Chapter 8: Testing One, Two, Three: Developing Your Own Lab
6. Test several of the tags, one at a time, that were of the highest attenua- tion value and see how far they can be read and then repeat the
The tag characterization test resolves one of the biggest issues you’re likely to face if you use an automated print-and-apply solution: the lack of standard- ized quality from tag vendors. For example, a print-and-apply machine might be able to read a particular tag if it’s a few inches away from the antenna, but might not be able to read it from a foot away or at a high speed because the chip is poorly mounted to the antenna.
The only way to tell the good tags from the bad tags is to characterize the quality of various tags and understand the lowest performance threshold that is acceptable given your mandates. In other words, if you know a tag attenuated to a certain point can’t be read beyond a foot, you want to set up a verification station after your print-and-apply station (or work with the print-and-apply vendor) attenuated down to the lowest threshold acceptable.
By using the tag characterization data to set up a verification station, you can stop bad tags from being shipped out to your clients and not getting read. If all your tags are read, you get paid quicker and with less hassle by folks like the Department of Defense (DoD), Wal-Mart, Target, and others.
Additional tests
The four tests that I describe in this chapter give you great insight into the performance and characteristics of RFID systems. You can add more tests as you put your creativity to work and as you graduate up the learning curve.
Eventually, you will develop tests that are unique to your application or pro- duction environment. If you’re tracking top-secret hard drives covertly, for example, you may want to test how antennas perform when they are behind sheetrock or how tags perform at different orientations. If you’re producing roller skates, you may end up slicing up a standard tag and experimenting with ways of affixing the tag onto the boot. Creativity and repeatability are the keys to good experiments.
Chapter 9
Tag, You’re It: Testing for Best Tag Design and Placement
In This Chapter
Examining the material properties of your products Determining the optimal spot for testing
Carrying out application testing Testing with physics
Deciding how to encode and apply tags Ensuring maximum read success
W.C. Fields’ constant search for the 25-cent cigar is second only to Linda Dillman’s search for a 5-cent RFID tag. Dillman, as the CIO of Wal-Mart, has been hounded by suppliers implementing RFID who claim that the only positive ROI comes with a 5-cent tag. Although this might be true for companies that count their product margins in fractions of pennies, many current RFID implementations demonstrate acceptable ROI using today’s tags, which range in price from 22 to 50 cents each. The trick is knowing the difference between tags and how they interact in your inventory environment so that — regardless of price — you can make the right decision for your organization.
Think of choosing the best tags for an RFID system as picking the right bat to play a game of baseball. (Without a bat, there’s just not much point in play- ing, but you can’t just go out and grab any ol’ hunk of wood.) You need to evaluate the bat’s attributes, try it out in batting practice, and see how it fits with your swing if you want to hit ’em over the fence. The same is true for tags — you have to test them in their working environment to get successful results from your RFID system.
The ideal RFID system automates any counting function such as shipping, receiving, picking products for shipment, asset management, security, and so on. It does this by capturing data without line of sight.Although RFID promises to achieve this automated, non-line-of-sight communication, in many cases
this communication succeeds only through careful tuning of key system para- meters, including tag selection and placement, reader selection and configura- tion, interrogation antenna selection, and a thorough understanding of the environment. This chapter deals with the tag-selection-and-placementparame- ter, specifically with testing tags on difficult-to-read items.
In this chapter, you find out how the right tags help you get better read ranges and rates and improve the efficiency of your RFID network. You also discover why different products make tags behave in unique ways and the implications of these unusual behaviors as you create your system architec- ture. A little bit of understanding of tags goes a long way, whether you do the work yourself or enlist a third party to do it for you.
Ready, Set, Test!
The biggest error many people make when setting off on an RFID program is thinking that a single solution will work well for every situation. Unfortunately, there is no silver bullet. RFID systems are always custom designed for both the environment and the properties of the item to be tagged. Getting the right tags for various products is a critical and ongoing process; you may find that the optimal tags and their placement change as you change inventory and as the technology evolves and prices change. The first step to finding the right solution is to test the items that you will be tagging (or to have someone test them for you). A good assortment of tags to test and a rigorous methodology get you the answers you need to build a good RFID network. The two primary ways of testing for proper tags and placement are
Application testing(also known as trial and error) using a conveyor or dock door.
Scientific static testingto evaluate the way a product is affected in an RF field.
A solid testing procedure, as in any data-gathering exercise, is required.
Figure 9-1 shows the three critical steps for incorporating tag testing into your organization and the factors influencing those different steps:
Product assessment: Determines the optimal areas to place a tag to be tested, so you don’t need to cover every inch of every product with each different tag. You can base this assessment on a tester’s experience using a method like the RF pyramid I show later in the chapter, or you can use an automated RF Visualizer to show a product’s various RF-friendly areas.
This step tells you where the best place to test is — essentially it helps you find the sweet spot.
Product testing: After determining the five or six sweet spots on the product that look optimal for testing tags, you need to choose the tags you plan to test, document the test locations, conduct the tests, and
then determine the best performing tag. For purposes of comparison, you should record the performance measures specified in a universal and scientific manner such as using the ODIN tag performance index (or TPI) for each tag, location, and orientation. See “The ODIN tag perfor- mance index” sidebar later in this chapter. In order to test tags effec- tively, you need an RF-friendly location without ambient electronic noise, testing hardware, and (to make life easy) testing software that produces efficient and easy-to-understand results across the entire fre- quency band.
Employee training: Incorporates tag testing into your normal inventory processes by making sure that you have a system in place and docu- mented procedures that can be taught to end-users simply and con- cisely. Manufacturing employees or packaging workers are the likely recipients of your product-testing training, so keep their needs in mind when designing your program. And create a user’s manual for your test- ing process that appropriately addresses this intended audience.
Chapter 13 discusses user training in more detail.
Tag testing must isolate the performance of the tags for a given orientation (vertical, horizontal, random) and location on the case (top, back, side). Tag testing should not be a test of reader performance, antennas, or any other variable. The tag is the only thing that should vary during the testing. Keep this in mind when designing your testing methodology because the software on readers can vary results greatly, and some automated methods don’t test beyond one or two channels in the ISM UHF band. (There are 124 channels, and you need to know the performance across the whole band.)
Recently, my company demonstrated the use of Trifecta for a client. Trifecta is an ODIN technologies software tool designed to simply and accurately pro- vide a tag performance index so tags can be compared universally across the frequency band. I tested one of the client’s difficult-to-read SKUs, which was
1 2 3
Product Assessment
Product Testing
Employee Training
Experience
Automation Software Application
Facilities
Hardware Manuals
Figure 9-1:
The tag testing process.
161