🚀 go-pugleaf

In article <959@obelix.UUCP>, per-el@obelix.UUCP (Per Elmdahl) writes:
> >	This is also, BTW, one of the ways those stolen-item detectors in
> >stores work.  The big plastic clips contain some sort of passive tuned
> >circuit, with a diode detector.  The gizmos on either side of the door emit
> >RF at some frequency (presumably up in the several hundred Mhz range) and
> >listen for harmonics comming back.....
> 
>   A girlfriend of a friend of mine found one detector in a box she had
> got from a shop. I opened it and found a coil with about 10 turns and 
> a capacitor in it. No diode! I was puzzled, because I did also think
> it used harmonics for detection. I measured the resonant frequency
> with a grid-dip meter, and found it to be about 5.5 MHz. Later I went
> into a radio-shop and listened on a shortwave radio and found a strong
> signal around 5.5 MHz. I started to think about what principle it worked
> it used. 
> 
>  1. Send a signal and measure the loss when the resonant circuit enters
>     the e-m field. (Would that work? Would it not be very sensitive for
>     other things entering the field.)

	You got it!  All theft-prevention systems that I am aware of use
this principle.  The operation is not unlike a giant fix-tuned grid-dip
meter.
	The detector element consists of an LC circuit, which when placed
in an electromagnetic field at its resonant frequency will absorb energy
from the field, thereby affecting the power drawn by the oscillator
circuit in the detector apparatus which generates the field (this is somewhat
over-simplified, but close enough).
	One of the first theft-prevention systems was manufactured by a
company called Sensormatic, and used a UHF excitation frequency.  The
detector elements were rather small and were like a flexible printed circuit
sandwiched between two pieces of cardboard.
	Since other metallic objects (keys, jewelry, etc.) could conceivably
cause UHF energy absorption within the sensitivity threshhold of a UHF device,
HF systems in the range of 2.0 to 10.0 MHz have become more prevalent.  While
HF detector elements are larger than the UHF variety, they offer a greater
detection range and fewer false alarms.  Stores are _awfully_ concerned
about false alarms - since they usually generate ill will among customers.
An example of an HF system using detector elements about 1-1/2 by 2 inches by
1/8 inch thick is manufactured under the trade name "No-Go".
	The HF systems generally use loop antennas to create a confined
electromagnetic field, and thereby prevent interference to adjacant detector
systems and other radio services.
	There are still UHF systems on the market, and I believe that at
least one of them uses an element having two absorption frequencies to
minimize false alarms.

>  2. I talked to a service man at a shop. He was adjusting the boxes at
>     the door. I looked into them but could not figure out how they worked.
>     The service man did not seem to know either, only how to 
>     adjust it. He talked about *two* frequencies, one at longwave and one
>     at microwave, but I did not believe that. 

	There could be two frequencies in some theft-detection systems, but
they are probably not very far apart.

>  3. (My favourite theory!) The boxes sends out pulses of 5.5 MHz radio
>     energy. They deliver energy to the resonant circuit. When they stops
>     transmitting, they start to listen. If a resonant circuit is near, 
>     it will send out energy a short period of time after the excitation.
>     The boxes hears that and starts the alarm.

	Not a good idea.  While you could "ping" the detector element with
a short high-energy RF pulse and listen for received damped oscillations,
the receiver would have to be rather sensitive with a rather fast T-R
switch; without getting expensive and complex, such an arrangement could be
unreliable and susceptible to false alarms.

> Does anybody have other theories, or does anybody know how this works?

	The use of a varactor diode to emit a totally different frequency
from that of the excitation frequency would be nice, but it is presently
cost-prohibitive for retail store theft-detection system applications. 
The advantage of such a varactor diode scheme would be virtual elimination
of false alarms since the RF excitation could be pulsed, with the expected
received frequency signal being fed through a coincidence circuit.

<>  Larry Lippman @ Recognition Research Corp., Clarence, New York
<>  UUCP:  {allegra|ames|boulder|decvax|rocksanne|watmath}!sunybcs!kitty!larry
<>  VOICE: 716/688-1231        {hplabs|ihnp4|mtune|seismo|utzoo}!/
<>  FAX:   716/741-9635 {G1,G2,G3 modes}    "Have you hugged your cat today?"