13.56MHz shielding re-radation nightmares

[msinger]

I am working on an RF based access control system. Small 13.56MHz transmitters are detected by a receiver/controller to grant or deny access to a doorway.

The problem is the 13.56MHz signal couples easily to metal surfaces and other wiring and re-radiates near the receiver triggering many false alarms. Site surveys and location of receivers can solve many re-radiation problems but system interconnect wiring also conducts this signal causing false alarms.

I have tried shielded wiring only to find the signal travels well on the shield also. I have grounded the shield to earth ground, circuit ground, single ended and on both ends to no avail. The only thing that has worked so far is to install ferrites on the offending wiring, at certain distances from the reciever. As the ferrites are a fairly expensinve solution I am fishing for any tips or alternatives to what I have tried so far.


Thanks,
Mike S

 

[nbucska]

I would suggest more complex digital code -- instead of relying on the attenuation of the carrier
Or use IR instead of HF

<nbucska@pcperipherals.com>

 

[VEBill]

Are the transmitters running all the time? If so, why not add a 'push to open door' button?

HF is a bad choice anyway - try tuning a SW radio to that frequency (including plus or minus your bandwidth) and see how much other junk is there.

 

[msinger]

The committment to RF has already been made, but we have thought of opto-isolation (IR) to bring the offending wires down to the controller (it is an expensive option). Transmitters output a 13.56MHz on/off keyed (data is the carrier) ID signal with a random delay. The purpose of the system is to prevent Alzhiemers patients from exiting the care facility and getting into harms way while alowing normal access by staff and visitors. A push button is used on the outside, but using one on the other side would defeat the purpose of the system. We seem to have good noise immunity, it is the re-radiation of our own transmitter signals that fakes the receiver into believing a patient (transmitter) is near a monitored door.

 

[VEBill]

No, no. The suggestion about the push button relates to the immediately previous question about the transmitters running all the time. The description of your problem seems to indicate that these little transmitter are running all the time (continuous duty cycle), and you're trying to detect if they are close to the door or not. The coupling into the power lines and so forth is defeating the entire plan (not surprising).

The suggestion is to put a button on the transmitters so that if someone wants the door open then they can push a button on the remote control. At that point it doesn't matter how the signals get coupled in - so long as it is reliable.

You've set yourself up a difficult problem if you're trying to create a 'proximity detection system' in the HF bands. If I knew your digital code, then I could open your door from halfway around the world on a good day.

The only solution that I can imagine is to try to use multiple drectional antennas (ferrite loopsticks) and &quot;AND&quot; the data so that someone must be in a particular location to trigger the door. Even that wouldn't be perfectly reliable (probably not even acceptable)

It would be cheaper to purchase a commercial 'access control' system.

 

[nbucska]

Or this ID chips they use for cows, pets, kids, etc.

<nbucska@pcperipherals.com>

 

[hacksaw]

the good thing about 13.56 Mhz is that is the same freq used by RF Diathermy, RF Induction heaters, and other industrial users. of all the freq's you could have picked....

to keep the RF from crawling all over your wires, and shields, you may have to resort to twin ax (shield (balanced) twinline with a braided shield(96% coverage). Additionally you may have to use external capacitive chokes. These work as &quot;band&quot; stop filters for any external rf sneaking down the lines.

Baluns may be in order if you only have &quot;single-ended&quot; outputs from your sources...