RFID Reader - Intermittent operation?

Del Tapparo

I am using the Parallax RFID Reader with 50mm and 30mm World Tag Unique tags in a model railroad application. The reader is mounted under a G-Scale boxcar with a microcontroller and other electronics inside. The tag is sitting on top of the railroad ties approximently 1.25" below and parallel to the reader. I want to scan the tag as the car passes over the tag to identify a uniquie location on the layout, which will then trigger some events (slow down, blow whistle, etc.) I am using a BX-24 microcontroller and X-Basic. I am able to read the tags with my software MOST of the time. But not very reliable yet. I have tried varying the distance between tag and reader, speed at which the car passes over the tag, different tags (50mm and 30mm) and can't seem to get it. The reader is enabled all of the time. (I have tried enable, read, then disable also).

In debugging my code, it appears that I either get a full 12 byte + transmission from the tag, or else I get nothing. Why I am getting nothing is the question. This will happen about 20% of the time, even at very close proximity and very slow motion. The metal rails (no power in rails) don't appear to affect it. Same problem on a table top. I am 99% confident this in not a software issue. My program monitors the serial input buffer and waits for 12 bytes of data before trying to process it. Any ideas? Anyone else have this sort of problem?

Jon Williams

The processor on the RFID reader will not dump anything out of its serial port unless it has a valid tag ID. I would be more concerned about the environment causing interference than anything else. As an experiment you might try mounting the reader vertically next to the cars and tape a tag on the outside (this is just an experiment); it would be interesting to see if moving the reader away from the metal tracks has any effect on your system.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Jon Williams
Applications Engineer, Parallax

John R.

Del;

I've been working off and on doing the same thing only backwards (tags on the rolling stock, reader under the track) in N Gauge. I've been using 12mm "capsule" tags mounted to the bottom of the cars.

I've seen the inconsistant reads, but was able to get close to a 95% read rate with a very slow moving train. Too slow to be "workable".

I was starting to do some "field mapping", taking a tag at documented 3-d postions and orientations, activating the reader, using "count" to see how long before I got a read (BS2). I had get things setup and was about to start this when "the boss" indicated that I needed to do some things on the "honey do" list, one of them turned into the remodeling project from hell. I should be back "in business" next weekend if all goes well.

I see you have the tag and readeer parallel. Is the reader laying "horizontal"? Is there a metal "weight plate" in the floor of the car? The frequencies used on these tags are not the greatest for penetrating materials, especially metal. I don't think the plastic would be a problem, but this shouldn't be ruled out.

I did not see any difference between leaving the reader enabled vs. read/off/pause/enable. I was also able to have tags fairly close together (2" or so) and still read each tag.

As another option, have you looked as "Asyncronous DCC" as an another alternative? The early applications for this appear to be exactly what you're after.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
John R.

8 + 8 = 10

John R.

Jon;

Our posts crost each other. In my testing (with N gauge, much smaller than G) I did not see any difference between powered track, unpowered track, and no track (rolling the car over the reader). Very different situation to be sure. Just for reference.

I'd be more concerned about the possibility of a metal "weight plate" embedded or sandwiched in the car floor.

Except at very slow speeds, I also saw the inconstant read success with large tags (as well as small) when passing them by hand in front of the reader. Smaller tags were harder to read (as expected.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
John R.

8 + 8 = 10

Del Tapparo

No metal weight plates. There is nothing between the reader and the tag. The only metal in the vicinity is the wheels and the track. I tried plastic wheels, no change. I can also get the inconsistent readings when rolling over a plastic table top instead of on the track.

Speed really doesn't seem to be an issue. It works the same fast or slow. "Fast" is a pretty good clip for a train. I can of course really zing it, and it doesn't work, as would be expected. But, as I said before, it also skips reads when moving slow. And shouldn't I get continuous reads if it is parked over the tag? (I don't).

Washer Medic

I was wondering if their is·a way to pre excite the tag before it gets to the reader or·has any one tried active tags

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
There's nothing a good wack with a hammer won't fix

Darn I let the white smoke out again

Goten

Forgive me if I interrupt.

But I would like to know in which language you are developing the programs.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Best Regards from Chile

Goten

Del Tapparo

Goten said...
Forgive me if I interrupt.

But I would like to know in which language you are developing the programs.

I am using the BX-24 processer by NetMedia, which uses BasicX software.

John R.

I don't think you get a repeating tag ID when "parked", I believe you need to have the tag leave the field and come back. I might have also done that behavior in my software, but I thought those smarts were in the reader. I can probably confirm that either this weekend, or sometime next week.

As far as "pre exciting": my understanding of these tags is that the field from the reader provides the energy to the coil in the tab. This is only "stored" in that there is a small capacitor in the tag, and it is immediately discharged in order to broadcast the tag ID. I'm not even sure the prime purpose of the capacitor is storage, as opposed to tuning the antenna of the tag.

I don't belive this frequency and protocol support active tags. I believe there was another forum member that seemed to talk about hacking a tag open and getting a battery in the circuit.

For Goten: I am working on an BS2 Stamp with PBASIC. I will be transitioning to an SX family and either SXB and/or Assembly.

I get the impression that these tags and readers may not be well suited to a "pass by" application, and may be more targeted as an "entry control" or more other type of interactive ID system where a user can "fiddle about" with the tag until a feedback signal of some type is seen/heard.

I have done some looking (I would not call it extensive), and have not found anything except this technology in what I would call "affordable" prices. The next level up "price wise" looks to be about $300-$700 for the readers. Tags were harder to track down, and the ones I saw were rather large.

There have been other suggestions in the form of bar codes and or "one wire" buttons, etc. I haven't given up on the RFID, but I have not gotten far enough to say it will work either.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
John R.

8 + 8 = 10

Chris Savage

John,

·· As I recall if you hold the enable line low the Read will in fact repeat the tag output.· I have done this.· That is the benefit of releasing the enable line.· You can stop the reader from sending anymore data while the line is high then return it low when you're ready to receive more data.· Most demo programs bring the enable line high while processing the data received and then return it low when ready again.· This also makes the indicator turn green when the tag has been read.


▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Chris Savage
Parallax Tech Support
csavage@parallax.com

John R.

I may indeed have done some work with the enable pin, as well as using software to throw out consecutive identical reads. I'll have to check my work and report back, hopefully over the weekend. If I had to guess whose recolection was correct, I'd put my money on Chris

I really, really, REALLY want to find a way for this to work for "car" id past various points. Kind of the "mirror image" of what Del is trying to do.

Dell: Are you using DCC for engine control?

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
John R.

8 + 8 = 10

Del Tapparo

I was in error about not getting continous readings when "parked" over the tag. I do.

I am using radio control with battery power for train control. The BX-24 is also handling the R/C (very simple).

I have some glass tags on the way. I'll see if they do any better. They are suppose to work better around a metal environment (wheels and track).