• State TI Thinks Resolved
  • Locked Locked
  • Replies 2 replies
  • Answers 1 answer
  • Subscribers 48 subscribers
  • Views 969 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.

OPA2365: Analog Pulse Width Modulation

Rock Su
Intellectual 2220 points
Part Number: OPA2365

Dear colleague,

In our http://www.ti.com/lit/ug/slau508/slau508.pdf,

Customer asked me:

1. Why output error is larger when input voltage is in 1.3V-2V?

2. For slowing down the output error, how should we do ?

Best Regard,

Rock Su

  • 0
    TI__Genius 15565 points

    Hello Rock,

    This increase in error is due to the tolerances of the components used. While ideally, the equation for the transfer function would be Vout = -Vin+2.5 , it is explained that the gain is measured to be -0.9978V and the offset 2.4973V. The equation then becomes Vout = (-0.9978*Vin)+2.4973. A graph of the percent error between these two equations is shown below:

    As you can see this graph very closely resembles figure 17 of the TI Design, with the exception of the error being strictly positive in the figure. In order to minimize this error, you would need to further decrease the tolerances of components used so that they are as close to ideal as possible. 

    Best,

    Hasan Babiker

  • 0
    Guru 140200 points

    Hi Rock,

    the errors of this circuit can very easily be simulated. In the following simulation R1, R2, R3 and R4 shall be free of any error:

    Everything is fine. The error of output voltage is almost zero.

    But when you assume production tolerances for these resistors the error can become relevant. In the following simulation all resistors R1, R2 and R4 shall be perfect. Only R3 shall have an error of +1.0%:

    You can see that at an input voltage of -2.00V the output voltage shows an error of -0.7%. At an input voltage of 0V the error is -0.5% and at an input voltage of +2.00V the error is +1.5%.

    TIPD108_Error Amplifier.TSC

    You can easily simulate the errors of the other resistors as well and estimating the worst case error:

    The errors of the resistors are +/-1% here.

    With zoom:

    TIPD108_Error Amplifier_worst_case.TSC

    The simulations underline how important it is to use matched resistors showing low drift in this circuit.

    Kai