Crystals For Odd Frequencies?

[Ron Hubbard]

Uh, gee folks, haven't you ever heard of the first tenet of
engineering: Keep It Simple, [Stupid]!

I'm building this to keep cost, materials, and labor time down
to a bare minimum. If I had wanted FFT, DDS, microprocessors,
EPROM's, programming, etc I just pay for a commercial EEG and
forget about it.

Keith R. Williams wrote in message ...

In article <3f6b09ed.6265615@news.itd.umich.edu>,
no_spam@aol.com
says...
On Thu, 18 Sep 2003 10:42:40 -0400, Keith R. Williams
krw@attglobal.net> wrote:



WHat happens if there is more than one wave? Measuring a
single sine
wave is easy. My solution would be to use a micro to do a
Forier
transform on the signal and then look at the peaks.

--
Keith

Yours is the best idea I've heard so far, and by far the
easiest
to actually use. It will require an ADC with response
down to very low frequencies, which may rule out a lot of
sound
cards. But there are some with bottom ends around 2 Hz or
so
(maybe lower, it's been a while since I looked), so this
should be
easy to get going.

There are a couple of caveats. You will need an FFT that
covers
the range of interest with the desired resolution. You can
do that
using really huge FFTs at the usual audio sample rates, or
you
can try to slow the sample rate with sample rate
conversion. Either
way you need a lot of time to get enough data for each FFT,
so
to prevent slow, jerky screen updates you would want to do
overlap processing. Just maintain a big data buffer and
run your FFT
and display refresh as fast as possible, always taking the
most
recent N samples from the buffer, even if some of them were
used
for the previous FFT. This gives a nice smooth response,
assuming
your FFT and display code are fast enough.

I don't think it's going to be a huge problem at these
frequencies. One should look up some algorithms for a
"continuous" FFT. I know of someone who does this stuff at
RADAR
frequencies (using FPGAs), so at a few Hz things should be
simple. One thing you're going to need is a good filter on
the
input to prevent aliasing, which will also help with noise.

The big advantage of the FFT approach is that you can have
an
effective huge array of narrow filters, and see all their
outputs at
once. So you not only know if you are at your target, but
also how
far off you are, and how strong each component is. And you
get
immediate information if the noise floor increases due to
electrodes
getting loose, etc. Nice!

--
Keith

 

[Ron Hubbard]

I wouldn't bet on it. If it does work there ain't gonna be any
schematics: the original circuit is in a book called IC
Projects Design; I forget the author's name.

I had built the circuit's isolation amplifier, integrator, and
other circuit components up to the filter stage, then built in
the digital filter from the 1164-8's spec sheets; the four
7490s I used (the local store was out of 74HC390s) to divide
the crystal oscillators down 2000 times to get the frequency I
want is basic counter stuff..

I don't have the time nor the patience to redraw the
schematics, nor for that matter a CAD package that outputs
..jpg or .gif images. And my scanner has apparently died, so
don't hold your breath waiting.

Ron



Robert Monsen wrote in message ...

Well, good luck to you, and I'll look forward to seeing the
circuit on
alt.binaries.schematics.electronics if you get it working to
your
satisfaction.

Regards,
Bob Monsen

 

[Keith R. Williams]

In article <bkg3sm$1do06$1@ID-43450.news.uni-berlin.de>,
hubbard_ron@hotmail.com says...

Uh, gee folks, haven't you ever heard of the first tenet of
engineering: Keep It Simple, [Stupid]!

It is! You wouldn't know simple if it bit you in the a$$.

I'm building this to keep cost, materials, and labor time down
to a bare minimum. If I had wanted FFT, DDS, microprocessors,
EPROM's, programming, etc I just pay for a commercial EEG and
forget about it.

Microprocessors are dirt cheap, you have dozens in your dwelling
now. The algorithms are cheaper yet (freely available to anyone
who cares enough to look). You're looking for digital filters,
which aren't that different than a A/D and Micro.

....or perhaps you really don't want help, rather support for your
looney ideas. You're the one who asked for suggestions. You got
'em. If you're going to get all pissy because the suggestions
don't meet your unstated criteria, I suggest that perhaps the
good people here won't bother to help you in the future.

Sheesh! What a maroon! ...top poster and all.

--
Keith

 

[The Captain]

Keith R. Williams <krw@attglobal.net> wrote in message news:<MPG.19d681a3d124eb3598a6cc@enews.newsguy.com>...

In article <bkg3sm$1do06$1@ID-43450.news.uni-berlin.de>,
hubbard_ron@hotmail.com says...
Uh, gee folks, haven't you ever heard of the first tenet of
engineering: Keep It Simple, [Stupid]!

It is! You wouldn't know simple if it bit you in the a$$.

I'm building this to keep cost, materials, and labor time down
to a bare minimum. If I had wanted FFT, DDS, microprocessors,
EPROM's, programming, etc I just pay for a commercial EEG and
forget about it.

Microprocessors are dirt cheap, you have dozens in your dwelling
now. The algorithms are cheaper yet (freely available to anyone
who cares enough to look). You're looking for digital filters,
which aren't that different than a A/D and Micro.

...or perhaps you really don't want help, rather support for your
looney ideas. You're the one who asked for suggestions. You got
'em. If you're going to get all pissy because the suggestions
don't meet your unstated criteria, I suggest that perhaps the
good people here won't bother to help you in the future.

Sheesh! What a maroon! ...top poster and all.

This sort of thing really doesn't help. If you can't tell someone
that you disagree in a polite manner, I suggest you don't post at all.
calling someone a moron in a post is rather childish.

John

 

[Steve]

If you can't tell someone that you disagree in a polite manner, I suggest you don't post at all. calling someone a moron in a post is rather childish.

Sheesh! This little tiff makes aus.electronics look like a 3rd Reich
Kindergarten!

I think that it's just not worth getting so upset about such an
inconsequential debate with a complete stranger. But people still do
it. I've even heard of apprehended violence orders being served as a
result of pointless bickering on usenet! Insanity!

nifty

 

[norm d.]

You might have better luch looking for xtals that are multiples of what you
need. You can just devide the multiple frequency down in various ways to get
what you want.

"Ron Hubbard" <hubbard_ron@hotmail.com> wrote in message
news:bjqqo6$m69lp$1@ID-43450.news.uni-berlin.de...

I need a stable clock oscillator for a digital filter. Does
anyone know if it's possible to make a crystal oscillator to
produce square waves at odd frequencies like 400 Hz, 700, Hz,
830 Hz, and 2kHz?

 

[Peter Meyer]

Hello,

"Ron Hubbard" <hubbard_ron@hotmail.com> wrote in message
news:bjqqo6$m69lp$1@ID-43450.news.uni-berlin.de...
I need a stable clock oscillator for a digital filter. Does
anyone know if it's possible to make a crystal oscillator to
produce square waves at odd frequencies like 400 Hz, 700, Hz,
830 Hz, and 2kHz?
You can always synthetize an odd frequency by using the PLL technique.

You need a crystal of (sub)multiple of the frequency you want, frequency
divider and a PLL chip.
Peter

 

[GPG]

Why not:

--LPF-AMP-LIMITER-PLL at 100xFin--output
Measure op with counter also listen to signal?