Haking the newer Harbor Freight Driveway Sensor

[amdx]

I want to hack the newer Driveway sensor.
I had the old ones all figured out, now they went to all surface mount
parts, with a different design.
I just want to a add a button to trigger the peizo alarm in the receiver.
It has an 8 pin chip labeled AT12L and 319H.
I think the 319H is the relevant number, but not sure.
Might be a pic.
The receiver has an RF section, the RF section output feeds the input
to the chip above.
The signal looks like this.
> https://www.dropbox.com/s/et3bmc3pc5y2rw4/Driveway%20sensor%20before%20trigger.jpg?dl=0

After the the sensor/transmitter transmits to the receiver the signal
changes to this,
> https://www.dropbox.com/s/foe5lvaqhprl4zg/Driveway%20sensor%20after%20trigger.jpg?dl=0

The 319H has one 2.5V on pin 1, Pin 8 is ground.
Pin 3 outputs a square wave on trigger to drive a piezo.
Pin 4 is shown in the dropbox pictures above. I think that's the input
to the 319H.
Pin 7 goes high after trigger to drive Leds.
So, There's no nice place to pull up or down to trigger pin 3 to
make the peizo sing.

Can someone explain what is happening on pin 4 at trigger.
Is there a way to fool it with a simple switch, or is that a code that
the 319H is reading?

Mikek

[amdx]

Just bumping for someone to take interest. :-)
Mikek

[gnuarm.del...@gmail.com]

I would guess the AT12L is an indication it is an Atmel custom chip, likely an MCU with custom programming.

I can't say I understand what your device is. Is this a remote control device with a remote button to activate something, the piezo and LEDs only? I think you will have to wire up your own piezo to the button. Mallory makes some nice alerts that are designed to be panel mounted.

Rick C.

[bloggs.fred...@gmail.com]

On Thursday, August 2, 2018 at 9:45:14 PM UTC-4, amdx wrote:

All these modern systems transmit a code to identify the particular sensor type/location to the master control. It's probably simplest to get another dummy sensor and activate that to initiate the test.

[amdx]

Oh, sorry, it is the Bunker Hill Driveway Sensor sold at Harbor freight.
You have a transmitter that will sense when someone or something
enters it's area. Pretty sure it uses a PIR sensor.
The transmitter then sends a signal up to 400ft (?) to the receiver.

I think the 319H is the relevant number, but not sure.

>> Might be a pic. I think this, >
https://www.mouser.com/datasheet/2/268/40001897A-1108654.pdf
And then go back and read the section above.
Mikek

[gnuarm.del...@gmail.com]

Ok, so you are talking about the receiver which doesn't really care much what is doing the transmitting I expect. So the RF link is most likely the series of pulses you are seeing on the output of the RF receiver. Since this is battery operated, I don't expect a signal is being transmitted continuously as the transmitter uses a single 9 volt battery. This would say to me the signal you are seeing on pin 4 is not the received RF.

I don't have any suggestions on how to false trigger the thing from a push button without adding an MCU between the receiver chip and the MCU on the board. This MCU can generate whatever signal is required. But that would be some work and I expect you are looking for a quick fix.

What is the part number of the RF chip?

Rick C.

[Jasen Betts]

On Thursday, August 2, 2018 at 9:45:14 PM UTC-4, amdx wrote:
> I want to hack the newer Driveway sensor.
> I had the old ones all figured out, now they went to all surface mount
> parts, with a different design.
> I just want to a add a button to trigger the peizo alarm in the receiver.
> It has an 8 pin chip labeled AT12L and 319H.

Why not just buy a wireless doorbell?

> Can someone explain what is happening on pin 4 at trigger.
> Is there a way to fool it with a simple switch, or is that a code that
> the 319H is reading?

no. but do you know how pyroelectric sensors are read?
feed that input with a few Hz at low amplitude.

--
ت

[amdx]

On 8/6/2018 10:03 PM, Jasen Betts wrote:
> On Thursday, August 2, 2018 at 9:45:14 PM UTC-4, amdx wrote:
>> I want to hack the newer Driveway sensor.
>> I had the old ones all figured out, now they went to all surface mount
>> parts, with a different design.
>> I just want to a add a button to trigger the peizo alarm in the receiver.
>> It has an 8 pin chip labeled AT12L and 319H.
>
> Why not just buy a wireless doorbell?

I had one of these driveway sensors modified and it worked fine for
years. My business is 3 steps down, I put the sensor on the middle step,
then when someone walked down the steps, the receiver was triggered and
the obnoxious noise from the piezo would wake me up. Then maintenance
changed the steps, the jerk must have been very curious what was inside
this pvc pipe. So he cut the wire and took it. When I ask him about it
he didn't know anything about it. He is what we call a liar.
So, now I'm modifying another one. I have to modify it because it will
trigger on every car that goes by, every pedestrian and every customer
that stands in front of my while we do business. Just want it to trigger
in that two or three foot area as they descend the stairs.
Doorbell, yes, I just wanted the doorbell button to cause the same
piezo to sound, that's in the driveway receiver.

>
>> Can someone explain what is happening on pin 4 at trigger.
>> Is there a way to fool it with a simple switch, or is that a code that
>> the 319H is reading?
>
> no. but do you know how pyroelectric sensors are read?
> feed that input with a few Hz at low amplitude.
>
> --
> ت
>

No, I didn't but I did look at this site and got some idea.
> http://www.fuji-piezo.com/prodpyro.htm

I'm still lost, as the rf section has that constant pulsing output that
changes when the receiver gets triggered.

I took the perfectly working assy, wallwart, voltage regulators,
receiver and sensor, which I had working just fine at home, down to the
business.
I wired it all up and installed the sensor where it needs to be and
the receiver, just will not stop beeping.

I'm going to try another section of the fresnel patterned window and
mount it differently. I have the PIR looking through a 1/2" piece of
tubing with the window on the end. (same as the old unit)

I tried twice to clean this off my screen --> ت
Eyes aren't what they used to be.
Mikek

[mike]

Why can't you just leave the fresnel where it is and modify the shape of
the pipe to get the coverage you want? Have it view only the bottom step.

[amdx]

Maybe worth a try!

[Jasen Betts]

SFAIK they use OOK in the transmitter. The receiver adjusts the gain until
it sees 50% signal

It'll be receiving everything else that shares the same band (car
remotes, door remotes, doorbells, noise, etc. So you'll get binary noise
until a nearby transmitter comes on and the AGC pushes the noise floor down
below the threshold.

> I took the perfectly working assy, wallwart, voltage regulators,
> receiver and sensor, which I had working just fine at home, down to the
> business.
> I wired it all up and installed the sensor where it needs to be and
> the receiver, just will not stop beeping.

I hate it when that happens :(

> I'm going to try another section of the fresnel patterned window and
> mount it differently. I have the PIR looking through a 1/2" piece of
> tubing with the window on the end. (same as the old unit)

I've seen similar units with black electrical tape covering most of
the fresnel zones, resulting in a narrower view angle..

> I tried twice to clean this off my screen --> ت
> Eyes aren't what they used to be.

:) I was looking through unicode for some reason and came across that
character, and thought "He looks happy"

--
ت

[amdx]

This morning I may have noticed why that happened!

My business is on the water, 3 steps down to the dock.
The dock is against a concrete wall that goes below the water line.
I point my sensor at the concrete wall, not a problem, but,
when they replaced the steps they also replaced dock boards.
The original dock boards were 6'6" long, the replacements are 6 0"
long, leaving a 6" gap along the wall. I noticed this morning the sun
reflecting off the ripple of the water and hitting the wall.
My theory is the rippling reflection though the gap was causing the
continuous beeping.
On my next test, I'll put a board over that gap.
The marina owner saved money by cutting a 12' board in half, it cost
me when a dropped phone went into the water through that gap.
"It's always something"
Mikek

[Jasen Betts]

On 2018-08-09, amdx <noj...@knology.net> wrote:
>
> My business is on the water, 3 steps down to the dock.
> The dock is against a concrete wall that goes below the water line.
> I point my sensor at the concrete wall, not a problem, but,
> when they replaced the steps they also replaced dock boards.
> The original dock boards were 6'6" long, the replacements are 6 0"
> long, leaving a 6" gap along the wall. I noticed this morning the sun
> reflecting off the ripple of the water and hitting the wall.
> My theory is the rippling reflection though the gap was causing the
> continuous beeping.

I had a similar problem with sunlight through tree-leaves.

--
ت

[gnuarm.del...@gmail.com]

You guys know how the "motion" sensor works, right? They use a "lens" that causes a moving IR source to vary the intensity of the light falling on the photoreceptor. I haven't taken one apart, but I expect they use at least two photoreceptors so changes in overall brightness doesn't trigger the device. So light patches varying in intensity will trigger the device if they contain IR energy.

A friend was talking about a sensor that was getting triggered at night outside her mother's home which upset the mother thinking someone was trying to get in. The daughter though the IR sensor was picking up tree branches moving since it is a "motion" detector. I tried to explain that the branches were the same temperature and so wouldn't trigger the detector. But then I realized, if a strong light were behind the branches it might have enough IR to trigger when the branches moved. She didn't see the need to involve a light source.

Rick C.

[Jasen Betts]

As I understand it:

The sensor produces current as its temperature changes.
The pyroelectric element has two cells and the lenses in adjacent
zones of the window focus incoming infrared on opposite cells.
the two signals are fed to opposite inouts of a op-amp and the
amplitude of the output signal is checked

--
ت

[Joseph Gwinn]

On Aug 11, 2018, Jasen Betts wrote
(in article <pkm847$s43$1...@gonzo.alcatraz>):

It’s called a pyroelectric IR sensor:

.<https://en.wikipedia.org/wiki/Passive_infrared_sensor>

.<https://learn.adafruit.com/pir-passive-infrared-proximity-motion-sensor/how-
pirs-work>

Joe Gwinn