Searching \ for '[EE] PIR tester' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/index.htm?key=pir+tester
Search entire site for: 'PIR tester'.

Exact match. Not showing close matches.
PICList Thread
'[EE] PIR tester'
2010\05\22@110147 by Charles Craft

picon face
I've had a couple hotel rooms recently that use PIR to enable/disable
the HVAC system.
If the sensors are in a bad spot or not sensitive they shut down the A/C
when I'm asleep.

As I understand PIR, you need movement across the field to trigger the
sensor.
Short of building a robot with an IR LED that moves around the room any
thoughts on
how to trigger the sensor with a static piece of hardware?

2010\05\22@111320 by jim

flavicon
face
Charles,

Use a PIC to generate PWM, connect it to a hobby servo motor, and
periodically move the arm on the servo that has a small flag or card
attached to it.
This would be enough movement to be detected if you have it in view of the
PIR sensor.

Jim

{Original Message removed}

2010\05\22@111646 by Michael Watterson

face picon face
Charles Craft wrote:
> I've had a couple hotel rooms recently that use PIR to enable/disable
> the HVAC system.
> If the sensors are in a bad spot or not sensitive they shut down the A/C
> when I'm asleep.
>
> As I understand PIR, you need movement across the field to trigger the
> sensor.
> Short of building a robot with an IR LED that moves around the room any
> thoughts on
> how to trigger the sensor with a static piece of hardware?
>  
Three IR LEDs on a horizontal bar, occasionally pulsed A -> B -> C then
C -> B -> A, appropriate distance from PIR. Or possibly even three small
Metal Oxide  resistors!

even if 100mA peak, the average current could be < 1mA  so more than
2000 hrs on pair Alkaline AA cells (the  NiMH may self discharge too
fast) or a single iPod type replacement 2200mA LiPoly flat pack.

One PIC or 4 x 555 timers with a few diodes. Also 3 transistors to get
the peak current.

2010\05\22@113353 by Olin Lathrop

face picon face
Michael Watterson wrote:
> Three IR LEDs on a horizontal bar, occasionally pulsed A -> B -> C
> then
> C -> B -> A, appropriate distance from PIR. Or possibly even three
> small Metal Oxide  resistors!

The PIR sensors I've worked with are not that sensitive to near IR LEDs.
They work, but you'd need them close.

LEBs (Light Emitting Bulbs) make great IR emitters that PIR sensors pick up
easily.  You don't need to run them at full voltage either.  If you can see
a orange glow, they're putting out plenty of IR.  Used that way, they last
pretty much indefinitely.  Unlike resistors, they are actually intended to
radiate.


********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
(978) 742-9014.  Gold level PIC consultants since 2000.

2010\05\22@114725 by Russell McMahon

face picon face
> Three IR LEDs on a horizontal bar, occasionally pulsed A -> B -> C then
> C -> B -> A, appropriate distance from PIR. Or possibly even three small
> Metal Oxide  resistors!

That was the immediate thought - but why not just 2?
A / B / A / ...
OR
A / AB / B / AB / A / ...

Also, if what it wants is a change in one field relative to the other
it MAY be enough to use one LED or bulb (Olin) occasionally activated.

The servo solution is not a wholly silly one. Small. Cheap enough. You
now your PIC is watching over you while you sleep. Add a mosquito
swatter and whine detector for double duty.

For extra points get the desk lamp and rig a curtain or piece of paper
to flutter in the breeze from the air conditioning and modulate the
lamp. If no lamp you may need to use a mirror to divert light from a
fixed lamp :-)

You may need to take a roll of duct tape with you, the rest should be
on site. Worked for Apollo 13.


                Russell.

2010\05\22@115353 by Picbits Sales

flavicon
face
You could borrow my wife - she moves about so much at night the A/C would
never turn off.

You might not get any sleep and the snoring might annoy you but you'll be
nice and cool ;-)

Dom

{Original Message removed}

2010\05\22@121432 by Oli Glaser

flavicon
face
Everything that has been mentioned, plus I'm thnking an IR laser
diode/pointer pulsed directly at the sensor occasionally might work (sort of
a "brute force" approach, which may trick it into thinking movement is
taking place) Another reason for this is (not that I know anything about PIR
nor have I used it) doesn't it (usually) have to be a rather large IR
emitting body? i.e. so small amimals and such cannot activate the sensor
easily?  That's another option of course - bring a largish animal with you
and let it loose in the room overnight :-)


2010\05\22@171826 by YES NOPE9

flavicon
face
I presume an ordinary lamp with incandescent bulb that is fed enough  
current to make it glow
red would trip the PIR if it went on/off every few minutes.
gus


{Quote hidden}

2010\05\22@173230 by Russell McMahon

face picon face
> I presume an ordinary lamp with incandescent bulb that is fed enough
> current to make it glow
> red would trip the PIR if it went on/off every few minutes.

They tend to work on differences caused by a heat source moving across
their field. The plastic fresnel lens enhances the results of objects
moving through 'zones'. If the IR source increases relatively evenly
across the field of view or part of it it probably wouldn't work.


              Russell

2010\05\22@175327 by Dario Greggio

face picon face
YES NOPE9 ha scritto:
> I presume an ordinary lamp with incandescent bulb that is fed enough  
> current to make it glow
> red would trip the PIR if it went on/off every few minutes.


but then the strong light will wake you up... :)


--

Ciao, Dario
--
Cyberdyne

2010\05\22@191701 by Olin Lathrop

face picon face
Russell McMahon wrote:
> They tend to work on differences caused by a heat source moving across
> their field. The plastic fresnel lens enhances the results of objects
> moving through 'zones'. If the IR source increases relatively evenly
> across the field of view or part of it it probably wouldn't work.

But turning on a IR source suddenly in any one position is likely to work.
The only reason is wouldn't is if the illumination on both ends of the
sensor just happened to be the same for that particular position, which
would be quite tricky to arrange on purpose if you tried.  Two bulbs a few
inches apart would almost certainly work.


********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
(978) 742-9014.  Gold level PIC consultants since 2000.

2010\05\22@191856 by Olin Lathrop

face picon face
Dario Greggio wrote:
> but then the strong light will wake you up... :)

You don't have to turn on the bulb far enough to emit "strong light".  You
get plenty of IR well before that.  Use just enough voltage so that it glows
orange.  It can also be a small bulb close up.


********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
(978) 742-9014.  Gold level PIC consultants since 2000.

2010\05\22@192525 by Dario Greggio

face picon face
Olin Lathrop ha scritto:
> Dario Greggio wrote:
>> but then the strong light will wake you up... :)
>
> You don't have to turn on the bulb far enough to emit "strong light".  You
> get plenty of IR well before that.  Use just enough voltage so that it glows
> orange.  It can also be a small bulb close up.
>

I see, I never tried that though, only observed some of such behaviour
by chance.

I use NAIS-Matsushita Napion PIR in my home automation system, thay have
TTL output, quite good response and angle, size more or less like a 10mm
led.
They send RS485 messages to the PC, and the Windows software does its
best to track and store recent activity and presence/absence per room.

--

Ciao, Dario
--
Cyberdyne

2010\05\22@204451 by Russell McMahon

face picon face
>> They tend to work on differences caused by a heat source moving across
>> their field. The plastic fresnel lens enhances the results of objects
>> moving through 'zones'. If the IR source increases relatively evenly
>> across the field of view or part of it it probably wouldn't work.

> But turning on a IR source suddenly in any one position is likely to work.
> The only reason is wouldn't is if the illumination on both ends of the
> sensor just happened to be the same for that particular position, which
> would be quite tricky to arrange on purpose if you tried.  Two bulbs a few
> inches apart would almost certainly work.

Yes, maybe, sort of.
I wasn't intending to state a sole position but only a comment on that one.
I'd previously said

> Also, if what it wants is a change in one field relative to the other
> it MAY be enough to use one LED or bulb (Olin) occasionally activated.

so Gus and Russell and Olin are all essentially agreed, for a change :-).

The uncertainty lies in how intelligent and/or crude the sensor is.
Dim memory suggests that they use a single sensor with the lens
effectively providing a "grating" which enhances some areas and blocks
or desensitises others. As you move an IR source across the zones you
create a modulation of the level at the sensor.

OK - lets see what MrWpedia says ... time warp ...

  "The sensor is often manufactured as part of an integrated circuit
and may consist of one (1), two (2) or four (4) 'pixels' of equal
areas of the pyroelectric material. Pairs of the sensor pixels may be
wired as opposite inputs to a differential amplifier. In such a
configuration, the PIR measurements cancel each other so that the
average temperature of the field of view is removed from the
electrical signal; an increase of IR energy across the entire sensor
is self-cancelling and will not trigger the device. This allows the
device to resist false indications of change in the event of being
exposed to flashes of light or field-wide illumination. (Continuous
bright light could still saturate the sensor materials and render the
sensor unable to register further information.) At the same time, this
differential arrangement minimizes common-mode interference, allowing
the device to resist triggering due to nearby electric fields."

So, what we all said, or not :-).
And adding processing power to that. as some do, makes it more uncertain.

If using one only light / LED etc a relatively point source would seem
best. This simulates moving a warm body in and out of a single zone
rather than between zones. How well it works seems liable to be quite
dependant on the sensor and it's intelligence,

Trying it would be extremely easy.
I like the curtain flapping in the air conditioner breeze as the
modulator though - seems apposite to use the system to trigger itself
:-).


    Russell

2010\05\22@231616 by Charles Craft

picon face
On 5/22/2010 8:44 PM, Russell McMahon wrote:
{Quote hidden}

I haven't checked if the HVAC sensors have a status LED when they're
tripped.
If so then I'll start by pointing the TV remote at it and see if a point
source works.

For those that offered wives and furry animals - I'll have get approval
from my wife. :-)

thanks
chuckc

2010\05\23@081957 by Olin Lathrop

face picon face
Russell McMahon wrote:
> The uncertainty lies in how intelligent and/or crude the sensor is.
> Dim memory suggests that they use a single sensor with the lens
> effectively providing a "grating" which enhances some areas and blocks
> or desensitises others. As you move an IR source across the zones you
> create a modulation of the level at the sensor.

There are some fancy sensors, but all the ones I've seen in commercial
motion detector applications are optimized for price.  The sensor has a
rectangular window with each end being sensitive to IR seperately.  Sensors
with more than two "pixles" exist, but I've never seen one in practise.  The
output voltage is one end minus the other, usually plus about one diode
drop.  This is a very low level signal, so is usually amplified then run
thru a window comparator.  The motion sensor trips if the signal ever gets
outside the threshold.  To give you some idea of the signal level, in one
design I amplified the AC component about 1400 to feed it into a PIC A/D.

The lens is usually of the cheap Fresnel stamped plastic type.  For burglar
detection applications the two ends of the sensor are arranged horizontally,
and the lens is optically actually 10-30 vertical cylindric lenses.  This
creates a bunch of vertical stipes in the field being viewed.  A object in
the field moving horizontally will alternately shine on opposite ends of the
sensor, sooner or later creating a large enough differential signal to trip
the motion detector.


********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
(978) 742-9014.  Gold level PIC consultants since 2000.

2010\05\23@082341 by Olin Lathrop

face picon face
Russell McMahon wrote:
> I like the curtain flapping in the air conditioner breeze as the
> modulator though - seems apposite to use the system to trigger itself
> :-).

Except that it's positive feedback and therefore subject to hysteresis and
latching.  If the system is off for long enough, then it will stay off.


********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
(978) 742-9014.  Gold level PIC consultants since 2000.

2010\05\23@115711 by Peter

picon face
Charles Craft <chucksea <at> mindspring.com> writes:
> Short of building a robot with an IR LED that moves around the room any
> thoughts on
> how to trigger the sensor with a static piece of hardware?

The IR equipped robot would not help, wrong wavelength and too small. A small
computer fan with some blades removed spinning on reduced voltage and a power
resistor on a heatsink underneath with thermostat set to 40 deg C should do the
trick. The box should not be too small, nor the fan. Pocket book size should
work, or make one you can tape on the PIR "eye". The fan chopper should not be
faster than 10Hz. There are no LEDs for the 10um IR wavelength range on the
market afaik.

PIR triggered room lights and A/C units are evil and cheap, some are counting
direction type (mounted to view the only entrance, they count people entering
and exiting), that is less evil but occasionally miscounts in one direction or
in the other. Instead, "occupancy sensors" exist for that purpose. Some use the
principle of the sensors in contactless IR thermometer and do not require
contact or motion to exist (dual band far IR), but they are expensive and less
frequently used. See for example MLX90614 from Melexis.

-- Peter


2010\05\23@120449 by Peter

picon face
Olin Lathrop <olin_piclist <at> embedinc.com> writes:
> LEBs (Light Emitting Bulbs) make great IR emitters that PIR sensors pick up
> easily.  You don't need to run them at full voltage either.  If you can see

That might work but one would have to use truly tiny bulbs so the inertia would
not be too large. They have to turn on (orange or whatever) within 0.2 sec or so
for the signal to pass the high-pass filter in the PIR electronics. Finding a
location that does this could be "interesting" even if the "jammer" is Pattfix
glued directly to the PIR sensor. It is well worth a try though. I'd take a 2
AAA cell flashlight with tiny bulb, remove one battery and replace it with a
nail (!) and then tape it on the sensor and try to trip it with the flashlight
switch. If it works, there is hope.

-- Peter

2010\05\23@155815 by Olin Lathrop

face picon face
Peter wrote:
> They have to turn on (orange or
> whatever) within 0.2 sec or so for the signal to pass the high-pass
> filter in the PIR electronics.

First, the bulb should be easily emitting substantial IR in 200ms.  Second,
the PIR sensors I've seen have a high pass filter with a longer time
constant than that.  Think about how long it takes for a burglar to walk
between veritical zones.  At best the zones are about torso sized.
Detecting only burglars that move at least 5 torsos/second wouldn't be
smart.  Keep in mind the only purpose of the high pass filter is to filter
out slow environmental changes, and to eliminate amplifier offset error.
200ms time constant is way too short.


********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
(978) 742-9014.  Gold level PIC consultants since 2000.

2010\05\24@140506 by Al Shinn

picon face
About light bulbs.
1 the filament of a light bulb can be turned on to whatever desired
level within almost arbitrarily fast time if one uses either a carefully
profiled voltage pulse or an active photo feedback circuit. Of course,
you can't turn it off fast this way.
2 It doesn't matter for the subject at hand because the wavelengths that
most pir detectors use are blocked by the light bulb glass. So you need
the glass envelope to heat up at "speed"
A loop of thin wire in air can be heated up rapidly and will cool fairly
fast.
--

Looking forward,
Al Shinn

More... (looser matching)
- Last day of these posts
- In 2010 , 2011 only
- Today
- New search...