Hey guys and gals. FWIW, sonoluminescence is a real phenomenon.
I was suprised to find that somebody a couple offices over from
me is actually studying it. (But come to think of it, the guy
that got killed in that Cold Fusion experiment was over here
too...)

Sonoluminescence is the process by which water emits light when
it is bombarded with an energetic ultrasonic source. The exact
underlying physical mechanism is a mystery, though. I know of
no greater than unity processes associated with it (or with
anything for that matter).

----------------------

Antti,

I'm glad you spoke up to help Mr. Mail12707 (sorry for being,
presumptuous but he is mail). I think his desire to implement
vertical counters is the wrong solution. Instead, phase
accumulators are more appropriate.

To obtain an ultra stable 63.625kHz wave form, you have
two choices. First, you can hope that it is possible to write
an isochronous loop that executes in an integer number of
periods of the '84's crystal. Now considering the fact that
63625 = 5^3*509, I doubt very likely you'll find a crystal
satisfying this condition. The second choice is with a phase
locked loop. Take a look at National's CGS410 programmable
clock generator for example. This device implements a transfer
function:

 f_out = f_in * N / (R * P)

where
1 < N < 16384
0 < R < 1024
0 < P < 17
and the maximum frequencies are:
 f_in < 35MHz
 f_out < 135MHz

So if you had an isochronous loop that executed in say
20 instructions = 20 * 4*T_osc seconds

and you wish this to increment the 63.625kHz counter once
per loop, then

 1/63625 = 20 * 4 * T_osc
or
 T_osc = .196464 usec
 f_osc = 5.09 Mhz (remember 509 from above?)

Now if you had an ultra stable 10.0000000MHz ovenized crystal
as the frequency reference, then you could use the CGS410 and
have three degrees of freedom to choose the constants:

  5.09 = 10.0 * N/(R*P)
 0.509 N = R * P
For example, you could choose N=100, R=509, and P = 1.

This solves the ultra-stability problem. The other two
problems (100Hz to 5kHz complementary square output and
the 3 second timer) are trivially solved with the phase
accumulators. I can post the phase accumulator article
(again) or send it to those who request it.



Scott


PS. Do you realize that with an ionized solution of H+ and
OH- ions you simultaneously have an acid and a base?