• State
  • Locked Locked
  • Replies 8 replies
  • Subscribers 49 subscribers
  • Views 548 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

Load Regulation

Shaun Wigger
Intellectual 315 points
Other Parts Discussed in Thread: OPA132, TLE2071, OPA141

Normal 0 false false false EN-US X-NONE X-NONE MicrosoftInternetExplorer4

Background:

I have a customer that had a linear power supply designed by an outside house.  The main pass element is a transistor whose base is biased by the OPA132 Op Amp which is an 85C part and not capable of withstanding the internal heat dissipation of the power supply.  The non-inverting input to the Op Amp is fed from a 2.5V voltage reference through a 499R resistor (actual measured value at room temp. is 2.497V).  The inverting input is referencing a voltage divider of 2, 1.00kOhm .1% resistors.  The idea being that the Op Amp is forced to sink current from the base in an effort to maintain the output of the main pass transistor at 5.0000V.

Normal 0 false false false EN-US X-NONE X-NONE MicrosoftInternetExplorer4

The OPA132UA performs this task almost perfectly.  Both inputs are set at 2.497V as one would expect.

When the Op Amp was switched to the TLE2071 part something happened to the inputs.  The non-inverting input is fine and remains at the reference voltage.  The inverting input is 40mV higher at 2.535V.  The result is poor load regulation in that it appears that the steady state operation has the Op Amp in a state that is incapable of sinking enough current from the base.

Problem:

The customer is now evaluating the OPA141 and it is functioning in the unaltered circuit but does not regulate the load as well as the OPA132. No load to full load regulation is 7mV difference and they must achieve 0.5mV maximum.

They are trying to do this without a board spin.

Any ideas???

  • 0
    TI__Guru* 93105 points

    Shaun,

    Please post the applicaiton schematic.

    Thomas

    PA-Linear Applications Engineering

  • 0 in reply to Thomas Kuehl
    Intellectual 315 points

    Please let me know if the quality is sufficient (I had to re-size and crop customer information).

     

    Thanks.

  • 0 in reply to Shaun Wigger
    TI__Expert 4115 points

    Shaun,

     

    A comparison of the parts mentioned previously (OPA132, TLE2071 and OPA141) shows that the OPA132 has the best offset voltage (Vos) and open loop gain (Aol).  A larger open loop gain will allow for more correction in gain error.   The OPA141 has a larger offset voltage and smaller open loop gain compared to the OPA132 and the TLE2071 has the worst offset voltage and smallest open loop gain of the three devices.  So, that is why the customer sees greater error with the OPA141 compared to the OPA132 and even more error with the TLE2071.

     

    Now, even though the OPA132 is specified to 85 C, you can operate the device up to its absolute maximum of 125 C.  Has the customer had issues using the OPA132 at the temperature of the application? 

    Regards,

    Jared Casey

  • 0 in reply to Jared Casey
    Guru 21975 points

    Shaun and all,

    I'm suspicious of possible oscillations. With a power transistor inside the feedback look of a 10MHz op amp there is plenty of opportunity for trouble. Though the various op amps mentioned have similar GBP of around 10MHz, minor differences in gain/phase can cause trouble. The apparent 50mV offset in one case is a strong indicator.

    Have you looked at the output of the op amp with a scope? Oscillations may not be so readily apparent at the output of the regulator.

    Regards, Bruce.

  • 0 in reply to Bruce Trump
    Guru 21975 points

    Shaun,

    I'm straining a bit to read the schematic but took another look...

    The TLE2071 is probably failing in this circuit because its common-mode range to the negative rail is not sufficient. Typical performance of this op amp is about 3V to the negative rail and this circuit is operating with 2.5V. At this point, it may barely behave as an op amp but with high offset voltage.

    There is still a strong possibility of oscillations affecting operation and may account for the regulation issues with the OPA141. The circuit is in unity gain at high frequency due to the 10uF across the 1k feedback resistor (C33?). With the power transistor in the feedback loop there could easily be stability issues. Check for oscillations. They may be present even with the OPA132. Searching for an op amp that makes the circuit look okay could lead to problems down the road.

    There are various possible approaches to compensation of the circuit for improved stability but they would likely require some minor layout changes.

    Regards, Bruce.

     

  • 0 in reply to Jared Casey
    Intellectual 315 points

    I will see if the customer can test the OPA132 at max temp... Also if there are any other parts I am missing that you could suggest... Much appreciated. Thanks!

  • 0 in reply to Bruce Trump
    Intellectual 315 points

    Thanks Bruce... I was scheduled to visit the customer today but unable to due to weather. I will see if they have scoped for oscillations.

    I have the same feeling and the customer might have to do a board spin even though they are trying to avoid at all costs.

  • 0 in reply to Shaun Wigger
    Guru 21975 points

    Shaun,

    Having trouble reading some details that might help...

    What is the part number of the pass transistor?

    What is the value of the the resistor in Q8's emitter (R29?)?

    Bruce