Hey!
I'm trying to make a RFID access system for a local business (helping a friend). They have people coming in and out of the building and would like a system to display what kind of member the customers are, and if they are a member at all. They have over 150 members, so I'm afraid to run out of space if I save all the members on the code itself.
Another solution I came up with was to save all the RFID-IDs as a txt file on a SD card and include additional information in the txt file as name and membership, and connect it to the arduino. The problem is that the SD-module and the RFID-scanner uses the same pins.
I tried to see if it would work on a raspberry, but it's too advanced for me
You can hack something together using SD or fram, but as @Whandall intimated, it will be painful to keep the data up to date. Far better to have the Arduino talk to the server that has the latest member data.
I have been testing some software using SD cards quite a bit over the last couple of months, mainly on Arduino DUEs, but ATmega1284P as well.
Whilst is the SD\SDfat libraries do work, there are from time to time issues with SD cards just stopping working. Not a significant problem for hobby level projects.
If I were asked on a scale of 1 to 10 whether an Arduino SD card should be used as the core data for a 'professional\commercial' access control system I would vote -10
Whether it is the firmware or hardware, SD cards and Arduinos do not seem to mix. The fatal error rate is FAR too high, so now, I always send important data to be stored elsewhere.
I own a picture frame (some random commercial one) that damages the pictures on the SD card
randomly. I have to rewrite the SD about once a year, when too many pictures are damaged.
This is exactly the sort of system I used to design when I worked for an access control company.
It is not as simple as it sounds. You want the data on the door controller to make a quick decision about access, but you need for it to be updated from a central controlling computer for adding new members and removing them.
We used an RS485 network for this with non volition storage on the door controller itself. The door controller could be about 200M from the reader and door lock.
Even so searching all the possible entries would take too long and the data base search was speeded up by using a hashing algorithm on the entries and searching on the hash to get to the right part of the database.
Each token had a time profile so entry times could be restricted, the token also had a start date and end date attached to it. Also some places required two tokens, and hence people, before a door would open.
It was actually a lot more complex than that, but that is a simple overview.
