Hi Kiste,

2010/10/28 Oliver Seitz <[email protected]>

> >When powered @5V, I loose a lot of resolution, thus information. Even more
> when such a voltage divider would be designed for power supply up to 24V...
>
> That's right, you're loosing resolution. 5V max. on a divider designed for
> 24V, you've got a resolution of only about 24 milivolts. So, when measuring
> only 5V, accuracy can not be better than 0.5%.
>

OK, right. I'm starting to accept the idea of just not doing this :)


>
> Have you put this number into relation to the rest of your circuit? For the
> divider, are you using expensive precision resistors with 0.1% accuracy and
> do you compensate for temperature errors? Or do you use cheap resistors with
> 5%?


Something in between. I'm using 1% accurary resistors:
http://sirbot.org/sirbot-modules/dc-motor-controller-board/dcmc_rctank_schematic.png


> Is there a dedicated 5V reference, or do you simply use one supply to power
> different devices? Then your reference voltage could introduce errors of
> several percents.
>

Vref is the one used by the PIC itself. So yes, it may be shared with other
devices (but not the motors')


>
> The magic is quite simple: You're powering the Op-amp with 5V. As there is
> no magic inside the Op-amp, it cannot output more than 5V. You should use
> so-called rail-to-rail (R2R) op-amps, simple Op-amps will not even output
> 4V.
>

OK, I thought R2R was actually R/2R (ladder I think). I was quite lost :)


> >Have I absolutely misunderstood this whole thing ??
>
> In general, you understood very well :)
>

Well, that's a great emotional victory !


I was thinking that... maybe... Could I use internal vref output, as Vasile
mentioned, connected to Vref+ input, and adjust Vref output so I could match
a better resolution ? Say I have 5V as input, on a 24V divider. More or less
I would measure between 0-1V. If I lower Vref+ to 1V instead of 5V, with
internal vref output, I would now a better resolution, since 0-1V range
would now expand to full resolution.

If needed...


Cheers,
Seb

-- 
You received this message because you are subscribed to the Google Groups 
"jallib" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/jallib?hl=en.

Reply via email to