1. Limited time only! Sign up for a free 30min personal tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Faraday cage problem with RFiD

  1. Aug 4, 2010 #1
    I'm working in a tracking solution based on RFiD, that consists on tags without any type of battery or power source. When the antenna generates the signal in the correct frequency, an induction process occur in the tag, generating a response to the antenna (containing the tag's id number).

    The problem is: we need to place tags inside of an metal tripod (aluminium or iron, about 5mm of thickness), that has a few holes (1 cm each one). The frequency used is 900mhz. I believe that:
    - Could be a Faraday cage. The metal of the tripod is blocking the antenna's eletric field, so the tag isn't receiving the necessary amount of power to answer.
    - The metal is distorting the frequency, so the tag cannot receive any signal.

    I've some success using a piece of wire, connected to the metal in the tag and passing it to outside by one of the holes (like an antenna). If the wire touches the metal surface (it's properly isolated), the signal is interrupted.

    It's possible to pass a radiofrequency trough the cage? Using a frequency of 120khz or 13.5mhz could help (there are another tags with those frequencies. The 120khz one has a bad read range)?

    How can I calculate the effects of the cage (I've read that is calculated by the Gauss law)? I need some help to calculate it's effects on the radio signal frequency and intensity.
     
  2. jcsd
  3. Aug 4, 2010 #2
    I think you need to look at the 'skin depth' which is the distance EM waves can penetrate a conductor. I believe it decreases with frequency, so using a lower frequency may help penetrate the shielding. The 'optical' effects of the holes and the geometry of the metal will be much more complex to calculate...
     
  4. Aug 4, 2010 #3

    Born2bwire

    User Avatar
    Science Advisor
    Gold Member

    Best bet for this kind of work is computational electromagnetic simulations. For real applications, you are interested in detailed properties like read distance which really needs a proper solver to give you an accurate answer. The holes are not going to channel much radiation as 1 cm is too small to guide waves in any frequency desirable for RFID. For the most part it is going to look like a solid piece of conductor at such low frequencies and 5 mm is pretty thick too. Lower frequency will help as the skin depth will be much deeper and you may have a chance of penetrating the conductor. You could always do a back of the envelope calculation to see what kind of attenuation you can expect and use that as a rough estimate to find your read distance. Of course how you mount the tag matters a lot here since you are mounting it to a conductor. Mounting tags on a conductor is a whole other problem in and of itself.
     
  5. Aug 4, 2010 #4
    Ok, I've searched about skin depth and found this:
    [tex]\delta=\sqrt{{2\rho }\over{\omega\mu}}[/tex]

    I've applied it to 900mhz, 13.5mhz and 125khz. The results weren't what I was expecting:

    Using (aluminium):
    [tex]\rho[/tex]: 2.82×10−8
    [tex]\omega[/tex]: 2pi x frequency
    [tex]\mu[/tex]: 1.2566650×10−6

    The results:

    900mhz: 2.81721... x 10-6 m
    13.5mhz: 0.0000230024 m
    125khz: 0.000239048 m

    These calculations are correct? If it's indeed correct, I think that will be very difficult to penetrate metal with that radio frequencies.
     
  6. Aug 4, 2010 #5
    When I worked on a project to put Remote Terminal Units in remote areas of Alaska, the specifications required a stainless steel box for the equipment. (Apparently, in remote areas, people will shoot at anything.) When the equipment arrived, the local engineer called and asked if the radio antenna had to be on the outside of the box.

    The answer is Yes. The antenna has to be outside the Faraday cage for electromagnetic frequencies normally used for radio communications, including RFID.

    Why does the tripod have to be entirely metal? Are you tracking tripods?
     
  7. Aug 4, 2010 #6
    As you can see, trying to penetrate the skin debth with low frequency requires the metal to be too thin, which is pointless. It is better to increase the frequency and use a structure with holes in the metal. This way some penetration becomes possible and structural strength is maintained. You've probably seen some faraday cages that are really cages or screens, rather than solid metal. Because of the holes in them, they only behave as a Faraday cage at lower frequency. Microwaves get through them. And, of course you can see through them, which tells you somthing about even higher frequency penetration.
     
  8. Aug 4, 2010 #7

    Born2bwire

    User Avatar
    Science Advisor
    Gold Member

    Unless he's willing to drill holes in these tripods, which I assume is outside allowable options, 900 MHz is not going to cut it for 1 cm holes. The wavelength is around 33 cm and generally a quarter wavelength is a good cutoff size for the maximum allowable gap in a Faraday cage. I think his best options are to mount the tags on the exterior surface in line of sight of the reader or connect the tag to an external antenna (which almost defeats the purpose of the tag in the first place, you might as well just design a new antenna and smack the RFID chip onto it)
     
  9. Aug 4, 2010 #8
    Yes, I agree and realized that. What I'm suggesting to him is that this line of thought is potentially more productive than going to lower frequency. Maybe he can go to 2-3 GHz and perhaps the biggest allowable holes may let a small signal through. Perhaps some type of horn antenna with a waveguide down the tube could do the trick. I haven't really tried to understand the details of his setup, but perhaps there is room for him to brainstorm and make some modifications to the original design plan.
     
  10. Aug 4, 2010 #9
    We're tracking film equipment, including cameras, lens, tripods and other itens. The items are very expensive, demanding a monitoring solution. The client wants to use RFiD (we have been working with it by some time, so we've acquired some expertise). Another demand is that the tags cannot be visible (here is the main problem).


    Interesting. The client's tripod is thicker (maybe 2 or 3 cm), and could have bigger holes. The biggest holes are on the extremities (it's just like an iron pipe).


    We're having some success simply welding the extremity of a wire in the tag, isolating the tag and hiding it inside the tripod, with just some wire to outside (fixed on the tripop with insulating tape). The result is almost invisible and it's possible to read the tags. The signal is weak, and the read range is very reduced. In our client situation, the tripods will be stored together, so there will be to much metal.

    Possible solutions:
    - 3 tags per tripod, one by leg, with the same ID. That'll improve readings and reduce problems to detect tags.
    - Using HF tags (13.5mhz), less sensitive to metal.

    These solutions could work well? With one tag (900mhz), I'm having 30cm of read range. The ideal situation would be an improvement to a 2 m read range.
     
  11. Aug 4, 2010 #10

    Born2bwire

    User Avatar
    Science Advisor
    Gold Member

    RFID is not meant to be directional and is generally low frequency and are near field devices. 900 MHz is the highest frequency that I have heard being used for tags and it would not be useful for RFID to use higher frequencies. The idea is to have localizd scan regions of uniform strength so that you can pass product or items over a scanner rapidly and have consistent scans of the multiple tags at the same time. This like having a tray of medications passing over on a machine line belt, or items on a shelf that actively tell the inventory system when they have been removed, trucks leaving a warehouse and having their contents scanned, etc. The low frequencies help the signals penetrate packaging, limit their scan distances, and allow simple techniques like inductive coupling of the RFID antenna and scanner.

    It's hard to say. My first inclination is that you are not going to really get any improvements in penetration of the equipment due to the thickness of the metal. In addition, unless these tags are designed to be placed on metal then just attaching to a metal surface can degrade performance due to the image currents that are produced in the metal but that is not an effect that I have done any research on myself. The fact that they can be stacked or stored together can cause other complications. You would want the tag to be uncovered when the items are together. Still hard for me to say as with the lower frequencies this is all near field and thus my intuition with scattering problems goes out the window.
     
    Last edited: Aug 4, 2010
  12. Aug 5, 2010 #11
    Just a question. Could be possible to connect the antenna of the tag (using a wire or other thing) to the metal structure and use it like a big antenna? If I put the tag inside of the tripod, it's not possible to read it. But, holding it with my hand in the same position, it's possible to read it. I'm starting to believe that in this case it used my body as antenna. If I could do that with the tripod, maybe I could get some reads.


    The possible frequencies are: 125khz, 13.5mhz, 900mhz and 3.5~5.8mhz (microwaves). The lower one would be ideal, but his range is limited to 50cm, maybe 1m. With 900mhz, we can obtain 10~15m. I believe that 13.5 mhz tags could reach up to 3m.

    We've two possibilities to track the items:
    - RFID antennas all over the building. Tha tags would be monitored all the time, and an alarm would be activated if one tag dissapears.
    - RFID antennas in doors, windows and other exits. The items wouldn't be monitored, but the alarm would be activated if an item exits the building without properly authorization. Another advantage is that we would't have to read a lot of items togheter, it would be one by one.



    The idea isn't penetrate the metal, what looks really impossible. I'm hiding the tag inside the tripods, with a wire welded on it. Then, I just left a tip of the wire out of the metal (hiding it on the tripod). It's only visible in a very close investigation.

    It's possible to read it, but the range is limited to 50cm (I'm improving it changing the position of the wire). The idea is to use 3 tags (with 3 wires on them), one by leg, all with the same ID, and read it with more antennas. I think that this could work well with 13.5 mhz (medium range, less problems with metal) and using the second strategy that I described above.

    Putting the tags outside of the tripod is a solution. There are tags designed to work over metal. We could isolate normal tags to, using a layer between the tag and the metal, giving to it the necessary distance. Actually, we had similar problems using tags to track racing times (running). Each competitor had a tag attached under over the clothes. We had to isolate the tags, because the contact with the skin caused problems to read the tags.
     
    Last edited: Aug 5, 2010
  13. Aug 5, 2010 #12

    Born2bwire

    User Avatar
    Science Advisor
    Gold Member

    These are questions that would be best posed to the RFID tag manufacturer. When a tag is designed, the antenna on the tag is designed under certain assumptions about how it will be mounted. In addition, it is tuned to couple efficiently at the desired frequency and tuned to have a matching impedance with the RFID chip for efficient transference of power. Adding a wire to the antenna changes the antenna's impedance and will adversely affect the efficiency of the tag. These kinds of questions and such should be something that the manufacturer can answer for you. I don't think we can give you very detailed answers without knowing intimately the tags that are being used, the manner in which you need to use them and the performance constraints you are working under. Sure you can run a wire off of the tag to try and pick up a signal but I can't say if that will work well enough for your purposes. Electrically isolating the tag and then shorting the antenna to the surface of the tripod. Sure, that might work, but again it is using the tag in a manner that it isn't designed for and I can't say if it would work to your satisfaction. It could, but I can only say that its performance should be worse than if the tag was attached in accordance to manufacturer's specifications.
     
  14. Aug 6, 2010 #13
    The performance is indeed worser (10 m is the normal, I'm getting 1 m now with some improvements in the position of the wire). The support of the manufacturer of the tag is a problem. When we contacted them, it was clear that we're knowing more about the product than them.

    Unfortunately, with the tag placed inside metal, an external antenna is our only choice. I'll test som other possibilities, like active tags (more range and don't depend of the antenna power, because they already have a battery integrated) and more powerfull antennas.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook




Similar Discussions: Faraday cage problem with RFiD
  1. Faraday cage (Replies: 3)

  2. Faraday cage problem (Replies: 9)

  3. Faraday cage ? (Replies: 5)

  4. Faraday cage (Replies: 10)

Loading...