Shop OBEX P1 Docs P2 Docs Learn Events
Hall effect sensor? — Parallax Forums

Hall effect sensor?

InteractivesInteractives Posts: 83
edited 2005-07-12 19:55 in General Discussion
Hi there, Ive been working on a project that will turn on an alarm if one object comes within about 5" of another specific object. Ive built the alarm circuit, and I've got some hall efect sensors, but I can only get the hall effect sensor to trip the circuit at about 1" away! Im using very strong NIB magnets, so I know thats not the problem. Someone on this forum once said that I could achieve 5" trigger distance with "a linear output and use an op amp and comparator to adjust range and sensitivity." Anyone have any idea how I could use an Op Amp and comparator to achieve this? confused.gif

Comments

  • KenMKenM Posts: 657
    edited 2005-07-02 17:47
    I believe the linear output mentioned may have referred to the hall effect sensor.

    If you have a digital output hall effect, the op amp and comparitor circuit will not help you.

    For starters, could you please give the part number of the hall effect device you have?



    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Ken
  • Chris SavageChris Savage Parallax Engineering Posts: 14,406
    edited 2005-07-02 19:31
    Hello,

    ·· There are other options for detecting approaching objects.· For example, we have the PING))) Sensor, although it's not perfect for detecting people due to the lack of reflectivity, but could possibly work.· It has a range of 2cm to 3.35m.

    ·· We also have the Sharp GP2D12, which can detect objects from 10cm to 80cm.· Not sure the full range you are looking to detect, but these could see past what the hall effect sensor will.



    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Chris Savage
    Parallax Tech Support
    csavage@parallax.com
  • InteractivesInteractives Posts: 83
    edited 2005-07-02 20:59
    Thanks guys, actually the hall sensors I have are HI300, from Jameco, they opperate at 3-20 V * AMPS, then I also have some fancy bipolar ones from Allegro. part number A3212. The ping thing wouldnt work, because Im trying to detect a specific object. Think of a watch with a NIB in it, that would trip an alarm every time the wearer got within 5" of the circuit.
  • InteractivesInteractives Posts: 83
    edited 2005-07-02 21:00
    (thats 8 amps.)
  • KenMKenM Posts: 657
    edited 2005-07-02 22:03
    If you have tried both of your hall sensors at 5" and neither works, there is not much else you can do other than a stronger magnet, and you mentioned you already have a very strong magnet.

    The type of hall sensor you have will make the output switch low (provide a ground) after a certain level of magnetic flux density, and apparantly the magnet 5" away is not enough to trip the hall sensor.

    The linear hall sensor (which seems hard to come by) basically outputs a voltage based on flux density. The stronger the flux density the higher the voltage.

    With the linear sensor at a distance far from the sensor a small voltage might be produced, but that voltage too small to be read by your stamp......the small voltage could be amplified with an op-amp and fed to a comparitor.....something the stamp can deal with.

    I ordered some linear hall sensors from Allegro about 6 weeks ago and am still waiting for them to arrive.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Ken

    Post Edited (KenM) : 7/2/2005 10:40:22 PM GMT
  • InteractivesInteractives Posts: 83
    edited 2005-07-02 22:16
    thanks Ken, I wonder... the linear sensor changes according to the flux desity, so wouldnt the Op amp be increaseing the signal thats coming from the Hall sensor? I think if thats what your saying, then five inches probably still wont work. But thanks so much for the info- Ive been sitting here fiddling all day, thinking that I've been hooking it up wrong or something. I've soldered it up twice today, and twice thrown the damn thing across the room. I think I might just start to look for a different way to acheive my goal. Poo.
  • KenMKenM Posts: 657
    edited 2005-07-02 22:40
    Correct,
    There is a chance that there is enough·signal·present at 5", and if there is enough it be amplified. (too little signal and you·only amplify noise).
    But that remains to be seen. I don't have my sensors yet, so I have no idea how sensitive the linear sensors are.
    About the object to be detected, can you give a little detail about the object, will it always travel in the same path?
    Interactives said...
    thanks Ken, I wonder... the linear sensor changes according to the flux desity, so wouldnt the Op amp be increaseing the signal thats coming from the Hall sensor? I think if thats what your saying, then five inches probably still wont work.
    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Ken
  • InteractivesInteractives Posts: 83
    edited 2005-07-06 13:48
    Yes, the object to be detected will be fairly random. I was thinking of putting magnets on a dog collar, and having an alarm go off (more or less) when the dog goes thru the doggie door. The details are fuzzy here, and for that I apologize: I'm actually working on this for·a friend who thinks he will be able to patent it and make a million bucks. So suffice to say, there are minor variations involved, but essentially, that's the trigger threshold we'll be working with. I was thinking of looking into RFID stuff, since the hall sensor doesn’t seem to have the capabilities I was looking for. I was looking into the EM electronic theft control devices they use in department stores. Apparently that is done magnetically as well, but it's pretty hard to find schematic or other info since it's proprietary. Karma I guess!
  • KenMKenM Posts: 657
    edited 2005-07-06 13:55
    Doggie door detection, that should be fairly simple. Somebody did this already in this forum for his cat.

    I don't have time at this second to write up an idea (getting ready for work) but I am sure others will chime in with good ideas.

    The first thing that comes to mind is an IR beam, or reflective beam.

    keyence.com

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Ken
  • InteractivesInteractives Posts: 83
    edited 2005-07-06 13:59
    Thanks Ken, you consistantly contribute good ideas! IR would be nice and cheap, but I'm worried that it would be hard to ensure that the beam would have an unbroken path to the detector. I keep picturing the collar sensor part rotating, as collars sometimes do.
  • KenMKenM Posts: 657
    edited 2005-07-06 14:06
    Another thought....http://www.parallax.com/detail.asp?product_id=605-00004

    This would mount on the plastic door near the top of the door where it hinges. The bending motion would (should) deflect the sensor enough to give a reading the stamp can interpret.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Ken

    Post Edited (KenM) : 7/6/2005 2:12:10 PM GMT
  • KenMKenM Posts: 657
    edited 2005-07-06 14:09
    Now that we know it is a doggie....the ping sensor would work great to keep the dog from coming in or going out.....although us humans cannot hear past ~20kHz,·
    a dog can easily hear a ping sensor, and they HATE it. I have built and used one as a discipline device training dogs. It ran at 40KHz.
    Chris Savage (Parallax) said...

    Hello,

    ·· There are other options for detecting approaching objects.· For example, we have the PING))) Sensor, although it's not perfect for detecting people due to the lack of reflectivity, but could possibly work.· It has a range of 2cm to 3.35m

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Ken

    Post Edited (KenM) : 7/6/2005 2:12:24 PM GMT
  • KenMKenM Posts: 657
    edited 2005-07-06 14:11
    The beam receiver and emitter, or receiver and reflector would be mounted on opposite sides of the door, so a dog, cat, or anything passing through the beam would trip the beam.
    Interactives said...
    Thanks Ken, you consistantly contribute good ideas! IR would be nice and cheap, but I'm worried that it would be hard to ensure that the beam would have an unbroken path to the detector. I keep picturing the collar sensor part rotating, as collars sometimes do.
    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Ken
  • InteractivesInteractives Posts: 83
    edited 2005-07-06 14:20
    hhhmmmm... ping sensor: here's the complicated part: if a cat walks in the doggie door, I dont want it to go off. It's never simple with me!
  • KenMKenM Posts: 657
    edited 2005-07-06 14:46
    hmmm, sense a dog only, and possibly only one dog.....things are going to get a little harder now. Need to think a bit on this one.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Ken
  • Jon WilliamsJon Williams Posts: 6,491
    edited 2005-07-06 15:18
    How about an RFID tag in the collar?

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Jon Williams
    Applications Engineer, Parallax
  • steve_bsteve_b Posts: 1,563
    edited 2005-07-06 15:44
    Is the dog larger than the cat?
    Put up a couple of IR beam break detectors....one at the bottom and one near the top height of the dog.

    Or, put a scale down in front of the door that the dog would weigh out differently than the cat....

    Just other ideas!
    RFID would be the cleanest and use up less space.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    ·

    Steve
    http://ca.geocities.com/steve.brady@rogers.com/index.html
    "Inside each and every one of us is our one, true authentic swing. Something we was born with. Something that's ours and ours alone. Something that can't be learned... something that's got to be remembered."
  • InteractivesInteractives Posts: 83
    edited 2005-07-06 16:34
    Actually I'm playing with parallax RFID right now, just got it in the mail! One question though: I down loaded the program to read the IDs from the chips, but am having trouble making an IF - THEN statement work with the number. When it comes up in the debug window, it's in ASCII. I translated it into characters, but I can seem to figure out how to make STAMP differentiate the two tags. Perhaps this could be a separate post in its self.
  • Jon WilliamsJon Williams Posts: 6,491
    edited 2005-07-06 16:38
    Since the tag number is several characters it takes a little work to differentiate -- but as the attached program shows it can be done. What you end up doing is creating a table of known tags and scanning through it, comparing the tag you're currently looking at to enties in the table.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Jon Williams
    Applications Engineer, Parallax
    bs2
    bs2
    Rfid.bs2
    5K
  • InteractivesInteractives Posts: 83
    edited 2005-07-06 16:44
    God bless you, you wonderful man!
  • Jon WilliamsJon Williams Posts: 6,491
    edited 2005-07-06 16:59
    Wow, I haven't been called that in a while. Thanks! And good luck with your project.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Jon Williams
    Applications Engineer, Parallax
  • KenMKenM Posts: 657
    edited 2005-07-06 19:19
    Could somebody post a link to the RFid product? I cannot find it at the Parallax sight.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Ken
  • ForrestForrest Posts: 1,341
    edited 2005-07-06 19:23
    RFID Reader Module www.parallax.com/detail.asp?product_id=28140
  • KenMKenM Posts: 657
    edited 2005-07-06 19:50
    Thank you Forrest

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Ken
  • MrGMrG Posts: 1
    edited 2005-07-12 11:02
    Hi,

    I am doing a project a a university, I have to dectect an oil pipe cleanng pig in the pipe using a hall effect sensor op amp etc, can anyone give me an idea of other comonents required for this?

    Thanks
  • KenMKenM Posts: 657
    edited 2005-07-12 19:55
    A couple of comments.

    I don't quite understand what you want to do. Can you try again to describe what the end goal is?

    You might want to consider starting a brand new thread, you will likely get more views and comments that way.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Ken
Sign In or Register to comment.