Ultrasonic remote control switch problems.

Hello,

My intention is to build a reliable ultrasonic remote control switch
powered by 9V batteries with a desired range of 50M.

I've had a poor time of it so far.

I started with one of the few circuits seemingly available on the net.
http://www.electronic-circuits-diagrams.com/sensorsimages/7.gif
http://www.electronic-circuits-diagrams.com/sensorsimages/sensorsckt7.shtml

Replacing the transmitters unavailable SL100 and SK100 transistors
with BD139 and BD140 transistors I achieved a range of 5cm. Asking for
some advice in this group I changed them to BC548/558 added 220uF caps
across the battery supply rails and tried to reverse the transistors
but couldn't get that to work. I also added 20% 0.1 and 0.001 ceramic
caps across the transmitters supply rails, 20% IC decoupling caps, 1%
resistors and used 10% caps for the other caps along with shielded
cables and managed to achieve a 1.5M range. The 1.5M range was with
the transducer driven directly from the 555 and not by the transistors
which only managed less then half that range.

Finding the seemingly only available kitset (24kHz) on the net I see
that it uses few components with an advertised 50ft range. The kitset
uses a 16pin Dip, three resistors, a cap, trimmer, LED and push button
switch for the transmitter.

Can anyone help me build an ultrasonic tx/rx switch like it or better?

One particular electronic kitset site that sells the 24kHz kit also
sells the LM388 14Pin (now obsolete) as an ultrasonic driver.

LM388
http://sales.goldmine-elec.com/prodinfo.asp?prodid=8501

Kit
http://sales.goldmine-elec.com/prodinfo.asp?prodid=2103

Better photo of the kitset
http://www.gibsonteched.net/ctc68160.html

This led me to search National Semi for ultrasonics and I found the
LM386 to replace the LM388
http://www.national.com/pf/LM/LM388.html
http://www.national.com/pf/LM/LM386.html

At first I thought a sine wave was required and then found that a
square wave is fine and in the LM386 PDF datasheet, page 5, there is a
square wave circuit but for only 1kHz.
I don't know how to convert this circuit to 40kHz for the combined
transtmitter/receiver transducers I have or how the transducer is
connected. I think the circuits Vo should go through a low value
resistor and capacitor before the transducer to V- but I don't realy
know.

Any help is greatly appreciated.

Andrew.

 

[Robert C Monsen]

<Rubicon> wrote in message
news:4101e2af.5174101@news.netaccess.co.nz...

Hello,

My intention is to build a reliable ultrasonic remote control switch
powered by 9V batteries with a desired range of 50M.

I've had a poor time of it so far.

I started with one of the few circuits seemingly available on the
net.
http://www.electronic-circuits-diagrams.com/sensorsimages/7.gif

http://www.electronic-circuits-diagrams.com/sensorsimages/sensorsckt7.shtml

Replacing the transmitters unavailable SL100 and SK100 transistors
with BD139 and BD140 transistors I achieved a range of 5cm. Asking
for
some advice in this group I changed them to BC548/558 added 220uF
caps
across the battery supply rails and tried to reverse the transistors
but couldn't get that to work. I also added 20% 0.1 and 0.001
ceramic
caps across the transmitters supply rails, 20% IC decoupling caps,
1%
resistors and used 10% caps for the other caps along with shielded
cables and managed to achieve a 1.5M range. The 1.5M range was with
the transducer driven directly from the 555 and not by the
transistors
which only managed less then half that range.

Finding the seemingly only available kitset (24kHz) on the net I see
that it uses few components with an advertised 50ft range. The
kitset
uses a 16pin Dip, three resistors, a cap, trimmer, LED and push
button
switch for the transmitter.

Can anyone help me build an ultrasonic tx/rx switch like it or
better?

One particular electronic kitset site that sells the 24kHz kit also
sells the LM388 14Pin (now obsolete) as an ultrasonic driver.

LM388
http://sales.goldmine-elec.com/prodinfo.asp?prodid=8501

Kit
http://sales.goldmine-elec.com/prodinfo.asp?prodid=2103

Better photo of the kitset
http://www.gibsonteched.net/ctc68160.html

This led me to search National Semi for ultrasonics and I found the
LM386 to replace the LM388
http://www.national.com/pf/LM/LM388.html
http://www.national.com/pf/LM/LM386.html

At first I thought a sine wave was required and then found that a
square wave is fine and in the LM386 PDF datasheet, page 5, there is
a
square wave circuit but for only 1kHz.
I don't know how to convert this circuit to 40kHz for the combined
transtmitter/receiver transducers I have or how the transducer is
connected. I think the circuits Vo should go through a low value
resistor and capacitor before the transducer to V- but I don't realy
know.

Any help is greatly appreciated.

Andrew.

One problem is that the power of your sender is limited by the voltage
you put across it. 50m is a pretty far distance. Consider how much
energy you would use to yell intelligably to somebody 50m away...

One way I've gotten more energy across the transducer is to use an
inductive spike. You have an N-MOSFET attached to the gate of your
40kHz oscillator. The source is grounded, and the drain is attached to
one side of an inductor, possibly 1000uH or so. Your transducer (which
is, in effect, a capacitor) is in parallel with the inductor. In order
to limit current through the transducer, you'll need a resistor of
about 100 ohm in series with it.

Assume you are generating a 40kHz square wave. Then your mosfet is on
for 12.5us. That means that at the end of 12.5us (with a 9V supply)
your current will be about 112mA. Thus, the energy across the inductor
on each cycle is now

1000uH * (112mA)^2/2 = about 6.3uJ. The power is thus

6.3uH / 25us = about 1/4 W

You might be able to hear that with your receiver circuit at 50m.

Make sure the power mosfet can handle 100V, since the spikes will be
something like 80V.

Note that some transducers are voltage limited; if so, you'll need to
increase the inductor, or decrease the duty-cycle of the oscillator.

Regards,
Bob Monsen

 

Thankyou for the information.

Reagrds,
Andrew.


On Fri, 23 Jul 2004 21:33:45 GMT, "Robert C Monsen"
<rcsurname@comcast.net> wrote:

Rubicon> wrote in message
news:4101e2af.5174101@news.netaccess.co.nz...


Hello,

My intention is to build a reliable ultrasonic remote control switch
powered by 9V batteries with a desired range of 50M.

I've had a poor time of it so far.

I started with one of the few circuits seemingly available on the
net.
http://www.electronic-circuits-diagrams.com/sensorsimages/7.gif

http://www.electronic-circuits-diagrams.com/sensorsimages/sensorsckt7.shtml

Replacing the transmitters unavailable SL100 and SK100 transistors
with BD139 and BD140 transistors I achieved a range of 5cm. Asking
for
some advice in this group I changed them to BC548/558 added 220uF
caps
across the battery supply rails and tried to reverse the transistors
but couldn't get that to work. I also added 20% 0.1 and 0.001
ceramic
caps across the transmitters supply rails, 20% IC decoupling caps,
1%
resistors and used 10% caps for the other caps along with shielded
cables and managed to achieve a 1.5M range. The 1.5M range was with
the transducer driven directly from the 555 and not by the
transistors
which only managed less then half that range.

Finding the seemingly only available kitset (24kHz) on the net I see
that it uses few components with an advertised 50ft range. The
kitset
uses a 16pin Dip, three resistors, a cap, trimmer, LED and push
button
switch for the transmitter.

Can anyone help me build an ultrasonic tx/rx switch like it or
better?

One particular electronic kitset site that sells the 24kHz kit also
sells the LM388 14Pin (now obsolete) as an ultrasonic driver.

LM388
http://sales.goldmine-elec.com/prodinfo.asp?prodid=8501

Kit
http://sales.goldmine-elec.com/prodinfo.asp?prodid=2103

Better photo of the kitset
http://www.gibsonteched.net/ctc68160.html

This led me to search National Semi for ultrasonics and I found the
LM386 to replace the LM388
http://www.national.com/pf/LM/LM388.html
http://www.national.com/pf/LM/LM386.html

At first I thought a sine wave was required and then found that a
square wave is fine and in the LM386 PDF datasheet, page 5, there is
a
square wave circuit but for only 1kHz.
I don't know how to convert this circuit to 40kHz for the combined
transtmitter/receiver transducers I have or how the transducer is
connected. I think the circuits Vo should go through a low value
resistor and capacitor before the transducer to V- but I don't realy
know.

Any help is greatly appreciated.

Andrew.






One problem is that the power of your sender is limited by the voltage
you put across it. 50m is a pretty far distance. Consider how much
energy you would use to yell intelligably to somebody 50m away...

One way I've gotten more energy across the transducer is to use an
inductive spike. You have an N-MOSFET attached to the gate of your
40kHz oscillator. The source is grounded, and the drain is attached to
one side of an inductor, possibly 1000uH or so. Your transducer (which
is, in effect, a capacitor) is in parallel with the inductor. In order
to limit current through the transducer, you'll need a resistor of
about 100 ohm in series with it.

Assume you are generating a 40kHz square wave. Then your mosfet is on
for 12.5us. That means that at the end of 12.5us (with a 9V supply)
your current will be about 112mA. Thus, the energy across the inductor
on each cycle is now

1000uH * (112mA)^2/2 = about 6.3uJ. The power is thus

6.3uH / 25us = about 1/4 W

You might be able to hear that with your receiver circuit at 50m.

Make sure the power mosfet can handle 100V, since the spikes will be
something like 80V.

Note that some transducers are voltage limited; if so, you'll need to
increase the inductor, or decrease the duty-cycle of the oscillator.

Regards,
Bob Monsen