• State Resolved
  • Locked Locked
  • Replies 7 replies
  • Subscribers 30 subscribers
  • Views 2529 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.

Surface Mount (SMT) Inductor with LDC0851

Sanika Joshi
Prodigy 80 points
Other Parts Discussed in Thread: LDC0851

Hello,

I am developing a small handheld device and need to use proximity sensor LDC0851. My device operates on 3V battery. Target distance is approx 5mm and switching distance is 1mm.

Since I have constraint on the PCB area, can I use unshielded surface mount inductors instead of Sense and Reference coil? 

Thank you.

Sanika

  • 0
    TI__Genius 13335 points
    Hello Sanika,
    Unshielded surface mount inductors can certainly be used, but there are two concerns I would caution.
    1) The LDC0851 switching operation relies on the matching of the two inductive elements (LSENSE and LREF). If these two are not matched or they vary part to part then the switching distance will also vary. PCB inductors can be matched very well and therefore this is not a concern.
    2) The switching distance is a function of the magnetic field strength. Typically for a PCB coil this is dictated by the coil diameter. We typically say that the LDC0851 should be used to switch within 40% of the PCB coil diameter. Unshielded inductors may also work, but they will not get very good sensing range. We suggest to prototype the solution, but likely you will not get more than a couple mm.
    Regards,
    Luke
  • 0 in reply to Luke LaPointe
    Prodigy 80 points

    Hello Luke,

    Thank you for the answer. My switching distance requirements are approx 1-2 mm only.

    As per your feedback, I tried this with the LDC0851 EVM and 10uH un-shielded inductors. So far it seems to work for my application. Planning to use the same for my prototype.

    Thanks,

    Sanika 

  • 0 in reply to Sanika Joshi
    Prodigy 80 points

    Hello Luke,

    I am using SMD inductors for Lsense and Lref with LDC0851 in my prototype design. In some PCBs I am seeing issue due to inductor value mismatch. I have used these inductors : https://www.digikey.com/product-detail/en/murata-electronics-north-america/LQH2MCN100K02L/490-4046-1-ND/1016306. Their tolerance is 10%.

    I am setting ADJ input in my program to offset the inductor value mismatch. So for each device the value differs according to the Lref and Lsense value mismatch.

    In most of the cases it works well. But when inductor values differ a lot the switch doesn't work properly. 

    I can't use a pcb coil because of the area constraints. 

    Can please you recommend any solution for this issue? Should a lower/higher value inductor be used? Could 5% tolerance solve this? 

    Thanks,

    Sanika

  • 0 in reply to Sanika Joshi
    TI__Genius 13335 points
    Hello Sanika,
    The problem you are seeing has to do with the tolerance of the components. The best solution is to use inductors with tighter matching. This is why we recommended the PCB coils which can be very well matched especially if they are on the same PCB. However, since you are unable to use PCB coils, you may consider using two different values of inductances such that Lr - 10% is always greater than Ls +10%. For example, if you're picking 10µH for Lsense, then pick 12.5µH for Lref. This could potentially result in the output always being ON (LOW) unless you choose the appropriate ADJ setting. Also, because the tolerances on the components vary part to part you could potentially get a worst case mismatch of ~53% which could significantly impact your switching distance, however, you would be able to roughly tune this out with the ADJ setting, which has the ability to subtract ~55% offset to the reference inductance at max value.
    Regards,
    Luke
  • 0 in reply to Luke LaPointe
    Prodigy 80 points

    Hello Luke,

    Thanks a lot for providing me with a solution to this issue.

    So with my current inductor values I am seeing issue because Lref sometimes goes bellow Lsense. And that can't be adjusted with the ADJ value. Right?

    Instead of using 10uH and 12.5uH with 10% tolerance, can I use 10uH and 12uH with 5% tolerance? So in worst case scenario Lref will be 12.6uH and Lsense will be 9.5uH. Would this work if my switching distance is around 1mm?

    Thanks,

    Sanika

  • 0 in reply to Sanika Joshi
    TI__Genius 13335 points

    Hello Sanika,

    The ADJ setting only subtracts an inductance offset from the LREF channel. If the nominal LREF inductance is already below the LSENSE inductance then using the ADJ setting to subtract more inductance will not help your sensitivity.

    You can also use inductances with tighter values. I gave an example using 10%, but certainly improving the tolerance will help. 5% with the values you've chosen will work better and give you more options with your ADJ setting. 

    The switching distance will depend on the strength of the field and the matching of the inductors. The longest switching distances are achieved with the best matching between channels. Because there is tolerance on the inductor, it will likely require tuning the ADJ setting to find the best matching for each board unless you pick an ADJ setting that works for all of the values and are able to accept the large variation on the switching threshold as a result. I would also imagine that package size and type (wirewound vs ceramic, etc) will have an impact on the switching distance as well.

    Please feel free to post your results here of your testing for the benefit of others as well.

    Thanks!
    Luke LaPointe

  • 0 in reply to Luke LaPointe
    Prodigy 80 points

    Hello Luke,

    Thanks a lot for detailed explanation. Yes, I will try with 10uH and 12uH inductor values and will post the results.

    Thanks,

    Sanika