glossing over that buffer.
> >
> > I'll have a look at it.  Parking the ADC on the ground input would
> > probably be a good technique to make whatever effect this has on the
> > reading more consistent.
> >
> > Thanks.
> >
> > -----Original Message-----
> > From: piclist-bounces@mit.edu [mailto:piclist-bounces@mit.edu] On
> > Behalf Of alan.b.pearce@stfc.ac.uk
> > Sent: Thursday, May 5, 2016 2:44 AM
> > To: piclist@mit.edu
> > Subject: RE: [PIC] Glitches on analog input
> >
> > > > A possible cause is that you are seeing the charge from the ADC's
> > > > internal sampling capacitor. Another possible cause is leakage
> > > > caused by an analog input clamping a voltage above the supply rail.
> > >
> > > I would hope that wasn't the case.  Shouldn't there be at least a
> > > buffer amplifier between the capacitor and the switch?
> >
> > Nope, not when you look at the simplified schematics they put in the
> > data sheet and family manuals. That is one reason why there is a
> > significant restriction on signal source impedance to maintain the ADC
> accuracy.
> >
> > --
> > http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
> > View/change your membership options at
> > http://mailman.mit.edu/mailman/listinfo/piclist
> >
> > --
> > http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
> > View/change your membership options at
> > http://mailman.mit.edu/mailman/listinfo/piclist
>
> --
> http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
> View/change your membership options at
> http://mailman.mit.edu/mailman/listinfo/piclist
>
> --
> http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
> View/change your membership options at
> http://mailman.mit.edu/mailman/listinfo/piclist
>
--=20
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist
.