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.

  • Resolved

TLV70012-Q1: TLV70015DSE

Part Number: TLV70012-Q1

Hi,

I've found an issue whereby when a TLV70015DSE regulator is used to supply an op amp mid rail bias the reference 1.5V output is not stable and gradually rises to reach a voltage of approximately 2.18V.

 

On further investigation I found that op amps draw about 0.7uA but believe that the current is sourced from the op amp Vcc via the op amp long tail pair (Vcc= approx 3.0V) that seems to cause the regulator to voltage rise.

 

To try to prove this I replicated the op amp and replaced with a high value resistor of 2M2 to simulate a similar current.  When this resistor was connected to the Vcc rail a similar effect with a rising voltage could be seen.

Note that the same 3.0V rail supplies the input of the TLV70015DSE regulator.

I found that the voltage could be made to return to the correct regulated voltage by connecting a 100k resistor from the 1.5V ref to ground.  My thinking behind this was to provide a return current via ground to the regulator.

 

I would like some help to understand why this is happening as would assume that a regulator would be able to provide a solid voltage source o/p.

As far as I can see the regulator is connected as per recommended in the data sheet and with the recommended decoupling on I/P and O/P of 1uF ceramic capacitor.  An outline schematic  and circuit of the bal to unbal circuit are attached.

 

Thanks

 

Chris

  • Hi Chris,

    Will you help me to understand where you actually connect the LDO in your opamp circuit? Are you using the LDO as a DC reference at the positive node of your opamp?

    Regards,
    Jason Song
  • In reply to Jason Song:

    Hi Jason,
    Since it is a differential to single op amp configuration the LDO was used as the mid rail ref for the positive node of the op amp but connected via 10k resistor R91 as shown in the circuit attached to the original query.  Ie LDO o/p=VREF.
    Hope that's clear.
    Regards
    Chris
  • In reply to Chris Woolrych:

    Hi Jason,

    Just wanted to know if you've any further feedback or answers to my questions on the operation of this part?

    Thanks

    Chris

  • In reply to Chris Woolrych:

    Hi Chris,

    I apologized for the late reply. Will you help me to understand what are the other circuitry that is connected to the positive side of the opamp? When you mentioned the voltage rose to 2.18V, at which point did you take the measurement?

    Could you also try to explain how you replaced the op amp with the high value resistor? Can you draw me a simple schematic for the connections at the output of the LDO?

    Regards,
    Jason Song
  • In reply to Jason Song:

    Hi Jason,

    The circuitry connected to the positive side of the op-amp is as shown in the schematic snipet.   Additionally the VREF connects to an identical op-amp circuit for the other channel.

    The measurement was taken at "Node A" as shown on the updated digram attached

    Also see diagram showing how 2M2 resistor was connected to VREF.

    Thanks 

    Chris

    Vref issue #2.pdf

  • In reply to Chris Woolrych:

    Hi Chris,

    If you only need a voltage reference, instead of using a LDO, voltage reference will be a better option. From the circuit you provided, it does not seem that the regulator needs to drive any load for your application. For more information on voltage reference, you can find it here:
    www.ti.com/.../overview.html

    Regards,
    Jason Song
  • In reply to Jason Song:

    Hi Jason,

    As you mention in this design we only need a voltage reference.

    However since this design was adapted from a previous design  and we are already in mass production we don't really want to modify the circuit to change to a voltage reference just now.

    Therefore we'd like conformation that the modification we made to add a 100k resistor load from VREF (1.5V to ground to provide some loading for the voltage regulator) that we found prevented the voltage from rising is a viable solution.

    Thanks

    Regards

    Chris

  • In reply to Chris Woolrych:

    Hi Chris,

    Yes, you can add a 100KOhms resistor load from Vref to ground to solve the issue. As long as it works for your application, this is a viable solution for your application.

    Regards,
    Jason Song
  • In reply to Jason Song:

    Hi Jason,

    Thanks for your conformation.

    Regards

    Chris

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.