• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Subscribers 31 subscribers
  • Views 404 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.

DRV5055-Q1: Standard error and batch errors

Felix Röpke
Prodigy 10 points

Part Number: DRV5055-Q1

Tool/software:

Hello here smart people,

we are currently working with the DRV5055-Q1 hall sensors for the detection of a ISO26262 safety relevant lock position. For this we are doing a error analysis on the possible output voltages where we summed of the tolerances of the DRV5055-Q1's datasheet up to 5%. Is this a realistic error to assume or do you have some measurements for this sensor available, showing the standard error distribution on the DRV5055-Q1 outputs gair and offsets or the error on the production batches?

best regards

  • 0
    TI__Mastermind 24490 points

    Felix,

    Welcome to E2E.  Unfortunately we do not have any published histograms showing the distribution of these specifications.  I'm not sure I understand what you mean by summing the errors up to 5%.  I think you are saying that  you using 95% of the min/max range and then applying all errors into a single calculation.

    This is likely more realistic than combining all of the extreme distribution values on top of each other as the probability of obtaining a device that is an outlier in every category simultaneously is extremely low. Since the errors tend to be Gaussian in nature, I would typically combine similar terms using the root of the sum of squares.  

    Thanks,

    Scott

  • 0 in reply to Scott Bryson
    Prodigy 10 points

    Hey Scott, 

    the 5% I assumed is the summed up errors from page 6 of the datasheet to take the worst-case into account. My bad for not specifying this. 

     

  • +1 in reply to Felix Röpke
    TI__Mastermind 24490 points

    Felix,

    This makes more sense!  Thanks for clarifying.  

    So, since these all affect sensitivity gain, it is possible to combine them using the method I mentioned above:

    Sqrt ( 1%2 + 1%2 + 2.5%2 + 0.5%2) = +/-2.915%

    This is the typical method for combining statistical distributions.  You could use the 5% to be very conservative in your approach, but it is much more likely that your actual results will fall within 3%.  

    Thanks,

    Scott

  • 0 in reply to Scott Bryson
    Prodigy 10 points

    Hey Scott, 

    thanks for the answer. This helped us quite a bit :)