> My questions: I want the ADC to be 12-bit.

There are some model rocketry folks on this list that I'm sure will pipe up,
but I don't understand why you need 2.5 foot accuracy for your altitude.  Is
the sensor even that accurate?  Can't atmospheric effects cause at least
this kind of noise when using pressure to determine altitude, even assuming
calibration right before launch?  And what about Bernulli effects?

Again, I have no experience with model rocketry, but my knee jerk reaction
is to look at a 16F876.  This part has built in 10 bit A/Ds, which will give
you 10 foot altitude resolution over your 0 to 10,000 foot range.  It also
has 256 bytes of EEPROM, which could store about 200 10-bit altitude values.
I don't know how long a modle rocket flight lasts, so I don't know whether
200 data points is reasonable.  I could imagine sampling more often on the
way up than on the way down, since the rocket is (presumably) going much
slower down.

The 16F876 has a built in UART so that the altitude data can be downloaded
easily after the flight.  The 16F series also come with some features to aid
low cost debuggers, which can be useful if you're just getting into PICs.
And all this comes integrated in a 28 pin package.

> An important part- when I get the converted output into
> the PIC, I still need to do some calculations. I figured out a good
equation
> that is accurate +/- 2.5ft from 0-10000ft, but uses long coefficients (the
> equation is at the bottom). Can the PIC handle this well, and can it do it
8
> to 16 times per second

Your equation is a simple third order polynomial.  If you are measuring
pressure to 10 or 12 bits, there is no need for 15 digits of precision in
the coeficients.  24 bit floating point with 16 mantissa bits should be more
than adequate.  Yes, the PIC can handle this sort of thing well under your
62mS limit.  In one project I have a 16F876 at 20MHz doing a lot more PID
controller calculations than that every 10mS with at least 3mS left over for
other stuff.

> or should I have it simply save the raw ADC value to
> be computed on the ground?

I think that depends on which representation is more compact and if you need
real time altitude for any other reason.

> Finally, does the general idea sound doable?

It does to me.


*****************************************************************
Olin Lathrop, embedded systems consultant in Devens Massachusetts
(978) 772-3129, olin@embedinc.com, http://www.embedinc.com

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
email listserv@mitvma.mit.edu with SET PICList DIGEST in the body