• State TI Thinks Resolved
  • Locked Locked
  • Replies 6 replies
  • Answers 1 answer
  • Subscribers 31 subscribers
  • Views 151 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.

TMAG5170D-Q1: Accurate and repeatable linear and rotary absolute position sensing

Rod Watt
Prodigy 10 points
Part Number: TMAG5170D-Q1
Other Parts Discussed in Thread: TMAG6180-Q1, TMAG6181-Q1, LDC1612

Tool/software:

Hi,

We are researching technologies to implement absolute encoding onto some safety critical actuators for the automotive industry.

Some target specifications might be:

  • Linear - 50mm range, accurate to 10µm or less
  • Rotary - 180° range, accurate to 0.05° or less.

At this stage we have no restrictions on supply voltage or output signal.

Would it be possible to get some guidance with selecting appropriate sensors for our application? We have been having a look at the automotive versions of your linear Hall, multi-axis linear & angle position, and inductive sensor AFEs and are a little overwhelmed by the variety of options. 

Our aim is to develop confidence in a technology that we can implement at scale so accuracy, repeatability, low noise, and small packaging are top priorities.

Please let me know if I can provide any more information.

Kindest regards,

Rod

  • 0
    TI__Mastermind 18180 points

    Rod,

    Thanks for reaching out.  There are quite a few challenges in achieving these targets, which I'm sure you are aware. 

    Linear - To sense 50mm with 10um accuracy over that entire range will likely not be possible with a single sensor.  It might be possible using a large magnet (>5cm) to create a linear field ramp, but you may still have to perform some linearization on the result to achieve this result.  A particular concern with such resolution will always be the SNR of your position calculations.  You will need to make sure that you use as much of the input range as possible to perform such calculations.  What might be easier is to create an array of sensors along the 5cm path to shorten the range that each device needs to be maximally linear. There are some details about this configuration here:

    https://www.ti.com/lit/an/slya051b/slya051b.pdf

    https://www.ti.com/lit/ab/sbaa540a/sbaa540a.pdf

    Rotary - To sense angle below 0.05 degree can be very challenging with most technologies.  My recommendation would be to consider TMAG6180-Q1 for this application.  You will need to perform a system level calibration to eliminate mechanical errors from the result and to cancel device level errors.  A possible approach you could also take is to use a multi-pole ring magnet and to track that with TMAG6181-Q1.  This device has a turns counter that increments and decrements each time a the magnetic field changes quadrant, so you could track multiple pole pairs across this 180 degree rotation.  You might be able then to upscale the electrical resolution to reach the mechanical tolerance you need.

    Another approach could be a pair of rotating magnets where one magnet has one more set of poles than the other. Place one pole pitch at an outer diameter and a second pole pitch along the inner diameter.  If you place a sensor to detect either side of this configuration, you can track the difference in the angle these devices measure for a finer resolution as well.

    Thanks,

    Scott

  • 0 in reply to Scott Bryson
    Prodigy 10 points

    Hi Scott,

    Thanks so much for the detailed response, links, and suggested sensors! We are indeed becoming more and more aware of how challenging this is as we continue to explore sensor options, so, it is incredibly useful to receive professional advice and direction.

    One follow-up question for now - do you know if your LDCxxxx inductive sensors would provide significant benefits over magnetic for achieving our target accuracy/resolution?

    Kindest regards,

    Rod

  • 0 in reply to Rod Watt
    TI__Mastermind 24855 points

    Hey Rod,

    Scott is out, so I will try to fill in for him here.  Our inductive sensors provide the benefit of being a little more robust and immune to external magnetic sources.  Another benefit of an inductive solution might be a single sensor for the linear solution.  You simply would need to adjust the target and sense coil lengths on your pcb.  Despite this benefit, it might not give you as good of resolution as the Hall solution.  I will reach out to our systems engineer to see if he can give some more guidance on what to expect for your performance concerns.

  • 0 in reply to Patrick Simmons
    Prodigy 10 points

    Hi Patrick,

    Thank you very much for the fill in and insights. Sounds like we need to get some sensors and start testing.

    Any further information from your systems engineer would be much appreciated.

    Thanks,

    Rod

  • 0 in reply to Rod Watt
    TI__Mastermind 24855 points

    Hey Rod,

    Our system engineer, who can provide more insight, is out for the holidays.  When he comes back next week, I will give you an update.

  • 0 in reply to Patrick Simmons
    TI__Mastermind 24855 points

    Hey Rod,

    Sorry for the late update, we presently do not have any reference designs regarding the linear measurements for the inductive devices that I am aware of.  However, there is a link here in which someone used our LDC1612 and we have some premade coils similar to what he used on our LDCCOILEVM.  As for rotary encoding, this reference design might be useful.