Post Go back to editing

Regarding current limit accuracy in MAX32010 FV mode

Category: Datasheet/Specs
Product Number: MAX32010

Using the MAX32010 evaluation board, I checked the current limit operation when operating in FV mode.

With 100Ω as load
・CLH=0.5V, CLL=-0.5V
・CLH=1.5V, CLL=-1.5V
・CLH=1.8V, CLL=-1.8V

and gradually increased the VIN voltage.

The current limit is effective, but it deviates considerably from the set value.
(Because of the 100Ω load, it was 0.28V when CLH=0.5, about 0.76V when CLH=1.5V, and about 0.9V when CLH=1.8V.)

What can be the reason for the discrepancy?

(The image shows the 200μA range, but the measurement was performed in the 20mA range.)



The image shows the 200μA range, but the measurement was performed in the 20mA range.
[edited by: UBOTI at 11:26 AM (GMT -4) on 30 Jun 2023]
  • Hi ,

    Sorry I meant to look at your issue today but did not get a chance. Let me look at it and get back to you early next week.

    Regards,

    Hooman

  • Hi ,

    I believe you've made sure that the current limit is enabled (D10 CLEN high, and D8 /HIZFRCB high) and your plot shows values which are higher than expected. For example, in the 20mA range (Rsense = 50ohm), going by the formula below, you'd expected a current limit of 7.6mA (VCLH = 1.8V), whereas your plot shows 9mA current limit (about 15% error).

    Unfortunately, I could not locate any accuracy information in the MAX32010 datasheet for the current limit. And also unfortunately, I could not get my MAX32010 EVK to run (I'd planned to duplicate your experiment and see what values I get).

    Judging by these EC table values, I'd expected an accuracy of about +/-10%:

    Because my EVK is still not running, maybe instead of the 100ohm load, you can use a DMM (Ammeter setting) to ground and see what the current limit would be when shorted to ground this way? Just to see if that makes any difference?

    Also, please make sure that J2 on the MAX32010 is shorted (to tie DUTGSNS to GND) to make sure the DUT Gnd sense is the same as GND on the board.

    In the meantime, I'll reach out to other folks and see if I can come up with any information on the expected current limit accuracy and share any findings I might get.

    Regards,

    Hooman

  • thank you for your reply.

    ・When J2 is shorted and when it is open
    ・When the load is 100Ω and ammeter (internal resistance 10Ω)
    Measured under the above conditions.


    I assumed that current detection was performed on the H side, and thought that there was no difference due to load impedance.
    However, I was able to confirm some difference between 100Ω and ammeter.

  • you should set the Iclamp according to Note 8 in the datasheet to make the FV accurate. 

  • Thank you for contacting me.

    I think that the setting of CLH=1.8V, CLL=-1.8V in FV mode satisfies the conditions of Note 8, but is it a misunderstanding?

  • Yes, I encounter the same problem as you. I think it is the IC's drawback.

    Rd=5KΩ @ 200uA range, CLH=1.8V means Iclmp=9uA<12.5%*FSR=50uA(first condition of Note 8), so FV is not linear. so you should set IclmpH>50uA at least at 200uA range.

    and I think the clamp acts near 20% less than the set clamp value, so the FV value you set at this time is meaningless.

    ---all the above are tested by myself and don't represent ADI.

  • thankyou for the explanation. 

    In my case, for example CLH = 1.8V, the clamp current setting is about 7.6mA from the datasheet formula.
    7.6mA for the 20mA range is 38%, so it satisfies the condition of Note 8.