Rune said:
The sampling frequency is 12800 Hz 12bits and the signals I'm going to
measure on is a three phase system with voltage and current sensors. In
order not to loose any phase information of the signals I need to sample all
6 channels simultaneously.
I have looked at two chips LF398 and SMP04.
It's part of the final project before I can get my degree. At the university
they have a similar project where they use the SMP04.
The chips need to be cheap and in dip package.
Rune Christensen
Ok, that's good but you didn't answer my questions.
Look up the specs on the chips you've described and figure out how to
translate your system requirements into numbers that appear on spec
sheets. This is your senior project, not mine.
You seem to be falling into the same trap many of us have.
There is no such thing as simultaneously. You can't just measure
everything arbitrarily accurately and do the math.
Engineering is the art of figuring out how sloppy the system can be
and still get the job done.
I'm making the rash assumption that if it's three phase, it's a motor
or something similar. System noise is likely to swamp what you're
trying to measure. I'd worry a lot about that. 12-bits of resolution
on something that might have 50% noise spikes may be overkill.
BTW, if it's really a sine wave of known or measurable frequency, you
may be able to use one a/d and correct for the time skew in the
measurements. Is there some reason you'd expect one phase to be
radically different from the others? Think about how many things you
can NOT measure and still get the job done.
mike
--
Return address is VALID.
Wanted, 12.1" LCD for Gateway Solo 5300. Samsung LT121SU-121
Bunch of stuff For Sale and Wanted at the link below.
Compaq Aero floppy,ram,battery.
MINT HP-41CV, 2-METER AMPS, 200CH SCANNER
http://www.geocities.com/SiliconValley/Monitor/4710/
