Swipe card controller

[jimmythefool]

Hi
I'm working on a PCB for an automatic gate. I want to make it compatible with swipe/proxy cards. I know the reader sends a data stream to the controller in the building which verifies if the card is valid, if so , where does the signal go from there? Do I get an output from the reader? Or does it activate a No Volt Contact/ Dry switch?
Thanks
Jim

 

[Nigel Goodwin]

Presumably the controller will be a PC, the program running in the PC will verify the swipe card is correct, and then send a signal telling the gate to open - plus it would store the card details and the time of entry. How you actually open the gate depends on how you want to - but I would suggest sending serial data from the PC and using a PIC near the gate to respond to it, and switch a relay that opens the gate.

If you don't want to record any details, you could do it all in a PIC, read the data, verify it, and open the gate.

 

[jimmythefool]

Thanks Nigel,

I have fitted loads of readers, but the connection is normally done by the building's security contractor, I don't get involved. Usually, it goes to a dedicated controller, and sometimes onto a central PC for authorisation. I'm just not sure how it comes back.

 

[saturn1bguy]

I'm working on a PCB for an automatic gate. I want to make it compatible with swipe/proxy cards. I know the reader sends a data stream to the controller in the building which verifies if the card is valid, if so , where does the signal go from there? Do I get an output from the reader? Or does it activate a No Volt Contact/ Dry switch?

Our doorlocks comprise an enclosure containing a local PC-on-a-card running Linux. A Magtek serial card reader (and/or proximity reader) sits at one of the serial ports, and an LCD display on the other. One bit from the parallel port controls an opto-coupler (a 4N39, an opto-SCR), which in turn drives the electric doorknob. (But you could use an electric strike here.) There's also a magnetic reed switch at each door to monitor open/closed state.

The local PC runs the doorlock software (written in C++), records all transactions and grants/denies access locally. Software running on a central server (another Unix box) can download new images of the doorlock code and/or ID card database. Each lock connects to the network via a 100BaseT connection.

So, each lock is self-contained, and needs only mains and a cat5.

 

[jimmythefool]

Thanks Corey,
I was only going to provide a couple of PIC inputs to be controlled by the building. I have seen a lot of systems like this, so I guess the building's security people provide the interface. I know a lot of big companies only allow their own people to make the connections into their system.
Regards
Jim