Looking for cheap, simple PIR detector module

[Richard Rasker]

Hi all,

I'm designing the electronics for a flatscreen TV lift system, for tucking
away a flatscreen TV inside a low design-furniture cabinet (see
http://www.klink-osse.nl/ for an example).

As safety is an essential requirement for this type of machinery (the lift
system is easily capable of squashing or even severing fingers), I'm
looking for a reliable PIR detector module. The idea is to have it "look
down" from the back of the TV into the machinery, so that the system will
stop immediately if a body part or (warm-blooded) household pet is
detected. The PIR detector should run off something between 6 and 12 volts
(with a preference for 12V), consume no more than 10 mA, and provide a
momentary low or high level signal upon triggering. A sensitivity setting
is paramount.

I tried several cheap PIR detectors which are used for automatic door bells
or proximity warning, but these have a tendency to trigger too readily;
maybe they react to the changing heat pattern at the moment the lift starts
moving (the TV is warm after use, and the movement probably makes the
detector see quick changes in received heat), or maybe the PWM signal
(though within permissible EMC levels) makes it go haywire.

And oh, the cost may not exceed some 20 dollars for a complete module.
Schematics for PIR detectors are also welcome, but no doubt, existing
mass-produced modules are cheaper than anything I can make.

And yes, I know that this will /not/ secure the whole thing against crushed
fingers when the lid closes; we're contemplating several other measures for
this. Also, the remote control is IR, with mandatory "button-hold" action,
so that the operator must always be in sight of the system, and nothing
happens automatically.

Thanks in advance,

Richard Rasker

 

[John - KD5YI]

I hate when people go off topic when I ask for help in a newsgroup,
but if you'll forgive me this once...

I'm using a PIR sensor from Radio Shack. It's from Parallax (Parallax
part number 555-28027). it's about $10. It's not adjustable for
sensitivity as far as I can tell, but it's cheap. It gives continuous
triggering or single trigger. It has a Fresnel lens on the sensor to
give a wide angle of detection. You might be able to remove/replace
this lens and get the directionality/sensitivity you want. Mouser
carries this same detector for about $10, and others for a LOT more
(hundreds of $).

Much cheaper here:

http://www.futurlec.com/PIR_Sensors.shtml

 

[Phil Allison]

"Richard Rasker"

I'm designing the electronics for a flatscreen TV lift system, for tucking
away a flatscreen TV inside a low design-furniture cabinet (see
http://www.klink-osse.nl/ for an example).

As safety is an essential requirement for this type of machinery (the lift
system is easily capable of squashing or even severing fingers), I'm
looking for a reliable PIR detector module.

** There is no such animal.

Your idea is dangerous and absurd.

The idea is to have it "look
down" from the back of the TV into the machinery, so that the system will
stop immediately if a body part or (warm-blooded) household pet is
detected.

** Very stupid and dangerous idea.

PIR detectors fail, fingers may be too cold to register.

And yes, I know that this will /not/ secure the whole thing against
crushed
fingers when the lid closes; we're contemplating several other measures
for
this.

** The problem is mechanical in nature.

Just limit the available torque on any motor to a safe value.

Also, the remote control is IR, with mandatory "button-hold" action,
so that the operator must always be in sight of the system, and nothing
happens automatically.

** You must make it fail safe by nature.

Relying on PIR detectors is 100% absurd.

Time for a re-think.


..... Phil

 

[Martin Riddle]

Hi all,

I'm designing the electronics for a flatscreen TV lift system, for
tucking
away a flatscreen TV inside a low design-furniture cabinet (see
http://www.klink-osse.nl/ for an example).

As safety is an essential requirement for this type of machinery (the
lift
system is easily capable of squashing or even severing fingers), I'm
looking for a reliable PIR detector module. The idea is to have it
"look
down" from the back of the TV into the machinery, so that the system
will
stop immediately if a body part or (warm-blooded) household pet is
detected. The PIR detector should run off something between 6 and 12
volts
(with a preference for 12V), consume no more than 10 mA, and provide a
momentary low or high level signal upon triggering. A sensitivity
setting
is paramount.

I tried several cheap PIR detectors which are used for automatic door
bells
or proximity warning, but these have a tendency to trigger too
readily;
maybe they react to the changing heat pattern at the moment the lift
starts
moving (the TV is warm after use, and the movement probably makes the
detector see quick changes in received heat), or maybe the PWM signal
(though within permissible EMC levels) makes it go haywire.

And oh, the cost may not exceed some 20 dollars for a complete module.
Schematics for PIR detectors are also welcome, but no doubt, existing
mass-produced modules are cheaper than anything I can make.

And yes, I know that this will /not/ secure the whole thing against
crushed
fingers when the lid closes; we're contemplating several other
measures for
this. Also, the remote control is IR, with mandatory "button-hold"
action,
so that the operator must always be in sight of the system, and
nothing
happens automatically.

Thanks in advance,

Richard Rasker

I agree with Phil.
Limit the torque of the motors. Maybe with auto reversing like
automotive power windows.
Perhaps Jim has an IC for that ;D

Cheers

 

[Phil Allison]

"Martin Riddle"

I agree with Phil.

** That dull thud you heard we me falling over ......

Limit the torque of the motors. Maybe with auto reversing like automotive
power windows.

** I suspect the issue is with * lowering* that flat panel screen.

The motor is then probably opposing the weight of the screen and toque
limiting is not relevant to finger safety.

The answer might be to add a counterweight system on a pulley ( like all
elevators use ) so the motor IS pulling the screen down.



.... Phil

 

[Phil Allison]

"Phil Hobbs"

Using a pyroelectric for this is a recipe for Kitty Squash. They're
charge transducers, rather than current, so it won't even see a stationary
animal. The reason for the segmented Fresnel lens on a porch light sensor
is that it forms a dozen or so images that cross a split (2 element
differential) sensor as you walk by. That makes an AC signal that the
back end has no trouble recognizing.

No motion, no signal.

** Nothing guarantees that an infant's finger or feline's paw will actually
always be any warmer than the surrounding surfaces either.

That's the worst kind of safety system--the kind that lulls you into a
false sense of security while actually doing bupkiss.

** Plenty of them about.

Like nearly every video camera set up to provide " security " ....


.... Phil

 

[Richard Rasker]

[snip]

Using a pyroelectric for this is a recipe for Kitty Squash. They're
charge transducers, rather than current, so it won't even see a
stationary animal. The reason for the segmented Fresnel lens on a porch
light sensor is that it forms a dozen or so images that cross a split (2
element differential) sensor as you walk by. That makes an AC signal
that the back end has no trouble recognizing.

No motion, no signal.

Stationary animals or children are no risk: the sliding top of the cabinet
is closed at all times, except when the lift is raised or lowered. This
means that it's impossible to get an animal or limb inside without physical
movement during the time that the machine is active.

Also, motor current is monitored, and the system stops whenever unexpected
resistance is detected. I just want to add an extra safety measure which
will usually stop the machine before body parts get stuck inside.

That's the worst kind of safety system--the kind that lulls you into a
false sense of security while actually doing bupkiss.

I fully agree with this notion. It's not my intention to just stick in a PIR
device and simply declare the machine safe -- even if I'm satisfied with
the overall safety, we're going to have it tested by an independent
certification authority. But I have to start somewhere, and test different
options, hence my question here. I've also considered ultrasonic devices,
but these are both more expensive and less relaible than PIR.

You can use a chopper, maybe on the drive motor shaft, to turn a static
temperature difference into an AC signal that you can detect with a
pyroelectric.

Hmm, thanks, I'll add it to my list of options.

Cheers

Phil Hobbs

PS: Philips used to make pyroelectric vidicons whose sensitive material
has the most jaw-cracking name in technology: deuterated triglycine
fluoroberylate. Try saying that three times fast.

Anyway, thanks again, best regards,

Richard Rasker

 

[Richard Rasker]

"Richard Rasker"

** There is no such animal.

Your idea is dangerous and absurd.



** Very stupid and dangerous idea.

PIR detectors fail, fingers may be too cold to register.



** The problem is mechanical in nature.

Just limit the available torque on any motor to a safe value.




** You must make it fail safe by nature.

Relying on PIR detectors is 100% absurd.

I agree, and I certainly don't intend to rely on PIR only. Motor torque is
already monitored, and any unexpected resistance causes the system to stop
immediately. Some more mechanical measures are in place -- most parts which
may otherwise pose a risk (e.g. with a shearing action) have covers and
guides which push objects aside rather than squashing them. Still, there
are some parts inside that can't be fully shielded this way. And even if
the motors have a torque trip, a finger or arm will still get stuck at that
point. We're looking into the option to reverse motion for half a second in
those cases, but that of course may not introduce new risks.

Anyway, we going to have to it certified, so I'm not the one to call it
safe. But if I can significantly improve safety with a simple measure, why
not?

Time for a re-think.

We're re-thinking it continuously ;-) But thanks for your reply.

.... Phil

Richard Rasker

 

[Phil Allison]

"Richard Rasker"

I agree, and I certainly don't intend to rely on PIR only. Motor torque is
already monitored, and any unexpected resistance causes the system to stop
immediately. Some more mechanical measures are in place -- most parts
which
may otherwise pose a risk (e.g. with a shearing action) have covers and
guides which push objects aside rather than squashing them. Still, there
are some parts inside that can't be fully shielded this way. And even if
the motors have a torque trip, a finger or arm will still get stuck at
that
point. We're looking into the option to reverse motion for half a second
in
those cases, but that of course may not introduce new risks.

Anyway, we going to have to it certified, so I'm not the one to call it
safe. But if I can significantly improve safety with a simple measure, why
not?

** False argument.

You CANNOT significantly improve the safety of your contraption with a cheap
PIR.

It will have gimmick value at best - and doing that is practically
criminal.

If you counter balance the weight of the screen, then only small amounts of
motor torque are ever needed.

I think you are an ass.

Go away.


..... Phil

 

[Phil Allison]

"John Fields"
"Phil Allison"

"Phil Hobbs"

** Nothing guarantees that it will be significantly different either - or
even traverse in a way that registers on a device that was intended to
detect only whole bodies.

Most PIR security devices are designed to IGNORE pets.

None I ever saw were designed to work at such closes ranges.

The OP's idea is 100% crackpot.


...... Phil