Introduction to Magnetic Stripe & Other Card Technologies.

presented at,
SCAN-TECH ASIA 97, Singapore, April 24, 1997

When we use the term "card technologies", what do we mean? The easy answer is - any technology that can be placed on a card. What is a card? Typically we think of our credit or bank card but there are other sizes and materials used for different applications. The card can be made of plastic (polyester, pvc, or some other material) or paper, card, or even some amalgamation of materials. The common point is that the card is used to provide "access" to something and it includes some form of ADC/ID (automatic data collection and identification) technology.

There are currently three main technologies we think of when we mention card technologies. These are magnetic stripe, smart cards, and optical cards. Other technologies can be put on cards as well (such as bar codes, touch memory, etc.). Often the card will have printing on it which may involve technologies such as Dye Diffusion Thermal Transfer (D2T2) direct to card printing.

I will cover optical cards and smart cards briefly here and then explain further why magnetic stripe is still a viable solution for many applications of card technology.

Optical memory cards use a technology similar to the one used for music CDs or CD ROMs. A panel of the "gold colored" laser sensitive material is laminated in the card and is used to store the information.

The material is comprised of several layers that react when a laser light is directed at them. The laser burns a tiny hole (2.25 microns in diameter) in the material which can then be sensed by a low power laser during the read cycle. The presence or absence of the burn spot indicates a "one" or a "zero". Because the material is actually burned during the write cycle, the media is a write once read many (WORM) media and the data is non volatile (not lost when power is removed).

The optical card can currently store between 4 and 6.6 MB of data which gives the ability to store graphical images such as photographs, logos, fingerprints, x-rays, etc.. The data is encoded in a linear x-y format and ISO/IEC 11693 and 11694 standards cover the details.

The biggest users of optical technology today are: Medical/Healthcare; Prepaid Debit Cards; Cargo Manifests; Admission Pass Season Tickets; Auto Maintenance records; and Retail Purchase Cards.

Smart cards are credit card-sized plastic cards that contain relatively large amounts of information in an imbedded micro-chip. Smart cards differ from magnetic stripe cards in two ways: the amount of information that can be stored is much greater, and some smart cards can be reprogrammed to add, delete or rearrange data.

There are several terms used to identify cards with integrated circuits embedded in them. The terms "chip card," "integrated circuit card", and "smart card" really all refer to the same thing.

There are two types of smart card. The first is really a "dumb" card in that it only contains memory. These cards are used to store information. Examples of this might include stored value cards where the memory stores a dollar value which the user can spend in a variety of transactions. Examples might be pay phone, retail, or vending machines. Another example of a "dumb" card is the memory that is plugged into a Personal Computer (PC Card - used to be called PCMCIA).

The second type of card is a true "smart" card where a microprocessor is embedded in the card along with memory. Now the card actually has the ability to make decisions about the data stored on the card. The card is not dependent on the unit it is plugged in to to make the application work. A smart purse or multi-use card is possible with this technology.

Smart cards are the technology of choice when fairly large databases must travel with an individual or an object. For instance, a version of smart card technology is used to record service histories for automobiles. The data travels on a small tag on the owner’s key ring. It can be reprogrammed, updated and accessed whenever the vehicle is serviced with any of that company’s dealers.

As there is a microprocessor on the card, various methods can be used to prevent access to the information on the card to provide a secure environment. This security has been touted as the main reason that smart cards will replace other card technologies.

The microprocessor type smart card comes in two flavors - the contact version and the contactless version. Both types of card have the microprocessor embedded in the card however the contactless version does not have the gold plated contacts visible on the card. The contactless card uses a technology to pass data between the card and the reader without any physical contact being made. The advantage to this contactless system is there are no contacts to wear out, no chance of an electric shock coming through the contacts and destroying the integrated circuit, and the knowledge that the components are completely embedded in the plastic with no external connections. The disadvantage to this is that there are some limitations to the use of the smart card.

Smart cards are not new, the first patent was filed in France in 1974 and the first cards were used in France in 1982. The technology was rapidly accepted in Europe because the high cost of telecommunications made on-line verification of transactions very expensive. The smart card provided the mechanism to move that verification off line, reducing the cost without sacrificing any of the security. In the United States, telecommunication costs have always been low compared to other countries. This meant that the impetus to implement smart cards has taken longer to reach the momentum needed.

The possible benefits of the acceptance of smart card technology depend on the application in use. However, the ability to move large amounts of data with little or no increase in the security of the data will lead to many new applications being created that we haven’t even begun to think about.

There are many smart cards in use today throughout the world. In 1993 approximately 330 million cards were produced by the major manufacturers. Of this number only about 12% were true "smart cards", the rest were simple memory cards. This was projected to grow to approximately 580 million cards in 1995 (about 10% being "smart") and 990 million in 1996 (approx. 10% "smart"). Of the cards issued in 1993 approx. 260 million were used in phone systems; 25 million in health applications; and 23 million in banking. The rest were used in various small projects and trials.

The smart card future is extremely bright. Many changes are happening in the electronics world today that will increase the capabilities of the technology. As the state of the art in manufacturing integrated circuits improves, we get smaller ICs which run on lower voltages, giving us less power requirements and the ability to include more memory of processing power. We also need to see an increase in the speed that a card can be addressed. Currently the initialization of a smart card can take several seconds and even a single transaction may take longer than is tolerable under some circumstances.

Magnetic stripe technology is everywhere. We use cards with magnetic stripes on them everyday without even thinking about it. The technology has been with us for many years, but there are still many new things going on in the industry.

The first use of magnetic stripes on cards was in the early 1960’s. London Transit Authority installed a magnetic stripe system in the London Underground (UK). By the late 1960’s BART (Bay Area Rapid Transit) (USA) had installed a paper based ticket the same size as the credit cards we use today. This system used a stored value on the magnetic stripe which was read and rewritten every time the card was used.

Credit cards were first issued in 1951, but it wasn’t until the establishment of standards in 1970 that the magnetic stripe became a factor in the use of the cards. Today financial cards all follow the ISO standards to ensure read reliability world wide and along with transit cards constitute the largest users of magnetic stripe cards.

With the advent of new technologies many people have predicted the demise of the magnetic stripe. However, with the investment in the current infrastructure this is not likely to be any time soon. Magnetic stripe technology provides the ideal solution to many aspects of our life. It is very inexpensive and readily adaptable to many functions. The standardization of high coercivity for the financial markets has provided the industry with a new lease on life. This coupled with the advent of the security techniques now available means that many applications can expect to be using magnetic stripe technology for the next ten to twenty years.

A magnetic stripe is the black or brown stripe that you see on your credit card, or maybe the back of your airline ticket or transit card. The stripe is made up of tiny magnetic particles in a resin. The particles are either applied directly to the card or made into a stripe on a plastic backing which is applied to the card.

The material used to make the particles defines the Coercivity (see below) of the stripe. Standard low coercivity stripes use iron oxide as the material to make the particles, high coercivity stripes are made from other materials like barium ferrite. These materials are mixed with a resin to form a uniform slurry which is then coated onto a substrate. In the case of a credit card or similar application the slurry is usually coated onto a wide plastic sheet and dried. The coating is very thin and the plastic allows the coating to be handled. It is then sliced into stripe widths and applied to the card during the card manufacturing process. The methods of application include lamination (where the stripe and backing is laminated into the card), hot-stamp (where a heated die is used to transfer the oxide stripe from the backing onto the card after the card is cut to size), and cold-peel (where the oxide stripe is peeled from the backing, and then laminated into the card). Each of the methods have their own advantages and are largely irrelevant to the user of the card.

Another method of putting a stripe on a card is direct coating. In this case, the oxide slurry is coated onto the card (usually paper or card rather than plastic) during the manufacturing process for the card. There can be some manufacturing cost reductions by using this technique, though there may also be some quality trade off.

Once the slurry is coated onto the substrate (plastic backing or direct to card stock) the particles in the slurry are aligned to give a good signal to noise ratio. This is the equivalent of eliminating those pops and bangs you hear on old tape recordings. The tape with the wet slurry is passed through a magnetic field to align all the particles. With the iron oxide particles this is relatively easy for two reasons. The particles are low coercivity so do not need a large magnetic field to orient them, and the particles are acicular (needle shaped) with an aspect ratio of approximately six to one. The acicular particles have an easy axis of magnetization along the length of the particle which makes the alignment an easy process. This process is not so easy with the high coercivity materials. The particles used in most of the high coercivity materials are not acicular, they are platelets. These platelets have an easy axis of magnetization through the plate, which means the alignment field has to stand the particles on edge and they have to stay that way to get the best performance from the stripe. Obviously the particles want to fall over as soon as the field is removed from the stripe so part of the skill in making a high quality stripe lies in designing a process that can keep those particles on their side until the slurry sets.

Unfortunately, the lack of alignment can cause some major problems in the read and encode process of the magnetic stripe. The waveshape of the read process can be distorted by the lack of alignment. This distortion can cause significant problems for some read systems.

In all of the above processes, the final card has the familiar brown or black stripe on it. The stripe can be encoded because the particles (like iron filings) can be magnetized in either a north or south pole direction. By changing the direction of the encoding along the length of the stripe this allows information to be written on the stripe. This information can be read back and then changed if required as easily as the first encoding.

The end-user defines the requirements for the magnetic stripe including the signal amplitude expected, the coercivity of the stripe, the encoding method and the bit density. The card manufacturer uses the first two points to select the type of magnetic material to use. The system designer is concerned with all four of the parameters.

As explained above, the stripe is made from many small particles bound together in a resin. The density of the particles in the resin is one of the controlling factors for the signal amplitude. The more particles there are, the higher the signal amplitude. The density (or loading) combined with the thickness give a method for controlling the amplitude. Signal amplitude is important because it defines the design of the readers for the cards. Standards exist (ISO/IEC 7811) which define the signal amplitude for cards that are used in the interchange environment (such as banking). By conforming to these standards, a user ensures that the magnetic stripe can be read in any financial terminal world wide.

The bit density of the information is selected based on the user requirement. The ISO/IEC standards (7811) give requirements for bit density for cards used in the interchange environment. These standards define tracks one and three as 210 bits per inch and track two as 75 bits per inch. The bit density in conjunction with the data format (see below) dictate how much data is encoded on each track.

Each character that is encoded on the stripe is made of a number of bits. The polarity of the magnetic particles in the stripe are changed to define each bit. Several schemes exist to determine whether each bit is a one or a zero, the most commonly used schemes are F2F (or Aiken BiPhase) and MFM (Modified Frequency Modulation).

The ISO/IEC 7811 standards specify F2F encoding. In this encoding, each bit has the same physical length on the stripe. The presence or absence of a polarity change in the middle of the bit dictates whether it is a one or a zero. The width of a single bit always remains the same but some bits have an extra polarity change in the middle and these are called ones.

MFM encoding is more complicated. This type of encoding allows twice as much data to be encoded with the same number of flux reversals (edges). For more details on MFM the reader is referred to the AIM Inc. publication "Modified Frequency Modulation (MFM) for Magnetic Stripes" available on the AIM Inc. World Wide Web site.

The choice of encoding scheme is determined by the application and the user. If the application is one where conformance with ISO/IEC 7811 is necessary then F2F encoding is the choice. For applications where large amounts of data must be encoded, MFM may be a more suitable choice.

Once the encodation scheme is chosen, the format of the data must be selected. ISO/IEC 7811 specifies two different schemes for use on interchange cards. These are four bits plus parity and six bits plus parity. The four bits allow only the encoding of numbers plus some control characters, the use of six bits allows the full alpha numeric set to be encoded. The parity bit is used to help determine if an error occurred in the reading of the data. The total number of "one" bits in a character is added up, in odd parity this must equal an odd number. If the total is odd, the parity bit is set to a zero, if the total is even the parity bit is set to a one.

Although the encodation schemes are defined in ISO/IEC 7811, it is only necessary to follow them if the application requires conformance with 7811. Some applications depart from this scheme by allowing different bit density/encoding scheme combinations, others depart significantly by using "proprietary" schemes down to the bit level. As an example, an identification card may use two bits to determine eye color (00 = blue, 01 = brown, 10 = green, 11 = other). This is much more efficient in encoding space, but means the data cannot be read in a standard interchange terminal. For some applications this is not important and the extra space available is very important.

Measured in Oersteds, coercivity is the measure of how difficult it is to encode information on the magnetic stripe. A standard bank card has a coercivity of approximately 300 Oe (Oersteds) and is considered to be low coercivity. In Japan there is a second stripe on the credit cards with a coercivity of 600 Oe. The trend is to move towards higher coercivity with values of 2100, 2750, 3600 and 4000 Oersteds being common. High coercivity magnetic stripes bring a new collection of parameters to the magnetic stripe world and higher is not always better.

Initial coercivity is defined by the type of particles used to manufacture the stripe. Gamma Ferric Oxide will give you a low coercivity stripe, Barium Ferrite will give you a high coercivity stripe. The material alone does not define the final coercivity of the stripe as the manufacturing process will change the value usually in the downwards direction. It is possible to raise the coercivity of particles by including other agents in the slurry.

Coercivity is NOT a measure of signal amplitude. Early versions of high coercivity stripes often had high signal output. This is not a requirement of high coercivity and is not usually a good thing. Most readers available today are setup to read signal levels similar to those defined in the ISO/IEC 7811 standard. Keeping the signal output in this range makes the range of available readers much greater.

Early versions of the high coercivity magnetic stripe were marketed with the name High Energy. This name suggests high output levels and often causes confusion amongst users of the technology.

The advantage of high coercivity is that it is harder to encode the information on the stripe. This also means that the it is more difficult to erase the information and so problems of accidental erasure are diminished. It is still possible to erase the information, but common household magnets are not usually powerful enough. This means the person who puts the transit card on the refrigerator will not usually damage the encoding on the stripe.

The disadvantage is that although the encoding can be read in a standard low coercivity reader the encoder must be designed to encode high coercivity stripes.

Although the coercivity is a factor in erasing a stripe, it is by no-means the only factor. When a stripe is declared to be a 4000 Oersted (Oe) stripe, it really means that the nominal value is 4000 Oe. There are also lots of particles in that stripe with coercivities of other values. The distribution of the coercivities will typically follow a bell shape curve. The steepness of the bell shape defines the percentage of particles at the stated value, a sharp (steep) curve shows that are a large percentage are the nominal value. A flat curve shows that there are many other coercivities present in the stripe. This is important because it is used to define something called "squareness" of the stripe.

Squareness is a parameter that can be used to help define the susceptibility of a stripe to erasure. A 2700 Oe magnetic stripe with high squareness (sharp curve) has a large number of particles at the nominal coercivity. To erase that stripe, a magnetic force greater than the coercive value will have to be applied to the stripe. Another stripe with low squareness may have a higher nominal coercivity but because there may be a large proportion of low coercivity particles it may be very easy to erase the stripe.

Everyone uses magnetic stripes. The most visible use is your bank (credit, debit, and ATM) cards, but these are not the only places. Take a look at your Airline Ticket and Boarding pass (ATB) the next time you travel. Many of these are now including magnetic stripes on the cards. Other places include your phone card, your transit (bus or train) ticket, and even your parking lot ticket.

Magnetic stripes are not all the same. On the outside they are all made of a magnetic material coated in some way on the document. However, as explained above, there are different ways to coat the material on the document and different ways to make the magnetic material. These all affect the performance of the material in some way.

The properties of the magnetic stripe are all defined during the manufacturing process. These properties define the signal strength of the encoding, the coercivity of the stripe, the ability to resist erasure, even the waveshape of the recording. These parameters are not controlled by the user but they can have a tremendous effect on the performance of the system and should be defined by the user.

Even the method of coating the magnetic material on the document can influence the performance of the stripe. A direct coating on a paper ticket may produce a stripe that is much more abrasive than the stripe on a laminated plastic card. This abrasiveness will affect the life of the magnetic heads being used.

Some magnetic stripes have coatings over the stripe to protect the stripe from abrasion thus prolonging the life of the stripe on the card or ticket. This coating may affect the performance of the stripe in other ways.

Yes there are. The most commonly quoted standards are the ISO/IEC 7810, 11, 12 and 13 series of standards. These standards are written for the credit and debit card market and so include information on the embossed characters on the cards as well as the track locations and information on the magnetic stripe. ISO/IEC 7811 has six parts with parts two and six specifically about low and high coercivity magnetic stripes. These standards include information on the magnetic properties that guarantee that the stripe can be read in a magnetic stripe reader in the U.S.A. as well as in Japan. The companion to the ISO/IEC 7811 series of standard is ISO/IEC 10 373. This document details the test methods for the ISO/IEC 7811 series of standards.

AIM Inc.

Work is just about to start on three new American National Standards (ANSI) standards that relate to magnetic stripe performance. These are:

• Effective Magnetic Parameters of Magnetic Stripes

• Suggested Magnetic Parameter Values for Applications

• Magnetic Stripe Readers and Encoders - Equipment Specifications

The first two of these new standards are related to the AIM Inc. published document listed above, turning it into an ANSI standard. The third item is work that is new in the magnetic stripe world in that the goal is to create the first standards that are relevant to the equipment manufacturers. Details are available from the AIM Inc. office.

If you are not intending to use your cards in the banking system then you can do anything you want. The ISO/IEC 7811 series of standards define track one as a read only track with 210 bits per inch and 6 bits plus a parity bit per character. This allows for a full alpha-numeric encoding. Track two and three both use four bits plus a parity bit (number characters plus A to F) only, with track two at 75 bits per inch and track three at 210 bits per inch. If you don’t have cards that have to be read in the banking system then you can use any encoding scheme and bit density on any track you wish. In fact this gives you some added security, as it makes it more difficult for someone to copy your cards.

Magnetic stripes are not inherently secure. The problem with being easy to manufacture and encode is that it also makes it easy for the crooks to do the same. Several schemes are available for creating a secure encoding on a magnetic stripe, Watermark Magnetics, XSec, Holomagnetics, XiShield, Jitter Enhancement, ValuGard, and MagnePrint are a few. The contacts for some of these technologies are listed below. Each of these technologies exploits some aspect of the magnetic stripe, the card, and the data on the stripe to tie everything together to make counterfeiting the card in some fashion very difficult.

• WatermarkMagnetics http://www.tssi.co.uk/Products_x_Services/Document_Security/watermark.html

• XiShield - http://www.xico.com/index.html

• MagnePrint  - www.magneprint.com

The security schemes all work in basically the same way. They focus on some part of the card/magnetic stripe/encoding and record the information that makes it different from any other card. This could be the noise in the magnetic stripe, an intentional permanent signature in or on the stripe, or some external feature on the card that is permanent.

The advantage to using one of these techniques is that the card and data become tied together making the duplication of the data very difficult. The disadvantage to these techniques is that they cost money and are for the most part, proprietary. Several of the techniques have been used in large applications where the system demanded some form of extra security.

This is a rumor that started during the mid 1980's at a time when eel skin wallets had become very fashionable. The most common way of providing a clasp on these wallets was to use a magnet. This magnet was usually powerful enough to erase a magnetic stripe if the two came into contact. The popular press picked up the problem and very quickly the rumor that the eel skin was capable of damaging the magnetic information was spread. In fact the eel skin is no different from any other kind of leather and was not the problem, the magnet was the sole cause of the problems.

This is a complicated question to answer that can only be properly answered after the card has been analyzed by some test equipment. The likely problems are dirty or scratched stripe, or erased stripe. The stripe on a card is not delicate but a few simple measures will increase the life of the stripe. Try to keep the card in a clean place when you are not using it. A gritty wallet, kept in the back pocket of a pair of pants, will probably end up scratching the stripe (and probably warping the card). A scratched or dirty card will eventually not work.

Keep the card away from magnets. The two most likely examples of magnets we see are the refrigerator magnet and the Electronic Article Surveillance (EAS) Tag demagnetizer in a store (this is the box that some stores have on the check out counter that they pass a book or clothes over so that you do not set the alarms off when you leave the store).