Yet Another Gaussmeter

[salvarsan]

A little joke

This circuit sets the AD22151 sensitivity to 1.3mv/Gauss,
same as the Allegro A1302.

[RedHouse]

after last posts my questions make more sense.... without oversampling/average you put to work a PIC18F to obtaining results very worse than the simple sensor+voltmeter solution

Nice Elepro, somehow I would expect no less from you

If I really cared, I might ask what makes you think "put to work a PIC18F to obtaining results very worse than the simple sensor+voltmeter" but, I don't, so I won't ask. Seems from your posts of late it's appearing that it's bothering you that others can do what you've done, or are you just being mis-understood.

To be civil I'll respond by saying my 18F is running the code from the 16F, but as I said in this post above, the switch it to do with future projects that will need the additional ADC's.

MicroChip has a similar oversampling doc, AN1152:
http://ww1.microchip.com/downloads/e...g%2001152A.pdf
and includes source code, albeit for their dsp processors:
http://ww1.microchip.com/downloads/e...rce%20Code.zip

Thanks for the links again, I'd actually downloaded that app note before but it's always nice when someone goes the extra mile and posts links.

I am currently using a different approach although it's not set-in-stone and I may change it to better utilize the AD22151 which I've just started working with really. I've been using the A1302 and OHS3150U then just recently started applying the AD22151 which has been sitting on my bench since the group-buy back in August.

Currently I'm applying a sample/sort/average routine like a rolling-loop array sort to filter away the odd readings instead of using them to extrapolate missing data. Like a bell curve algorithm with 256 sample/sort/avg per displayed reading.

In my first prototype a minumum resoution of +/- 3.7G (using the A1302) isn't a big issue to me but now working with the AD22151 seems I may need to change my approach, we'll see.

[salvarsan]

... I might ask what makes you think "put to work a PIC18F to obtaining results very worse than the simple sensor+voltmeter" but, I don't, so I won't ask. Seems from your posts of late it's appearing that it's bothering you that others can do what you've done, or are you just being mis-understood.

Oh, he's okay.
He means that he first thought we did no data smoothing
but now he understands that we _do_ oversample and average.

No problems, Elepro.

In my first prototype a minumum resoution of +/- 3.7G (using the A1302) isn't a big issue to me but now working with the AD22151 seems I may need to change my approach, we'll see.

As I reconsider the A1302's +/-3% worst case inaccuracy,
Its reproducibility is much better, perhaps good enough
to merit 12 or 13 bits at the ADC.


Compared to the Analog Devices AD22151, its shortcomings
are sensitivity to power supply and temperature.


The through-hole A1302 sensitivity is 1.34mv/Gauss
compared to 1.30 for the SMT package. Given its
inherent inaccuracy, scaling by 1.33 or 4/3 is
accurate enough in either case.

Most important, Newark Electronics stocks the A1302
but the AD22151 is back-ordered everywhere.

-drh

[RedHouse]

...

Most important, Newark Electronics stocks the A1302
but the AD22151 is back-ordered everywhere.

-drh

A big 10-4 to that! Allied also has them in stock (at least last time I checked)

[Jedd]

...because we can.


Seldom does good+fast+cheap come in one package as it does with oversampling.


-drh

I saw a poster once that said we can do two out of three... good, fast, cheap. You pick which two.