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: test result for samples with DRV421 was not right

Part Number: DRV421

Hello!

I made some samples with DRV421,2000turns coils. but the test results was not right, the output for one sample was 4.96V, the other was 26.7mV, it was wrong ,it should be 2.5V.

why i have these unnormal results? it is because the IC welding unreliable ? the supply power was 5V. thanks very much!

  • Jie,
    Does "two samples" mean you have two magnetic cores or two PCBs, or both? I would try to debug this by removing as many variables as possible and adding them back in until it breaks. In this case, perhaps try the device without the core, and then with the core a few windings and a known current (or no current, as you have shown here). If you have two PCBs and one core, then the two PCBs are behaving differently. If you have one PCB and two cores, then the cores are behaving differently. Correct?

    Is the PCB a DRV421EVM ?
  • i have two PCB with one magnetic core.

    please see the schematic diagram, it is basically the same thing with a DRV421EVM.

  • Hello Jason! Happy new year!
    i doubt if some pins of IC were not welding on the pcb or some pins are short circuit or some other resaons. generally ,what factors will lead to this outputs 4.96V and 26.7mV?
    thanks!best regards!
  • Hello Jie Qian,

    I noticed that you have both a pull-up and pull-down connected on the GSEL* pins. So the GSEL pins are biased at mid-supply which is not a valid logic state. Each pin can either be logic high (~5V) or logic low (~0V). Please set the states of these pins properly for a valid gain setting.

    Best Regards,

    Harsha

  • Hello!Harsha!

    thanks very much , i choose the GSEL to "10", and the output seems ok, when there was no primacy current ,it was about 2.5V. but ,the value is unstable , it jumps from 2.496V to 2.506V.

    it's another problem...the GSEL i chosed did not fit ? or other reason?

    thanks very much!

  • Jie Qian,

    You have a 1nF cap load connected directly to the REFOUT pin which is most likely causing the internal reference source to oscillate. To find out whether this is the issue you can simply remove the 1nF cap on REFOUT. For low-pass filtering the reference input to the current sense amplifier, the cap should be connected to the REFIN pin, not REFOUT.

    What do you mean by GSEL did not fit? 

    Best Regards,

    Harsha

  • Hello!Harsha:

    I removed the cap on REFOUT and use a 1000 turns coils with 1J80 magnetic core, it can work stable,but the output isalso wrong ,there are two questions, please see the figure below,

    1.output not match

    according to the equation: Vout=Iprim*(Nprim/Nwinding)*Rshunt*4,Nprim=1(i use a copper bar), Nwinding=1000, Rshunt=5Ω(two 10 Ω parallel connection),when Iprim=10A, Vout should be2.5+0.2= 2.7V, but it was 3.386V. not match! may be there was something wrong.

    2.test range was too small

    H bridge peak current  was 210mA, 1000turns coils, the test range should be 210A, but actually ,when i set the Iprimiary =30A, the output =4.98V,when add current to 40A or 50A, the output has no change. and my Rshunt=5Ω and  R coil=1Ω, Rshunt+Rcoil=6<Vicom/Iccomp=4.4/0.25=16.8Ω.

    maybe it has something with the GSEL1/GSEL0? i select the GSEL1/GSEL0=10, my Gcore=4.145mT/A, and the inductance of the coils , what is the test frequency? i test on 1K Hz and the series mode, it was 0.9mH,Gmod=Gcore*Nwinding/L, what is the unit of inductance? H or mH?

    I have so much questions, thanks very much for your help.

  • Hi Jie Qian,

    GSEL and inductance set AC loop transfer characteristics. Based on your data your DC transfer gain (Nprim/Nwinding x Rshunt x 4V/V) seems to be incorrect. Please check the values of Rshunt and Nwinding. I suspect that you have only 225 or so turns instead of 1000. Also are you sure your primary current is accurate? I would recommend verifying with an ammeter at a low current setting on your source.

    Best Regards,

    Harsha

  • Hi Jie Qian,

    Can you share the dimensions and material of your core. I can simulate to get the expected inductance. The inductance you mentioned 0.9mH seems too low. Our part works with inductance between 100mH- 2 H. At 1KHz test should be fine. Did you test the inductance with it not connected to the DRV421 Icomp pins?
  • Hello !Harsha!

    you are right, the coils have some promblems ,i check the inductance of the compensation coils with LCR, it was only about 10mH, obviously ,the coils was shortcircuit,caused by the sharp edge of the magnetic core. i made  another coils with 1000 copper turns,and the inductance i checked with LCR was 100mH, it seems  more noramal, and the test result please see the figure below, do you think it 's ok? or have something to do to improve the accuracy?

    but i have another question, it is about the temperature drift, in conclution the temp drift was about 0.6% to 0.9%  (about 60-90ppm/℃), so do you have some suggesstions to minimize the temperature drift? by the way, i use the Rm was 25ppm, the same with the user guildline.

  • Hello! javier

    please see my reply, i think the reason was the coils was short caused by the shap edge, and i made a  new 1000 turns coils ,the inductance i checked was about 100mH, seems normal.

    but  i have a new problem, it is the temperature drift. it was about 0.7%-0.9%, about 70-90ppm/℃, was it normal? what i can do to minimize the temperature drift?

    thanks and best regards!

    Jie Qian.

  • Hello! javier

    please see my reply, i think the reason was the coils was short caused by the shap edge, and i made a new 1000 turns coils ,the inductance i checked was about 100mH, seems normal.

    but i have a new problem, it is the temperature drift. it was about 0.7%-0.9%, about 70-90ppm/℃, was it normal? what i can do to minimize the temperature drift?

    thanks and best regards!

    Jie Qian.
  • Hello Jie,

    The drift could be due to the sensor location movement. I do not expect the drift to be that large. Also I would check that that you have the GSel pins are set to operate at the lower value since your coil is about 100mH. In our datasheet we recommend larger than 300mH. It looks like you are using our EVM which has a low drift Rshunt resistor, but also make sure your Rshunt does not drift. Also is this drift, offset or full scale drift?
  • Hello,

    Haven’t heard from you so we’ll assume it’s resolved.