• State TI Thinks Resolved
  • Locked Locked
  • Replies 9 replies
  • Answers 2 answers
  • Subscribers 30 subscribers
  • Views 1465 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

IWR1642BOOST: IWR1643BOOST

peter mccormick
Expert 1110 points
Part Number: IWR1642BOOST
Other Parts Discussed in Thread: IWR1642, IWR1443, IWR1443BOOST

Oil sits on top of water.  Guided Wave Radar rely on large differences in the dielectric constant of layered fluids in a tank to determine the 'thickness' of each layer.  The dielectric constant of oil is about 2.  Water is about 70.  Radar partially reflects off of a lower dielectric constant then the remaining wave reflects off of the next layer....  Do you think that the IWR1642 would work trying to determine where the surface of the oil is and where the oil-water interface is several inches from the surface?  We will try it if you think it has a chance.

Thanks, Peter

  • 0
    TI__Guru 50240 points
    Hi Peter,

    Yes! Although we have not tested this yet, it is a priority for our mmWave team later in 2018. You may need a different chirp configuration but this is definitely a solid use-case for IWR1642.

    Let me know if you have more questions!


    Cheers,
    Akash
  • 0 in reply to Akash Gondalia
    Expert 1110 points

    We have tested the IWR1443 for accuracy up to 40ft and from what we recall, it is accurate to a few mm.

    Today, we tested a IWR1443 and it appears to be consistently off by 70mm.  If we measure distance to a metal plate or to water at 700mm, the Visualizer shows 770mm.  It is consistently off by 70mm.  If we subtract 70mm from the visualizer, that is the actual location of the IWR1443 from water or the plate.  The signal on the graph is very strong.

    I am confused what may be different since it has read accurately in the past?  Bad board, bad software load, necessary offset that we don't know about...?

    Thanks, Peter

  • 0 in reply to peter mccormick
    Expert 1110 points
    If we put the IWR1443 in a plastic box, the visualizer shows that the signal decreases substantially. Plastic boxes have a dielectric constant of 2-4 while water has dielectric of about 80. Will going through a plastic box be a problem? Sometimes wet dirt or sand may collect on the box with dielectric of about 20 but we hope the IWR1443 will see through the sand and plastic box. Do you see this as "killing" a reliable return signal to water of 1-10 feet? Peter
  • 0 in reply to Akash Gondalia
    Expert 1110 points

    Akash,

    We are getting good results with the IWR1443 high accuracy and have it doing auto configuration on boot up and sending [mm] readings at any baud rate among other features.

    I read that the IWR1642 gets "16xx has much higher accuracy than14xx".  For our tank level measurement application would we get better accuracy with the IWR1642?

    Can the high accuracy IWR1642 lab detect 2 objects at 1ft separation with high accuracy?

      Thanks, Peter

  • 0 in reply to Akash Gondalia
    Expert 1110 points

    Akash,

    You mention a lens in your tank level measurement video to focus the beam.  What is the typical angle of the IWR1443 beam?  I don't think optical lenses will work at 80Ghz but maybe it has some effect.  Is this a biconvex optical lens or some kind of foil Fresnel lens?

    Thanks, Peter

  • 0 in reply to peter mccormick
    TI__Guru 50240 points
    Hi Peter,

    Unfortunately we don't have any information on the lens at this time as the lens used was a prototype.

    The field of view for IWR1443BOOST is +/- 60 degrees in the azimuth and +/- 15 degrees elevation.

    I think you're right that an optical lens will work at such high frequencies and that you'll have to look at the recommended frequencies from the microwave lens manufacturer.

    As far as the High Accuracy 16xx question is concerned, I can confirm that the 16xx does have an improvement in accuracy and it only returns a single point although it is possible to alter this. If you'd like to have a discussion about that particular lab then I'd recommend opening a new thread.


    Cheers,
    Akash
  • 0 in reply to Akash Gondalia
    Expert 1110 points

    I understand part of the reason for multiple receive antennas so you can get angle.  Why do your have multiple transmit antennas?

    d=fc/2S=f*(3e8)/(2*100e12)=f*3e-6

    The distance "accuracy"  I think is 1/T where T=chirp time=40uS  1/T==25kHz            

    d=25e3 * 3e-6=.075m=75mm   Obviously, my math is wrong since your distance accuracy is much better than 75mm.  How do you get <1mm ?

    Peter

  • 0 in reply to peter mccormick
    Expert 1110 points
    CORRECTION
    S=4Ghz/40µS=100Mhz/µs=100e12

    d=fc/2S=f*(3e8)/(2*100e12)=f*1.5e-6

    The distance "accuracy" I think is 1/T where T=chirp time=40uS 1/T==25kHz

    d=25e3 * 1.5e-6=.0375m=37mm Obviously, my math is wrong since your distance accuracy is much better than 37mm. How do you get <1mm ?
  • 0 in reply to peter mccormick
    TI__Guru 50240 points
    Hi Peter,

    These questions are outside the scope of this thread. If you post them in a new one then I can ask our High Accuracy Expert to look at this for you.


    Cheers,
    Akash