In its first report, the RFID Alliance Lab, a not-for-profit testing center based at the University of Kansas, found that there were major differences in the performance of 10 different RFID tags based on the Class 1 and Class 0 Electronic Product Code specifications (see Test Lab Releases Tag Report). Now the Lab’s second report shows that there are significant variations in performance among tags of the same make and model.
Is that significant? You bet it is. Let’s say you purchased tags based on claims made by the vendor or on the recommendation of your systems integrator. Even if you chose a tag with good performance—that is, one that can be read from farther away than others—many of the tags you put on your products will not perform nearly as well as the best tags in the lot or even the “average tag.”
|
 |
That is, no doubt, why companies that have installed readers around dock doors are finding that they are not getting the performance they’d expected. Their system designs aren’t to blame; the problem lies in the differences in the performance of UHF Electronic Product Code tags.
For the new report, An Evaluation of the Performance of UHF EPC Tags, the RFID Alliance Lab evaluated nine commercially available UHF EPC tags under scientifically controlled conditions, to ensure that any difference in performance could be attributed solely to differences in the tags and not to environmental factors. The Lab determined the average performance for each tag. Then, it made nearly 1 million read attempts on more than 100 tags of each make and model to determine the variation from the average. The results are startling. The most consistent tag had a variation in performance of 3.5 dB, which means that the worst performing tags could be read at less than half the distance of the best performing tags of the same model. Other tags had much greater variance.
That’s not the only difference among RFID tags of the same make and model. The new report looks not just at the distance from which tags can be read but also how fast tags can be read, both when there is only one tag in the field and when there are many of the same tag in the read field. While some vendors boast that up to 1,000 of their tags can be read per second, the lab found that few tags performed as well as their marketers would have you believe. With some tags, fewer than 20 could be read per second.
And you’ve no doubt heard that some tags are dead when you receive them and others are “quiet”—that is, they can be read but only from very close range. Of the nine commercially available EPC tags tested by the RFID Alliance Lab, one model could be read 100 percent of the time. Another model was dead or quiet more than 19 percent of the time.
It’s not clear what accounts for the variance among tags. It could be differences in the microchips, the bond between the chip and the antenna or the antenna itself. The Lab has no way to determine what causes the variance, but it has done a great service to end users by quantifying the variance. The only way you can design a system that reads tags 100 percent, or close to 100 percent, of the time is to understand and account for this sweeping variance in performance. And having hard evidence of how many dead and quiet tags you are likely to get when you purchase tags from a vendor enables you to negotiate refunds for tags that don’t work.
RFID Journal helped to get the RFID Alliance Lab off the ground by putting up the funding. We also contribute to the Lab by selling reports through our site, but we don’t make any money from the Lab. All the funds go into creating the next report and buying more advanced testing equipment, if needed. We funded the Lab because we felt that end users needed high-quality information to deploy RFID successfully. And I believe that the Lab is fulfilling that mission in a big way.
When taken together, the first two reports provide companies with the information they need to choose the right tag and design a system that they know will perform reliably. I also think the reports will pressure vendors to improve the performance of their tags. Marketing spin just doesn’t cut it when your customers have scientific data to support their purchasing decisions.
Mark Roberti is the founder and editor of RFID Journal. If you would like to comment on this article, click on the link below.