• State TI Thinks Resolved
  • Locked Locked
  • Replies 4 replies
  • Answers 1 answer
  • Subscribers 32 subscribers
  • Views 598 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.

DRV425: Long term offset stability and hysteresis compensation

Jasper Keuning
Prodigy 70 points
Part Number: DRV425

Hi,

 

I have some questions regarding the DRV425 device.

My intended usage is to use several of these sensors to estimate the 3D position of a permanent magnet with respect to the sensor location. The target is to detect this magnet within a range from a few mm up to several (5+) cm. A rudimentary FEM simulation tells us that we would like to measure down to ~ 2uT at our sensor location and it is possible that the sensor will experience fields up to 200mT. We would like a measurement BW of ~100Hz.

I have a few questions about some situations that may apply to my product and I hope you can provide me with some more insights. My apologies in advance if things are clearly stated in the datasheet or have already been explained in the forum, and I have failed to notice them!

I believe that noise and temperature drifts using this sensor and a suitable ADC will remain below 2uT. I therefore think my biggest obstacle at the moment is initial offset and hysteresis.

Q: The worst case sensor offset is +/-8uT. Can you say something about the long term stability of this offset? In other words; If I compensate for this offset once after production, will this compensation be valid over the intended life-time of my instrument (several years) assuming conditions remain the same? Or do I need yearly/monthly/… calibration/adjustment.

 

Q: You mention a typical hysteresis of 1.4uT when a -/+10mT sweep is applied. Is there an easy answer to what causes this hysteresis? Does this number change when we apply +/-200mT? Is there some predictability to this hysteresis, if we can for instance predict the actual field strength and direction at the location of the sensor?

 

Q: If I place the sensor in a 2mT homogeneous field that is in the x-direction. What can I expect from the sensors sensitive in the y and z direction? Is there some crosstalk or is the sensor absolutely insensitive in the directions perpendicular to the sensitivity axis. (and what if this field is 200mT?)

 

Thank you in advance for your reply,

 

With kind regards,

 

Jasper Keuning

  • 0
    TI__Mastermind 26650 points
    Hello Jasper,

    Thanks for investigating the DRV425. We are looking into your questions and will be back with you shortly.

    Sincerely,
    Peter Iliya
    Current Sensing Applications
  • 0
    TI__Mastermind 26475 points
    Jasper,

    A few things to note before I get to the answer about the DRV425 behavior. A question would be is how you are going to eliminate any external magnetic fields or movement? Movement would introduce earths magnetic fields variation. If you have some shielding that would eliminate any external fields. Another issue is magnetic field variation from a magnet it self. I know that magnets variation over temperature are typically much more than the drift of the DRV425. How do you plan on dealing with that variation?

    Q1) I do not expect a large variation of the offset. If you are making these measurements to verify please make them in a zero Gauss chamber that eliminates any external fields and be careful that there is no varying currents surrounding the DRV425. Do not use a magnetic and use a Helmholtz coil to measure the drift of the device. Also please keep your Voltage supply constant as there is a power supply rejection issue that can vary the measurement when making precision measurements. See typical curve with Offset vs Supply Voltage in datasheet.

    Q2)This is due to the magnetic properties of our sensor. I have not got this data but I know of data that was collected with 2 Tesla and the device came back with no issues.

    Q3) The rejection will mostly be due to the angle on the sensor. The location of the sensor can vary in all axis(X,Y, and Z) by 0.025mm. This error in placement will give you the largest signal variation from fields in the non-sensitive direction.
  • 0 in reply to Javier Contreras4
    Prodigy 70 points

    Hello Mr Contreras,

    Thank you for your quick reply. 

    Currently we are still in a preliminary design phase and are evaluating (mostly on paper/simulation) sensor performance for our application. Our current idea is to use an IMU + eCompass that will be integrated in the probe for compensation for the Earth magnetic field. The goal is to have the IMU on the far side of the probe so the magnet we're trying to locate does not influence the measurements too much.

    We're not planning to deal with the temperature variation of the magnet itself. For now I see this as a gain-error, which is acceptable. 

    1) I am glad to hear you expect this offset to be relatively stable. Other factors such as a power supply and sensor temperature we can control/predict and factor in and/or compensate for. 

    2) So if I understand correctly 1.4uT is the worst case hysteresis we can expect, and our 200mT field will not degrade or damage the sensor?

    3) thank you for this comment, we have not yet included positioning accuracy in our model yet and if you say 25um will give us significant errors, I should try to quantify placement errors due to the manufacturing process and determine their impact as well.

    Thank you again for your response, I have some more homework to do. If I have further questions, I will post them here or open a new thread.

    With kind regards,

    Jasper Keuning

  • 0 in reply to Jasper Keuning
    TI__Mastermind 26475 points

    Jasper,

    Regarding Q2 I do not have the data.  I meant that after 2 Tesla they did not see major issues.  The actual level of hysteresis for that test I do not have the data.  The test was basically to determine if there were drastic failures and the part can recover.  Normally offsets in the lever of uT you need a zero-Gauss chamber to get any real numbers.

    Please let me know if you have further questions.  I will mark this thread as resolved you can add to this thread or open a new one like you stated.