Maker Pro
Maker Pro

A different kind of motion sensing security light?

J

John Doe

Correct me if I'm wrong on any of this, but... The motion sensor
(part) output constantly varies. Only when its output reaches a
certain level relative to its normal state, then it trips a switch
that drives the light for some period of time. Then the lightbulb
switches off and waits for another significant change in
detection.

What I would like to do is have a lightbulb or LED
brightness/intensity follow the motion sensor output part. In
other words... If a bird flies by, the lightbulb would momentarily
dimly light. If a person walks into close range, the lightbulb
would continuously shine brightly. If the person walked behind
some object, the lightbulb would go off, and then when they
reappeared, the lightbulb would come back on. Hopefully I'm not
over describing this, but I'll be happy to provide more
description if asked.

Does such a device (or close) already exist for purchase in a
store?

The sensor output part typically could not be connected to an LED
without putting an amplifier on it? Anybody do this before, any
schematic?

Thanks.
 
J

Jamie

John said:
Correct me if I'm wrong on any of this, but... The motion sensor
(part) output constantly varies. Only when its output reaches a
certain level relative to its normal state, then it trips a switch
that drives the light for some period of time. Then the lightbulb
switches off and waits for another significant change in
detection.

What I would like to do is have a lightbulb or LED
brightness/intensity follow the motion sensor output part. In
other words... If a bird flies by, the lightbulb would momentarily
dimly light. If a person walks into close range, the lightbulb
would continuously shine brightly. If the person walked behind
some object, the lightbulb would go off, and then when they
reappeared, the lightbulb would come back on. Hopefully I'm not
over describing this, but I'll be happy to provide more
description if asked.

Does such a device (or close) already exist for purchase in a
store?

The sensor output part typically could not be connected to an LED
without putting an amplifier on it? Anybody do this before, any
schematic?

Thanks.

look at ultrasonic distance sensors. You can get outputs that
are analog for a distance reading of an object infront of it or
a switched output when an object is with in range.

http://store.mp3car.com/ProductDeta...4f4d2d303738&gclid=CJ2d-vabmbUCFUid4AodwxIAIg


Jamie
 
T

Tim Williams

AFAIK, pyroelectric sensors are a "single pixel" sort of thing, so they
don't know if the scene is moving, only if the average IR changes. Which it
does when someone comes into view or jumps out of the shadows, but it
doesn't do jack if someone walks around freely while holding up a
room-temperature blanket, or moves in the scene while presenting an equal
viewing cross section (probably tricky to do).

The output is positive or negative charge, corresponding to the change in
temperature viewed. Into a high resistive load (say, like something Phil
might be working on right now :) ), this has a time constant (the sensor's
just an insulating crystal), so it automatically centers after a while. In
other words, it's a temperature differentiator (for pedants, that's the RC
kind). Obviously, a motion detector needs only a window comparator; you
could use a suitably calibrated gain instead (followed by a full-wave active
rectifier), and maybe a lowpass, or falling slew rate limiter, to enhance
the effect.

Still might not be quite as intended; if the detector response is equal
across the viewing angle, then a person walking across the path will cause
it to light up (or down), then not do so much (depending on how the sensor
sees a walking body), then light up again on leaving the frame.

Tim
 
J

John Doe

Tim Williams said:
AFAIK, pyroelectric sensors are a "single pixel" sort of thing,
so they don't know if the scene is moving, only if the average
IR changes.

That's all I want. It's not going to be the only part of the
system, and I can interpret the output. I guess it could even be a
very short (1 second or less) on time by a common passive infrared
motion detector, so that it would flash during its detection time.
 
H

harry

Correct me if I'm wrong on any of this, but... The motion sensor
(part) output constantly varies. Only when its output reaches a
certain level relative to its normal state, then it trips a switch
that drives the light for some period of time. Then the lightbulb
switches off and waits for another significant change in
detection.

What I would like to do is have a lightbulb or LED
brightness/intensity follow the motion sensor output part. In
other words... If a bird flies by, the lightbulb would momentarily
dimly light. If a person walks into close range, the lightbulb
would continuously shine brightly. If the person walked behind
some object, the lightbulb would go off, and then when they
reappeared, the lightbulb would come back on. Hopefully I'm not
over describing this, but I'll be happy to provide more
description if asked.

Does such a device (or close) already exist for purchase in a
store?

The sensor output part typically could not be connected to an LED
without putting an amplifier on it? Anybody do this before, any
schematic?

Thanks.

You can buy sensors that have a restricted/adjustable "target" size.
To eliminate nuisance operations from small animals.

There will always be an amplifier and relay in the circuit.
 
I don't know about this particular sensor, but most of the commercial
PIR elements for porch lights have the FET built in.  Maybe the 47k is
the source resistor of the JFET follower?

Cheers

Phil Hobbs


I doubt he's going to find a finished product that does what
he wants. The products ready to be mounted have not
only the sensor, but as he points, out the circuitry that
turns it into some kind of on/off output that is needed for
the product.

However those motion detector products use ICs which
he can probably find. Some of the ICs provide an analog
output that he could use together with his own circuit.
Here are examples from one company:

http://www.mouser.com/ds/2/281/s21e-6137.pdf
 
You can buy sensors that have a restricted/adjustable  "target" size.
To eliminate nuisance operations from small animals.

There will always be an amplifier and relay in the circuit.- Hide quoted text -

- Show quoted text -

Wrong yet again. harry is like a cosmic black hole of ignorance, no
real knowledge can exist anywhere near him.
 
M

Mark Zenier

AFAIK, pyroelectric sensors are a "single pixel" sort of thing, so they
don't know if the scene is moving, only if the average IR changes. Which it
does when someone comes into view or jumps out of the shadows, but it
doesn't do jack if someone walks around freely while holding up a
room-temperature blanket, or moves in the scene while presenting an equal
viewing cross section (probably tricky to do).

They're two "pixels" in an AC coupled differential circuit.
The optics are set so that the sensors see adjacent areas and
react to fast changes. (You can fool them if you move real slow).

The best analogy of the optics is to spread your fingers apart and put the
flat of your palms together so that the fingers interleave. One sensor
per "hand", with the views of the two sensors interlinked. Any movement
quick enough, and one sensor will decrease while the other increases.
The response is usually flat in one plane, determined by the way the
plastic lens is molded.

Radio-Electronics/Electronic Now magazine had a long distance pyroelectric
sensor project that used a motor driven mechanical chopper, back when
these things were new.

Mark Zenier [email protected]
Googleproofaddress(account:mzenier provider:eskimo domain:com)
 
J

Jamie

Tim said:
AFAIK, pyroelectric sensors are a "single pixel" sort of thing, so they
don't know if the scene is moving, only if the average IR changes.
Which it does when someone comes into view or jumps out of the shadows,
but it doesn't do jack if someone walks around freely while holding up a
room-temperature blanket, or moves in the scene while presenting an
equal viewing cross section (probably tricky to do).

The output is positive or negative charge, corresponding to the change
in temperature viewed. Into a high resistive load (say, like something
Phil might be working on right now :) ), this has a time constant (the
sensor's just an insulating crystal), so it automatically centers after
a while. In other words, it's a temperature differentiator (for
pedants, that's the RC kind). Obviously, a motion detector needs only a
window comparator; you could use a suitably calibrated gain instead
(followed by a full-wave active rectifier), and maybe a lowpass, or
falling slew rate limiter, to enhance the effect.

Still might not be quite as intended; if the detector response is equal
across the viewing angle, then a person walking across the path will
cause it to light up (or down), then not do so much (depending on how
the sensor sees a walking body), then light up again on leaving the frame.

Tim

we used a few of those for wire break and jerk detectors just as the
conductor comes out of the induction heater and prior to entering the
extrusion head assembly.

Since the conductor will jerk with a over sized section going through
or a breakage, it works out well. There is also a small air jet that
sits infront of the IR passive glass to keep the dust off.

Jamie
 
S

SoothSayer

I don't know about this particular sensor, but most of the commercial
PIR elements for porch lights have the FET built in. Maybe the 47k is
the source resistor of the JFET follower?

Cheers

Phil Hobbs

One must also be *VERY* careful when handling/installing the device as
well. They usually come with shorting wires on the leads they are so
worried.

Some FETs can be blown by an ESD field, particularly prior to being
installed. Does not even require contact.

That is why a smock is required at the 'proper' ESD safe workstation.
It 'contains' whatever fields you and your insulative clothing may be
carrying.
 
J

John Doe

Generally, no.

The common pyroelectric motion sensor outputs a single pulse
when the incident radiation which falls on the detector has a
higher or lower wavelength than that of the ambient radiation
within the field of view of the transducer.

I guess that means there is no analog output. So it's not like a
photo resistor. I've played with photo resistors, controlling a
variable frequency connected to a speaker. Aiming it towards a TV
produced amusing and rapidly changing frequencies. I wonder if
there are hypersensitive photo resistors that might do (with a
wide field of view), maybe infrared or whatever. Or maybe that has
already been suggested under different terminology.
Someone already mentioned a TOF SONAR sensor, which would be
ideal if it could resolve distance to the accuracy you require,
which will vary depending on temperature, humidity, and
atmospheric pressure.

Can you address that?

I don't need to know distance or speed. All I want to sense is
changes in radiation in the area. But if the changes in intensity
are great, that might suggest the object is either large or
nearby. I guess that would be a function of something like a
long-range omnidirectional (or wide angle, as long as it's view is
not narrow) infrared photo resistor/sensor if there is such a
thing. I'll look.

Distance to sense a person, to noticeably and distinctly change
the output, needs to be at least 20 feet, preferably 50-100.
 
M

mike

I guess that means there is no analog output. So it's not like a
photo resistor. I've played with photo resistors, controlling a
variable frequency connected to a speaker. Aiming it towards a TV
produced amusing and rapidly changing frequencies. I wonder if
there are hypersensitive photo resistors that might do (with a
wide field of view), maybe infrared or whatever. Or maybe that has
already been suggested under different terminology.


I don't need to know distance or speed. All I want to sense is
changes in radiation in the area. But if the changes in intensity
are great, that might suggest the object is either large or
nearby. I guess that would be a function of something like a
long-range omnidirectional (or wide angle, as long as it's view is
not narrow) infrared photo resistor/sensor if there is such a
thing. I'll look.

Distance to sense a person, to noticeably and distinctly change
the output, needs to be at least 20 feet, preferably 50-100.
https://www.google.com/search?num=1...34.6-1.1.0.les;..0.0...1c..2.serp.z6TfNc9RhwA
 
J

John Doe

To be clear...
actually, what I played with might be called a photodiode (not a
photoresistor), it looked like an ordinary white LED, it didn't have
the squiggly wires inside
 
J

John Doe

John Fields said:
The common pyroelectric motion sensor outputs a single pulse
when the incident radiation which falls on the detector has a
higher or lower wavelength than that of the ambient radiation
within the field of view of the transducer.

FWIW...
I will plan to use that, that will be easiest. I will see if the
output can be made to flash a light as long as motion is being
detected, without a long delay or long on-time.
 
J

Jasen Betts

What I would like to do is have a lightbulb or LED
brightness/intensity follow the motion sensor output part. In
other words... If a bird flies by, the lightbulb would momentarily
dimly light. If a person walks into close range, the lightbulb
would continuously shine brightly. If the person walked behind
some object, the lightbulb would go off, and then when they
reappeared, the lightbulb would come back on. Hopefully I'm not
over describing this, but I'll be happy to provide more
description if asked.

these sensors detect movement by having two "pixels" that are focused by
the lens onto alternating bands of the scene, the sensors only produce
a signal proportional to the rate of change of average temperature of
the area viewed by the pixel.

a person walking across the fiels od view would produce a signal that
alternates between the two sensors.
Does such a device (or close) already exist for purchase in a
store?

I've not heard of one.
 
R

Robert Macy

They're two "pixels" in an AC coupled  differential circuit.
The optics are set so that the sensors see adjacent areas and
react to fast changes.  (You can fool them if you move real slow).

The best analogy of the optics is to spread your fingers apart and put the
flat of your palms together so that the fingers interleave.  One sensor
per "hand", with the views of the two sensors interlinked.  Any movement
quick enough, and one sensor will decrease while the other increases.
The response is usually flat in one plane, determined by the way the
plastic lens is molded.

Radio-Electronics/Electronic Now magazine had a long distance pyroelectric
sensor project that used a motor driven mechanical chopper, back when
these things were new.

Mark Zenier  [email protected]
Googleproofaddress(account:mzenier provider:eskimo domain:com)

If you're going mechanical, why not make it a capacitive sensor.
Chopping the field from the intruder. Since you kknow the speed of the
chopping, you can synchronously detect and obtain EXACTLY the type of
signal you want. ...I think.
 
J

John Doe

John Fields said:
John Doe said:
To be clear... actually, what I played with might be called a
photodiode (not a photoresistor), it looked like an ordinary
white [clear] LED, it didn't have the squiggly wires inside
Then it's more than likely either a photodiode or a
phototransistor.

Can you post a schematic of the circuit you played with?

Google produces 280 results for my "Tonal Voltmeter". I'm sure it
was just that, using a photodiode to vary resistance/voltage. It
looked like a clear T1 LED. I have nothing left (except fond
memories) from experimenting with electronics.
 
J

John Doe

John Fields said:
John Doe said:
John Fields said:
John Doe <jdoe usenetlove.invalid> wrote:

To be clear... actually, what I played with might be called a
photodiode (not a photoresistor), it looked like an ordinary
white [clear] LED, it didn't have the squiggly wires inside
Then it's more than likely either a photodiode or a
phototransistor.

Can you post a schematic of the circuit you played with?

Google produces 280 results for my "Tonal Voltmeter". I'm sure
it was just that, using a photodiode to vary resistance/voltage.
It looked like a clear T1 LED. I have nothing left (except fond
memories) from experimenting with electronics.

--- eternalseptember.org, huh?

Yes?

As I said in two prior separate replies... I am FWIW satisfied
with the discussion. I will probably use a cheap motion detector,
and maybe try to shorten the on-time so that the light blinks
while motion is being sensed.

--
 
J

Jamie

Ian said:
If you mean PIR motion detectors, they usually have twin element
pyrometers (thin film thermistor) a motion of a IR source (person) is
"grated" to produce a gross change from one element to the other - this
is done in various ways such as Freznel type lenses (special IR
transparent material) and faceted surface reflecting reflectors.

That's strange that you would call them thermistors? TO me a thermistor
is a slow reacting device? But yet we have an application where we use a
basic pyroelectric 3 wire detector for detecting vibration of small
conductor as it exits from an induction heater. This vibration can get
up to around 1khz or more when the wire snaps for what ever reason.

I would think a thermistor device would be a little slow for this
wouldn't you?

Maybe the units I used have different smoke in side.

Jamie
 
J

Jasen Betts

That's strange that you would call them thermistors?

AIUI they are closer to capacitor than resistor, the device does sense
its own temperature though, they detect remote temperature by
black-body radiation changing the temperature of the sensor.

TO me a thermistor
is a slow reacting device? But yet we have an application where we use a
basic pyroelectric 3 wire detector for detecting vibration of small
conductor as it exits from an induction heater. This vibration can get
up to around 1khz or more when the wire snaps for what ever reason.

I would think a thermistor device would be a little slow for this
wouldn't you?

if you could make it thin enough it would work.
 
Top