[sdiy] while we're talking about scaling & biasing etc

[Jason Proctor]

thanks for the responses, all.

>He didn't say that exactly. He actually said he'd checked the 
>scale-and-offset hardware by testing it with a pot from +5V, and 
>that the hardware was fine.

ah, sorry if my phrasing was confusing.

tests done include -

- check that the scale/offset hardware is working by feeding it +/-5v 
and confirming that 0-5v appears on the correct pin.

- check that the ADC is working by detaching the scale/offset 
hardware from the pin, attaching a wire to a 0-5v pot instead, and 
confirming that the software reads values commensurate with the 
position of the pot.

>You're dead right though, Ingo: the important test to do is to check 
>the ADC with the simplest-possible hardware, usually a pot between 
>+5V and 0V.

indeed, that was test 2 :-)

>If the ADC works like this, then it's probably something to do with 
>the extra hardware.

indeed. it must be something to do with the scale/offset stuff. but 
since test 1 succeeded, we're looking at something a bit untoward. 
which is why i'm thinking it has something to do with the separate 
regulators.

when i first started building this module, i had a similar problem 
when the Ard was powered off USB but the pots etc were connected to 
regulator 1. i measured the correct voltages but the Ard couldn't see 
them. once i powered the Ard off regulator 2, everything worked fine. 
but the scale/offset stuff is powered off regulator 2, hence my 
theory.

i've just thought of something else to check -- i don't remember 
whether the Ard has an analogue reference ground or similar. i will 
take a look today.

thanks,
j