A Samsung Electronics employee demonstrates a 32-Gigabyte NAND memory card and chip. Explore more images of computer memory.
Chung Sung-Jun/Getty ImagesWe store and transfer various file types on our computers, such as photos, music, documents, PDFs, and more. However, sometimes your computer's hard drive isn't the ideal location for your files. Whether you're looking to back up important documents or ensure the security of your data, portable storage devices using a technology called flash memory might be the perfect solution.
Electronic memory comes in various forms, each designed for specific tasks. Flash memory is popular for quick and efficient data storage in devices like computers, digital cameras, and video game consoles. It functions similarly to a hard drive rather than RAM. Notably, flash memory is a solid-state storage device, meaning it has no moving parts and operates electronically rather than mechanically.
Here are a few examples of flash memory devices:
- The BIOS chip in your computer
- CompactFlash (commonly used in digital cameras)
- SmartMedia (frequently found in digital cameras)
- Memory Stick (primarily used in digital cameras)
- PCMCIA Type I and Type II memory cards (utilized as solid-state disks in laptops)
- Memory cards for gaming consoles
Flash memory is a form of EEPROM chip, which stands for Electronically Erasable Programmable Read-Only Memory. It consists of a grid of rows and columns, with each intersection housing a cell that contains two transistors (as seen in the image below).
The two transistors are separated by a thin layer of oxide. One transistor is called the floating gate, while the other is the control gate. The floating gate is connected to the row or wordline only through the control gate. As long as this connection is intact, the cell retains a value of 1. To change this to a 0 requires a special process known as Fowler-Nordheim tunneling.
In this article, we’ll explore how Flash memory works, delve into the various forms it takes, and examine the types of devices that use it. We’ll also discuss tunneling in more detail next.
Flash Memory: Tunneling and Erasing
Tunneling is the process used to move electrons within the floating gate. A charge, typically between 10 and 13 volts, is applied to the floating gate. The charge travels from the column, or bitline, into the floating gate and is then discharged to a ground.
This charge makes the floating-gate transistor function like an electron gun. The energized electrons are pushed through and captured on the other side of the thin oxide layer, which gives them a negative charge. These negatively charged electrons act as a barrier between the control gate and the floating gate. A device known as a cell sensor monitors the charge level passing through the floating gate. If the flow exceeds the 50 percent threshold, it indicates a value of 1. If the charge falls below this threshold, the value changes to 0. In a blank EEPROM, all gates are fully open, and each cell is set to 1.
The electrons in the cells of a flash memory chip can be restored to their normal state ("1") by applying a higher-voltage electric field. Flash memory uses in-circuit wiring to apply this electric field either to the whole chip or to designated sections called blocks. This process erases the selected area of the chip, which can then be rewritten. Flash memory operates much faster than traditional EEPROMs, as it erases not just a single byte at a time, but entire blocks or the entire chip, then rewrites it.
You might think that your car radio uses flash memory because it stores your preset stations. However, it actually uses flash RAM. The key difference is that flash RAM requires some power to retain its data, whereas flash memory retains its information without any external power source. Even when the power is off, the radio consumes a small amount of current to maintain the data in flash RAM. This is why the presets are lost if your car's battery dies or the wires are disconnected.
- How Computer Memory Works
- How RAM Works
- How Virtual Memory Works
- How ROM Works
- How Removable Storage Works
- How BIOS Works
Removable Flash Storage Cards
SmartMedia cardThere are several reasons to choose flash memory over a hard disk:
- It has no moving components, making it completely silent.
- It offers quicker access times.
- It is more compact and lightweight.
So, why don't we just use flash memory for everything? The main reason is that hard disks are significantly cheaper per megabyte, and they offer much higher storage capacities.
The solid-state floppy-disk card (SSFDC), commonly known as SmartMedia, was first developed by Toshiba. SmartMedia cards range in capacity from 2 MB to 128 MB. The card is quite compact, measuring approximately 45 mm in length, 37 mm in width, and less than 1 mm in thickness.
As shown below, SmartMedia cards are incredibly simple in design. A plane electrode connects to the flash-memory chip through bonding wires. The flash-memory chip, electrode, and wires are embedded in a resin using a method called over-molded thin package (OMTP). This technique integrates all components into a single package without requiring soldering.

The OMTP module is attached to a base card, forming the actual SmartMedia card. The electrode transfers power and data to the flash-memory chip when the card is inserted into a device. A notched corner indicates the power specifications: if the notch is on the left, the card requires 5 volts; if it's on the right, it needs 3.3 volts.
SmartMedia cards erase, write, and read memory in small chunks, typically 256 or 512 bytes at a time. This method ensures they perform quickly and reliably, while giving you control over the data you wish to retain. However, they are less durable than other types of removable solid-state storage, so caution is needed when handling or storing them. Due to the rise of smaller cards with larger storage capacities, like xD-Picture Cards and Secure Digital cards, Toshiba has largely ceased producing SmartMedia cards, making them increasingly rare.
CompactFlash cards, introduced by Sandisk in 1994, differ from SmartMedia cards in two key aspects:
- They are bulkier.
- They feature a dedicated controller chip.
CompactFlash cards consist of a small circuit board housing flash-memory chips and a dedicated controller chip, all encased in a durable shell that is thicker compared to a SmartMedia card. CompactFlash cards measure 43 mm in width and 36 mm in length, and come in two thicknesses: Type I cards are 3.3 mm thick, while Type II cards measure 5.5 mm thick.
CompactFlash cardCompactFlash cards support dual voltage, operating at either 3.3 volts or 5 volts.
The added thickness of CompactFlash cards allows them to hold more data than SmartMedia cards. Their storage capacities range from 8 MB up to 100 GB. The onboard controller chip can boost performance, especially in devices with slower processors. However, compared to SmartMedia cards, the larger size, extra weight, and added complexity come with CompactFlash cards.
Standards for Flash Memory
Samsung Electronics reveals the world’s first 30-nanometer 64-gigabit Flash memory device.
Kim Jae-Hwan/AFP/Getty ImagesBoth SmartMedia and CompactFlash, along with PCMCIA Type I and Type II memory cards, follow standards set by the Personal Computer Memory Card International Association (PCMCIA). These standards make it easy to use CompactFlash and SmartMedia products across various devices. Additionally, adapters are available that allow you to access these cards through standard floppy drives, USB ports, or PCMCIA card slots (found in certain laptops). For instance, games for Sony's original PlayStation and PlayStation 2 are compatible with the PlayStation 3, though the newer console lacks a slot for older system memory cards. Gamers wishing to transfer their saved data need to buy an adapter. Sony's Memory Stick is widely available in a variety of Sony products and is increasingly being adopted by other manufacturers too.
While many standards are taking hold, there are still many flash-memory products that are entirely proprietary, such as those used in some gaming systems. However, as electronic components become more interchangeable and able to communicate (thanks to technologies like Bluetooth), standardized removable memory will ensure your data is always within reach.
In September 2006, Samsung introduced PRAM – Phase-change Random Access Memory. This innovative memory type combines the high-speed processing capabilities of RAM with the non-volatile qualities of flash memory, earning it the nickname “Perfect RAM.” PRAM is designed to be 30 times faster than traditional flash memory, with a lifespan 10 times longer. Samsung planned to release the first commercial PRAM chips in 2010, offering 512 MB of capacity [source: Numonyx]. These chips were expected to be used in mobile phones and other devices, possibly even replacing flash memory in the future.
