Corrupt User memory

[Ray]

Hi we are utilizing RFID on one our production lines and use approx 25,000 alien 9710's a week, we have been getting tags that have misc characters in the User memory area and we also get multiple EPC's from the same tag (done in a clean room). We have used 450,000 rfid tags so far & have gotten approx 100 corrupt tags. Is this a normal expectation or do I have issues with RF interference or bad tags?

 

[Howard Heckman III]

Hi Ray,

Could you give some additional detail on your setup? Is this an installation that uses TracerPlus or ClearStream? Is the User memory bank something that you are encoding, if so what is the software/printer you are using?

As for the multiple EPCs from the same tag, could you elaborate on this a little more? Does this mean that if you scan it a second or a third time the value is changing?

Thanks, let us know on this.

 

[Ray]

We are currently using Zebra ZT410 RFID printers (changing to on line programmers/auto label). The Epc is coming from our Database and the data load in the user memory is part number and date all in one 16 character string. We are using Clearstream demo (we are purchasing) as a tryout to scan part bins (80 parts) Clearstream works great except the process closes prematurely sometimes. After bin scan the suspect bins get tagged for manual sort, this is where we use Tracerplus apps on a Motorola M9190z and also Ugrokit with cell phone. Both portable platforms return the same results. As for the multiple EPC's all you have to do is scan part with portable device for 10 sec and multiple EPC's show up. Upon looking at the EPC on tag it has been corrupted. It is my belief that the Zebra ZT410 RFID printer validates what it programs & if it cant it "Voids" the tag & goes to the next one. One more thing near the plant about 600ft away is a very tall radio tower with many antennae on it.

 

[Howard Heckman III]

Yes, generally the RFID printers will validate the tags and mark as void if they fail. I have never heard of the EPC banks becoming corrupt in the way that you are describing. If both the uGrokIt and MC9190z are reporting the same thing than it does sound as if the tags are bad.

Could you also try scanning the tag using some demo software on the device (something other than TracerPlus)?

 

[Ray]

We have tried the demo software that comes with the MC9190Z and also my bin scanner is utilizes Clearstream with an Alien 9900+ all same results. Find attached a good sample of our bin scan results. I'll go to our plant today & retreive the bad tags from our Quality people and post Screen shots.

 

[Ray]

I was just thinking of a cause for this, network interruption. Do you think this can corrupt the data as the data comes from a remote location (Corp office) and their is no redundant network. We have a real crappy connection to our Corp office and network drops quite a bit. I will also try and reprogram the corrupt rfid tags.

The tags reprogram fine no multiple EPC's etc.. so I dont think it's the tags. Could it be database formatting?

 

[Howard Heckman III]

Hi Ray,

I'm not sure what could be causing this. I guess if somehow the data that is going to be programmed to the tag is not correctly retrieved it could happen. Is it possible for you to pull all the data local before encoding the tags? If you do this for a while you might be able to narrow down whether or not it is a network issue.

I'd also be interested to see a sample of these tags that you say contain multiple EPC values.

 

[Ray]

Here is a sample read of one of those tags using Ugrokit & Android phone. 7 epc's!!!! FYi our process doesn't use letters in the EPC.