Re: How do memory flash cards store their information?

Hi Richard! No, it's not magnetism. Instead it's "electri-cism," and the data bits are stored as charges on microscopic capacitors. A capacitor is two metal plates separated by an insulator, and is used to store electrical energy in the form of separated opposite electric charges.

A similar question was answered here earlier:

How do EPROMS/EEPROMS/Flash Memory store code without batteries
http://www.madsci.org/posts/archives/mar2000/953088081.Cs.r.html

Flash memory is one member of a broader type of computer memory called "Electrically Programmable Read Only Memory", or EPROM for short. Flash memory actually falls under the category "Electrically Erasable Programmable Read Only Memory, or EEPROM.

EEPROM memory relies on the fact that one type of transistor can be turned on and off purely by voltage, with no electric current needed. Capacitors can store a voltage for a very long time, and if a charged capacitor is connected to the input of this transistor, the transistor switches on. If there are no paths for leakage, the capacitor can remain charged many tens of years. The transistor behaves as if it's turned on permanently, but it can be turned off again by discharging the capacitor. EEPROM is composed of millions of these transistors, each with its own capacitor connected to the input. Each transistor/capacitor stores one bit of data.

To "charge" any capacitor with energy, we never inject any electric charges into it. Instead we take charges from one plate and place them on the other plate. This leaves un-cancelled charges behind on the first plate, while also creating excess charges on the other. One plate becomes (+) and the other one is (-). In most electric circuits the charge is removed from the first plate by a wire, and then is driven through another wire to the second capacitor plate. However, in EEPROM capacitors things are different. If electric circuits were used to switch the path for charges, these circuits would create a path for leakage even when they were turned off. This would discharge the capacitor in a matter of hours or days. To solve this problem, EEPROM capacitors don't use switching circuits to charge and discharge their energy. Instead an exotic type of insulator is used between the capacitor plates. This insulator becomes a conductor when the voltage is high. To charge the capacitor, the circuit connects a high voltage across the capacitor, the capacitor's insulator becomes conductive, and charges are driven from one plate to the other. When the high voltage is switched off, the charges remain in place for many years. To "erase" the stored bit, the opposite polarity of high voltage is used.

All data in an EEPROM has a limited life. Over many years the little capacitors will slowly discharge, and the bits will start erasing themselves spontaneously. Capacitors aren't the only thing that decay over time; the tiny magnetized spots on hard drives and floppy disks aren't much better in the long run. Also, these types of capacitor-memory can only be erased and re-written a limited number of times. Each time the charges are forced to flow through the capacitor's insulator, some charges are trapped there, and this degrades the insulator. Most modern EEPROMs are only guaranteed for about 10,000 re-writes, so they cannot be used to store rapidly changing data.

Here are some other articles describing "flash" memory:

http://www.simpletech.com/flash/intro.html
http://www.toshiba.com/taec/main/faq/flash_faq.html#flash
http://developer.intel.com/design/flash/articles/what.htm
http://www.zdnet.com/products/stories/reviews/0,4161,2441113,00.html