Cadillac Intrusion Sensor

[Johndm1957]

Hello,

I have identified a faulty Ultrasonic motion sensor on my 2009 Cadillac STS which has been causing the alarm system to go off at random. The fault has got progressively worse over the last year, started with random alarms every few weeks, now it will trigger the alarm after only 15 minutes.

Inside there is no sign of any problem, although only the back of the circuit board is visible when removed from it's case

Since I removed the internal Ultrasonic motion sensor, the alarm has behaved correctly, but I now have a message on the information centre to 'service alarm system' as the system has a part removed.

A replacement part is $800, and the cost to import to the UK would be obscene, so I have the following questions.

Would there be any way to test and repair the sensor?

The sensor has 3 wires, would there be a way to bypass and fool the system that it had been replaced?

Help appreciated,

John

 

[Alec_t]

Are there any type/part numbers or other markings to identify the sensor? If so, some technical info might be found by googling, or availble from the manufacturer.

 

[Johndm1957]

GM Part No. 10388029

I can't find any tech info online.

 

[73's de Edd]

.

Sir John . . . . .

WOW . . . .WOW . . . .WOW . . ..what a price for that unit . . . . and they even make you think that you are getting such a DEAL already !

I finally found a pic of it from the frontal perspective :

Would I be correct in your disassembling the unit enough, to pull one of the units, that you have close eyeballed one or both of the two xmit-receive transducer units?

Is there a possibility of there just being a transducer connector pin loosened within its solder blob to the PCB , on either the xmitter or receivers ultrasonic transducers, thus accounting for the alarm triggering random indeterminacy.

Correct me from what you have seen but I am thinking of the proximity sensor aspect being the use of two ultrasonic transducers.
One is transmitting out an approximate 40Khz signal while the other unit is a receiver and receives the bounced signal from the xmitter.
A missing pulse detector associated with the receiver is timed for responding to an interrupted return signal, if being greater than a few seconds.
Then it gives a security breach response signal.

Look at my pic and see if its xmitter and receiver units look like this:

I've seen them with either the window screen protector on the front OR the perforated grill cover treatment.

I don't have my metric tape measure in me pocket now but seems like their overall diameter was right at 3/4 of an inch.

Waiting to read your response . . . . . .


73's de Edd


.

 

[Johndm1957]

Hi there, and thanks for the reply.

Here is the unit, left is marked TX and right RX. The cover on the other side won't come off easily, and not being an expert with electronics I don't want to force it and break something.
The units are approx. 9mm in diameter.

I was hoping to bypass somehow, or do you think a TV repair shop could fix something like this?

Regards,

John

 

[73's de Edd]

.


Sir John . . . . . . . .


Yes, that should be a piece of cake for a TV repair shop . . . PARTICULARLY, if they were active in the 1973 through 1980 time frame they would be familiar with this particular type of ultrasonic transducers.
That would be a man about 55 years of age or older. . . .just about your age . . . . .if you are a '57 model.
By the 1980's the Infrared technology was replacing it.
These transducers are still readily available.



I was hoping that you were telling me that upon inspection from a bright light that you could easily see the little aluminum-i yum-yum disc just back inside the unit, which is mechanically attacted to the ceramic
piezoelectric element at its rear.]

You will just have to try a pair to see if ONE OF THEM is your problem.
Or if you have to " fool" the units, a very small value of ceramic or silver mica capacitor coupling signal between the TX-RX units would give an all safe condition. At the loss of detecting anyone that successfully
broke and entered and was foraging around inside of your car.
But all an alarm does is tell you that your car is being breached, if you, or possibly a CARING bystander are within earshot of it sounding off.

Here are only two offerings of "new old stock" units that are available, and I know that SOMEWHERE ? in among my stored electronics, that I have in excess of 100 new units .


http://www.electronicsurplus.com/pa...rub40k25-40khz-new-ultrasonic-tranducers-rcvr

http://www.electronicsurplus.com/zenith-49-1267-02-ultrasonic-transducers-pick-up


( Did you stop to clean your glasses or do a double take on the offered .75 cent NOS item ? *****)

Can you provide a frontal closeup shot of that units components side, I'd just like to see its price justification, by virtue of seeing its component / technology/sophistication being utilized.

Five hunna dollah +*****. . . . . Jeeeeeeeeeez.


73's de Edd



.

 

[Johndm1957]

Thanks for the info, I am indeed a 57 model, actually born 1 day before there was any man made object orbiting the Earth

I can't get a picture of the component side without removing the cover, and would prefer to leave that to an expert so not do cause damage.

I'll contact some local 'Older' TV and Audio Repair Centres to see if they can help and report back ok

Those components you listed must have had a price increase making the unit so expensive on parts websites?

Regards

John

 

[Alec_t]

Do you have a wiring diagram of your Caddy? If we know what connects to all the three pins of the sensor (presumably two of them are the power supply) it might help to decide how to fool the Caddy.
I see the maker of the sensor (May & Scofield of Basingstoke, UK) is still in business. There's a faint possibility someone there might be willing to part with some info?

 

[Johndm1957]

I have so far been unable to find a wiring diagram, but I have just emailed May & Scofield to see if they still supply these boards

 

[duke37]

Put a cat in the car overnight to catch the mouse.

 

[73's de Edd]

.

Sir John . . . . . . . .

You are not alone . . . . .

Hot link . . . .
http://www.cadillacforums.com/forums/cadillac-forum/t-716170.html


73's de Edd

( With this being just the proximity sensor portion . . .of the whole alarm system . . . . . dare we ask what a whole alarm system would cost ? )


.

 

[Johndm1957]

My other thought was to just disable it as follows, bear in mind I ain't no sparky.......so would appreciate some advice so I don't do serious damage to the system!

2 of the pins must be positive and negative and 3rd pin would be a trigger, so if I identify pos and neg the cut the trigger wire to see what happens with the unit replaced.

If alarm still goes off, would the trigger wire would need a positive connection to it rather than just cut, or would connecting it to the cars 12v damage something?

 

[Johndm1957]

So I found pictures of the console and a plug diagram.
Would there be a way to bypass the faulty sensor and make the car think its in the circuit to stop 'service alarm system' message when the sensor is removed?

 

[duke37]

If you just want to disable the sensor, you could try covering the transmitter and receiver with a sound absorbing layer (felt?) and a thick plastic sheet.

 

[Johndm1957]

If you just want to disable the sensor, you could try covering the transmitter and receiver with a sound absorbing layer (felt?) and a thick plastic sheet.

Already tried covering with electrical tape, then tried Blu Tac but it still triggered the alarm system

 

[Alec_t]

If alarm still goes off, would the trigger wire would need a positive connection to it rather than just cut, or would connecting it to the cars 12v damage something?

I think leaving it just cut is unlikely to work, since the wire is described as a 'signal' wire. If you're going to risk connecting the trigger wire to either battery positive or the negative line (which could be either Pin B or Pin C, but my guess would be Pin C), then you could do so via a ~10k resistor rather than directly, so as to reduce the risk of damage.

 

[73's de Edd]

.

Sir John . . . . . . . .


I see that these have popped up in the interim:


" If you just want to disable the sensor, you could try covering the transmitter and receiver with a sound absorbing layer (felt?) and a thick plastic sheet."


" Already tried covering with electrical tape, then tried Blu Tac but it still triggered the alarm system"


The system xmitter transducer is flooding the interior with a 40khz ultrasonic bath and "missing pulse" detection is sensed by the ultrasonic receiver transducer, in any intrusion aspect.
Your covering it up, gave a fail condition in the magnitude of the presence of a baby elephant !


HOW TO . . . . . . " DO IT TO IT"

This proximity sensor COULD send an exotic serial data stream to the Security systems main unit, OR it could send a MERE LOGIC low or logic HIGH.
I'm sort of thinking that the latter is the case. In the overall picture, making this whole EXOTIC intrusion sensor 'thang" as dumb as an adjunct plain 'ole door jamb switch.( NOTICE. . . . ."Door is ajar" )
As far as now figuring it out, we neeeeeed to know which is which, as far as its three terminals are concerned. And its easiest access might be the 3 female portions of the plug that this unit mates with, since you now have that unit unplugged.
Initially you would need a ground for your metering / instrumentation and the only nearby access that I think that you might be finding for SURE will be the negative connection of your CIGAR lighter.
While you are there, also jab your DC metering positive lead up inside to the other terminal, to ascertain what your dormant battery voltage at this point is being.

Power up the alarm system so that normal voltages are now present at this female Berg connector.

Then remain keeping negative metering probe to the negative connection of your CIGAR lighter and probe the three cavities of the Berg strip connector.
If finding that you have no POINTED probe to reach inside of that small of a cavity, twirl on 10-15 turns of ~18-20 ga bare copper wire with the final short end length, it then being your mini pointee-makee-connectee probe

On reading the connectors three possibilities, you probably will find no reading on one pin, suggesting that one is being the NEGATIVE power portion of the connector and the other two pins might both be showing POSITIVE readings, and one might be close to the same value as your battery voltage.

Even the other POSITIVE reading might be pretty close to your battery voltage.

So o o o o o o here is how you sort out which is supplying a DYNAMIC battery voltage and current capability source, and the other is having but a mere STATIC voltage level with its WEAK current capabilities.

To confirm which is which, you get a 1K ( 1/4 W - 1/2 W - 1W - 2W- . . .what' cha' got cha' ?) resistor and fold its leads around so that it can "plug" into the Berg strip connector.
Then you power up the 'ole "Cataract" STV, SST, BLT* . . . . whatever . . . . in order to get its constant power to this alarm system connector .. . . . . . (* Bacon Lettuce Tomato)

Have one resistor lead plugged into the NEGATIVE and randomly select one of the other positive reading Berg connections in which to plug the other free lead of the resistor.
Meter across the resistor, and you probably will have to exert a slight sideways pressure to get a reading, since the resistors loose wire leads don't exactly even come close to making a tight friction PLUG in .

Take note of THAT voltage reading.

Do the same test with the other Berg pin possibility.

The connection which PULLED down your voltage reading the MOST is being the alarm logic output pin.

MAKING AN IN SYSTEM EVALUATION . . . . .of a procedure for "FOOLING" the system

To validate on the alarm systems " bypassing" of the intrusion alert feature.
If you were reading POSITIVE VOLTAGE in the alarm system " Sensor " pin of the Berg strip connector, the intrusion alarm portion will bring that connection to a logic LOW
on detection of intrusion.

I AM NOW ABRUPTLY STOPPING, since you might have only found ONE positive voltage on the Berg strip, in which case the system is using NEGATIVE logic.. . . . . . .
I await your findings, such that I will then only have to write up for the one pertinent condition.

Aside thought :
Remember that I had already given one solution in the manner, by the use of a low value cap to cross couple constant signal between the Xmitter an Receiver transducers.
But that involves the soldering access to the back side of the now covered up board.
The now conceived work around will only require access to the Berg plug.


73's de Edd




.

 

[Johndm1957]

Hi again 73's de Edd,

Most comprehensive reply there, just a couple of SMALL problems with your instructions....please bear in mind my electronic knowledge is rather....ahem...tiny

Anyways, I do have a basic el cheapo Multimeter with pointy probes, it has a black one AND a red one
Having googled 'Berg connecter' in now 100% in there with ya on that one...
First tests make sense, and there is a switch to temporarily disable the internal sensor if I want to leave kids, animals, fish or any other moving items in the car for the day, so I will test the berg plug with the alarm in full set state AND temporary internal sensor disabled to compare readings from the berg plug

It's the next bit where I'm getting lost, as I don't have any 1K ( 1/4 W - 1/2 W - 1W - 2W- . . .what' cha' got cha' ?) resistors.

Will report back when tests completed

 

[Johndm1957]

Ignore the above, I've found a guaranteed replacement on Ebay for £100 so gone and bought it.

If the replacement is faulty it will be returned for a refund, and.... I'll be back