building a sine wave oscillator

[Fred Bloggs]

Hi John,

Well, sheepishly, I must say, I screwed up my algebra again. The
capacitance is looking like around 82pf at 24khz.

Joe

You made a bad measurement obviously since there is no such thing as a
resonant resistor. What's resonating is your scope probe. By placing it
at the transducer node you have introduced significant shunt capacitance
to the L+C of the probe, causing more peaking.

 

[MooseFET]

Yes, the 555 square wave may be good enough, but why not use the
triangle waveform at pin 6 where the capacitor is ramping up and
down? Wouldn't a ramping waveform have considerable less harmonics?

-Bill

They fall as 1/N^2 instead of 1/N so the higher ones are much lower.

With an oscillator and a CD4017 and some resistors, you can make the
low numbered harmonics low.

 

[John Popelish]

Hi John,

Well, sheepishly, I must say, I screwed up my algebra again. The
capacitance is looking like around 82pf at 24khz.

I think that if you take this reading anywhere near the
mechanical resonance of the piezo, you are not measuring the
capacitance at all, but the energy bouncing around in the
mechanical resonator, through a capacitive coupling. Try
taking this data at a much lower frequency, like a kHz or
less. I recommend you adjust the resistor till the piezo
voltage is .707 of the applied voltage (and strangely
enough, the voltage across the resistor should also be .707
of the applied voltage). Then measure the resistance. It
should be approximately equal to the magnitude of capacitive
impedance at that frequency. Then you can use the
frequency and impedance to find the value of the
capacitance. This is what you want to resonate with a
series inductor with square wave drive.

 

[John Larkin]

I think that if you take this reading anywhere near the
mechanical resonance of the piezo, you are not measuring the
capacitance at all, but the energy bouncing around in the
mechanical resonator, through a capacitive coupling. Try
taking this data at a much lower frequency, like a kHz or
less. I recommend you adjust the resistor till the piezo
voltage is .707 of the applied voltage (and strangely
enough, the voltage across the resistor should also be .707
of the applied voltage). Then measure the resistance. It
should be approximately equal to the magnitude of capacitive
impedance at that frequency. Then you can use the
frequency and impedance to find the value of the
capacitance. This is what you want to resonate with a
series inductor with square wave drive.

Does the piezo impedance near resonance look similar to the
low-frequency capacitance? For a quartz crystal, it sure doesn't.

I'd expect a piezo to behave sort of like a quartz crystal, and go
through some serious impedance gyrations around resonance, maybe
capacitive-resistive-inductive over a modest frequency span or
something.

John

 

[John Popelish]

Does the piezo impedance near resonance look similar to the
low-frequency capacitance? For a quartz crystal, it sure doesn't.

I'd expect a piezo to behave sort of like a quartz crystal, and go
through some serious impedance gyrations around resonance, maybe
capacitive-resistive-inductive over a modest frequency span or
something.

Exactly so.
http://www.americanpiezo.com/piezo_theory/resonance_frequency.html

 

[[email protected]]

Dear M...

I already have a hydrophone. That is what I am going to use to receive
the 24khz. It will receive or transmit up to 25khz, the mfr says to
stay away from 25khz if transmitting, but I can receive at any
frequency within its range with no problem. Been there, done that.

As far as research, you might want to do a little yourself, start
here:http://www.massa.com/fundamentals.htm

The what you call 'loss' is proportional to 1/r^2, just like em waves.
Because of the spherical or hemispherical nature of the wave
propagating thru the water. There are also absorption losses due to
the manganese dioxide present in sea water. At 20khz, this amounts to
-3dB in a kiloyard (I know, unfortunately some folks still use the
imperial system). Now the 1/r^2 I am speaking of is the acoustic
energy flux density (aka intensity), NOT the sound pressure level. The
sound pressure level falls off as proportional to 1/r. I would refer
you to a great book on this very subject:
"Principles of Underwater Sound" by Robert Urick <sp>, and also
"Transducers and Arrays for Underwater Sound", by Sherman and Butler.
You can take a quick peek at them at amazon.

I know a little something about the physics of sound propagation in
different media.

My original question was about building a sine wave oscillator, which
has been answered, and I wish to thank everyone who replied. I know
what I am up against as far as sending sound through the ocean. I am
trying to design with those principles in mind, hence, the reason I
wanted to maximize the output from my multiplier.

Electronics has been a hobby for me, so that is why I come here and
ask questions on occasion. Most of the time, I just lurk and learn.
Oh, and screw up algebra.

Thanks again to all who replied. I believe I can make my way from
here.

Joe

Joe:

You stated a piezo transducer. That could be many things. If you said
hydrophone, it would have saved some time.

I designed this kind of gear (sonobuoy) in the 80s for use with the
P-3 Orion and quite frankly have no interest in digging up loss factor
of the ocean for something I did two freaking decades ago. Since you
have a chip on your shoulder, I'll cut this short. The loss I refer to
is not that of a 1/r^2 radiator. Rather, on the receiving hydrophone,
you equalize for the loss of the ocean. The filters had a rising
response. This was for a passive system obviously.

 

[Joe]

Joe:

You stated a piezo transducer. That could be many things. If you said
hydrophone, it would have saved some time.

I designed this kind of gear  (sonobuoy) in the 80s for use with the
P-3 Orion and quite frankly have no interest in digging up loss factor
of the ocean for something I did two freaking decades ago. Since you
have a chip on your shoulder, I'll cut this short. The loss I refer to
is not that of a 1/r^2 radiator. Rather, on the receiving hydrophone,
you equalize for the loss of the ocean. The filters had a rising
response. This was for a passive system obviously.- Hide quoted text -

- Show quoted text -

Dear M,

My apologies if I came across as having a chip on my shoulder. I just
wasn't sure you understood what I was asking about.

The type of transducer I have been looking at can be seen at :

http://209.41.160.145/site/index.cfm?DSP=Page&ID=101

On the right hand side of the page there is the sx23 and the sx22.

I guess you could call it a hydrophone, but it's my way of thinking
(which is not necessarily correct) that when I hear hydrophone, that's
for listening.
piezo transducer can be for listening or transmitting. Those are just
my definitions and that is probly why I misunderstood your good
intentions.

Since you were in the biz, I would summarize it by saying that I think
of a piezo transmitter like a loudspeaker (or underwater sound
projector), and I think of a hydrophone as the
water equivalent of a microphone in air. So, I already have one that
listens, now I am going to build one that transmits, and that was why
I was inquiring about some ideas for a sinewave oscillator.I know most
piezos can receive or send. My bad.

Again, thank you for your reply.

Joe

 

[Joe]

That, of course, depends on what form of modulation is being
considered.

Well, I think I understand what you are saying John. Right now I am
using a function generator (sine wave) at audio frequencies to
modulate my carrier. But there is no piezo transmitter on it yet
either. I have built the colpitts oscillator on the breadboard from
the link to hyperphysics that Joseph gave me (Thanks Joseph, BTW, I
didn't think you were snotty). Amplitude modulation.

Now, though, I am thinking about, in the past when I have used the
good ole 555 to modulate IR LEDs for various applications. I would
actually use a 556, one half would oscillate at 38khz, and the other
half would oscillate at, say 10khz. Connect the output of the 10Khz
astable to the reset pin of the 38khz astable, and now I am not sure
if it is the reset pin, so Let's say use 2 555's first one is
oscillating at 10Khz, the second one is set to oscillate at 38Khz. I
take the output from the 10khz and tie it to pin 4 ( pin 4 is
connected to ground thru a 33k resistor ) of the 38Khz 555, and,
presto, I can use one of those IR decoders made by sony, or whomever
to get my 10khz at the receiving end. In fact, I think you were the
one who told me about this method on the seb forum a few years ago.
The reason I changed this over to 2 555's is because I have the pins
memorized, but I did use a 556 for that app.

I am pretty sure I am going to be transmitting digital data. I may be
wrong, but that's zero's and ones, and no crossing zero ?? So I guess
maybe the multiplier chip I am using may not be the right one after
all. Although it may be more versatile, ie, I should be able to
modulate it with either analog or digital. Right now, I am modulating
it with analog (linear amplitude modulation) audio across the whole
audio spectrum. With digital, I should probably only need a bandwidth
of 5khz or so. So if the data is strictly digital, then I should be
able to use the 555 (or 556) the same way and not even worry about a
sine wave?

er, something tells me we've come full circle here, because you told
me that in one of your first posts.

I guess maybe I should just try it and see what happens.

Joe