• State TI Thinks Resolved
  • Locked Locked
  • Replies 3 replies
  • Answers 2 answers
  • Subscribers 30 subscribers
  • Views 384 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

LDC1614: Multiple LDC1614's for applications with more than 4 coils

abeaulieu-tri
Prodigy 50 points
Part Number: LDC1614
Other Parts Discussed in Thread: TIDA-01102, LDCCOILEVM, TS5A3159A

Hello TI,

I'm looking at an application that may involve eight coils that may be in relatively close proximity of one another. As I understand it, the LDC1614 cycles through activation/measurement of each coil. So far with only the 4 coils on my design I have not had any issues probably due to this cycling, but I am wondering if I use two LDC1614's with 4 coils each (again, in close proximity) what the potential for interference is?

I have also seen the TIDA-01102 https://www.ti.com/tool/TIDA-01102 reference design that uses a multiplexer with one LDC1614 to multiplex the coils into the IC. I am a bit worried that the multiplexer would add some noise/interference and also would like to sample the 8 coils as close to at the same time as reasonably possible. For that reason I would prefer multiple LDC1614's if possible.

Thanks for your help.

-Andrew

  • 0
    TI__Expert 7599 points

    Hi Andrew,

    Great question. Thanks for using our LDC1614 parts. 

    Assuming that you have designed the coils such that the interactions between the coils are taken care of in terms of mechanical isolation and spacing I can point you and provide guidance on pros and cons of using multiple LDCs and single LDC with a mux. 

    Using multiple LDC:

    Our LDC1614 is designed such that it can drive the coil measure the Inductance and switch to second coil and so on. The timing is controlled as to when we activate the drive current and you have control over setting the settling time as per the application. Adding another LDC there can be a potential that LDC2 maybe driving the coil at the same time as LDC1 and due to the device driving multiple coils there could be interaction between the LDC devices. We have not really tested this scenario. It does seem plausible as long we can control the LDC1 and LDC2 using an MCU.

    This also adds additional cost to the system. 

    Also, this setup will require multiple reference clk which is key to getting a high resolution and performance. It may be a challange to get the performance with 2 LDC devices. 

    Using a Single LDC and a Mux:

    This is a cost effective solution and the performance of the system is dependent on the single LDC device. 

    Using the Mux there can be additional series resistance added MUX which can impact the system as it adds additional series resistance to the PCB coil. As long as a low series resistance MUx and a well controlled system is implemented there can be less of an interference. 

    This approach depends on the system and the performance you like to achieve. The TIDA-01102 is meant for a touch application where we are detecting a change if the user has pressed a button by measuring the deflection in the metal. This is a binary application. 

    I highly recommend for your application ordering multiple LDC1614 EVMs with the LDCCOILEVM and try to evaluate using your system. 

  • 0 in reply to Arjun_Prakash
    Prodigy 50 points

    Using multiple LDC:
    I do not mind the added cost of multiple LDCs for this design. I was thinking about using the microcontroller to put one LDC into power down mode while measuring with the other. I expect this would work but it would have some amount of latency, that's probably non-negligible when trying to capture all channels. I was also considering that I might be able to use coils with resonant frequencies that are a bit further apart. Right now I think I could have 4 coils of one type and 4 coils of another. Any idea if an approach like this would have enough noise immunity if they're operated simultaneously? Mechanically these coils would still need to be in quite close proximity to eachother.

    Using a single LDC with a Mux:
    Are there other guidelines or reference designs for selecting an analog mux for this application? I know keeping Ron minimal would be best, but I imagine the on capacitance will also affect the performance (or maybe just tuning?) of the LDC coils. For example the TI TS5A3159A looks like a good candidate for this, but I'm unsure if I should look for something with even lower capacitance.

    Thanks!

    -Andrew

  • 0 in reply to abeaulieu-tri
    TI__Expert 7599 points

    Hello Andrew,

    If cost and size is not a concern I would highly recommend to use the approach with multiple LDC. With a single LDC and a mux you have highlighted the capacitance that can impact the design. Although the capacitance of the mux is relatively small it can contribute to errors. You can use our excel calculator here and change the capacitors and see the impact to the resonance frequency. 

    As you pointed out that you are going to manage LDCs by placing them in low power mode. In regards to designing the coils with different frequencies and to understand the noise immunity I would recommend to use a tool like FEMM to design and review the eddy currents with the target. 

    As you have pointed out you are going to manage LDC using the MCU keep the coils with the same frequency would be OK. We have not tested this and I cant point you to any app notes. My concern here is when the LDCs are driving at the same time the eddy current on the coils can interact with each other managing with LDC in low power would be OK here. If your application is such that you have a metal target that is going to cover all 8 coils I would highly recommend design the coils with the same frequency. Another approach to increase dynamic range and lower the noise you could consider using one of the Channel coils as reference coil such that you can compensate for any changes in temperature.  

    Let us know if this helps!