• State TI Thinks Resolved
  • Locked Locked
  • Replies 8 replies
  • Answers 2 answers
  • Subscribers 33 subscribers
  • Views 970 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.

DRV421: Magnetic shield design

Mattia Berton
Prodigy 210 points
Part Number: DRV421


Hello,

I'm developing a current sensor using a toroid in magnetic steel, the same used for classical 5A transformer. I made 1000 turns with an inductance of 380 mH. The DRV421 is inside a slot created in the core as you can see in the photo attached.

The DRV421 is powered at 5V. The shunt is 100 Ohm. 

If I leave with no current, i see this in my multimeter:

 

However, if I only put my chair (with some parts in metal) near to the sensor, the data output changes a lot.

 

It seems as if the sensor is able to capture the earth magnetic field, and in some way the metal chair changes this field. Do you think my idea could be correct?

It seems as if it is necessary to add a shield to the toroid. Where should I put the shield? What material do you think is better suited for this? 

Thank you,

Mattia 

  • 0
    TI__Mastermind 32680 points

    Hello Mattia,

    What is your magnetic gain?  The placement does not look optimized for the DRV421.  The DRV421 center should be at the air gap of the magnetic core and get as much of the fridge fields as much as possible.  This post goes over a lot of this with some tools that can help.  Most of the time the earths field is mitigated as the core diverts it away form the DRV421.  In this case I think your placement prevents from this and does not get the optimal magnetic gain. (Magnetic field seen at DRV421 per A at the primary)  See DRV421 System Parameter Calculator

    https://e2e.ti.com/support/sensors-group/sensors/f/sensors-forum/683921/drv421-question-about-use-case-condition

    I have two air gaps images below.   The DRV421 center must be placed in the highest fringe fields as possible in the direction of its sensitivity.  You can see the DRV421 if the slot is not in the center of the air gap the DRV421 does not get the field required to get the most of it.  The first image is ideal as you can see most field will be at placement of DRV421 device.

  • 0 in reply to Javier Contreras4
    Prodigy 210 points

    Hello Javier,

    thank you for your answer. I see that, with your correct position, the sensor now is much more sensible and has a lot of less noise. That is very good.

    However it seems that the sensor is still sensible to other source (i.e. if I move my metal chair next to it, the measurement changes). This depends a lot on the material of the core: if I use laminated steel, the changes is around five times bigger than with ferrite. We are talking about the field generated by 1-2 mA. What could it depend on in your opinion?

    Regarding the magnetic gain of the core: I'm not sure about my formula, but it should be Gcore=u0*ur/lm, is it right? u0 is the permeability of the air, ur is the relative permeability, lm is the magnetic path. If my formula is correct, than the Gcore is around 0,03T/A with the laminated steel.

    Thank you,

    Mattia

  • 0 in reply to Mattia Berton
    TI__Mastermind 32680 points

    Hello Mattia,

    I normally calculate the magnetic gain is by getting the gain of the core at the location of the DRV421 sensing location.  I normally do this with simulation as the mechanical placement of all matters.  From the E2E post I mentioned there is some simulations that go though that calculation.  This is the method I would use.  With lower magnetic gain then the DRV421 is more likely to pick up external fields with something external is applied.  Also the magnetic core should act as a shield from most external fields. 

    I don't think your placement is giving you enough magnetic gain therefore small changes in external field is interfering with your measurement.  

  • 0 in reply to Javier Contreras4
    Prodigy 60 points

    Thank you Javier.

    From your answer, I simulated the magnetic gain of the core; I see it depends on the magnetic permeability up to a certain point, from the geometry and what else? How can I increase the magnetic gain of the core? At the moment I have something around 600 uT/A, and the core is behaving well, but I'd like to know what else could I do to improve my response.

    Thank you,

    Mattia Berton

  • 0 in reply to Mattia Berton2
    TI__Mastermind 32680 points

    Hello Mattia,

    I don't understand the issue.  Is the direction of the field the same as your setup?  Can you share the location of the sensor and core geometry that is used?  How did you modify from your previous low gain issue to now?

  • 0 in reply to Javier Contreras4
    TI__Mastermind 23155 points

    Mattia, we have'nt heard back in a week, so I'm going to mark this thread as closed. If you need additional support, please feel free to reply to this thread, or open a new one. Thanks for using to E2E!

  • 0 in reply to Javier Contreras4
    Prodigy 60 points

    Hello Javier, sorry for my late answer, I have been on holiday! Anyway, sorry for my past question, it was quite confusing. What I'd like to tell is: what actions should I take to increase the magnetic core gain?

    Thank you,

    Mattia Berton

  • 0 in reply to Mattia Berton2
    TI__Mastermind 32680 points

    Hello Mattia,

    The best thing is to place the DRV421 at the center of the air gap for starters.   Just like the image above you want the field to be highest at the center of the DRV421 as that is where the sensor is located.  Making the air gap small will help to a point but the gain will mostly be limited by the mechanical limitation of getting the sensor(DRV421 center) close to the airgap.  From the post I shared above you can see how to download FEMM and simulate the magnetic gain.  There is a script in the post that will allow you to manipulate the core.  The issue is that the script makes a toroidal core and not  rectangular.  This should not have a large impact.

    https://e2e.ti.com/support/sensors-group/sensors/f/sensors-forum/683921/drv421-question-about-use-case-condition