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.


Other Parts Discussed in Thread: TLV3401, TLV3404, TLV3704, TLV3701


I am using TLV3401 to get the output of the Motion Sensor and Sound/Clap sensor as attached.I am giving 3.3 volt to the opamp and the comaprator.

Now Problem when i connect the R32 to my micro controller's (STM32F030R8T6) Wakeup Pin (PA0) my output voltage at R32, R86, R87 & R88 is becoming either 2volt or 0 Volt. But when i remove those resistors and check the output output is 3.3V (I am getting Pulse at those point which is correct)

Can you please help me to figure out why my output voltage is dropped down when i connect it to the Wakeup pin of the controller. Same time when i connect it to any other pin of the micro controller it is not dropping down, in short if i connect the output of the Motion circuit and the SOUND/CLAP sensor to wakeup pin of uC it dropped down and i am using it to wakeup the uC.


Krunal ShahExtracted pages from LB_PERIPHERAL_1.1.0.pdf

  • Hello Krunal,

    I wouldn't expect the kind of circuit behavior you describe having with the TLV3401. It is a nanopower comparator that has very limited output pull-down current capability and this odd result may be due to an output drive issue. If the microcontroller wake-up pin draws enough current when the TLV3401 U17 is high enough, then the voltage drop across 47 k, R71, could be large enough to prevent the output from pulling above 2 V. I checked the microcontroller wake-up pin specifications and no current level is specified.

    The even odder issue is why the output of U15-A, U15-B and U15-C would be affected by any loading at the output of U17. They are separate comparators from U17 that use a different Vcc supply, +3.3V_CLAP. My suggestion is you observe and record with a DSO the U15 outputs, U17 output, VCC_MOTION, and +3.3V_CLAP, with and without, the microcontroller wake-up pin connected.

    • Have you observed this behavior on more than one baord?
    • Did you obtain U17 from an authorized TI distributor?
    • What comparator is used for U15?
    • You comment, "But when I remove those resistors and check the output output is 3.3V." R32 and R86 - R88 are all in series with the output. Removing these resistors disconnect the circuits that follow them; that includes the microcontroller wake-up pin. Is that correct?

    Let me know what you find.

    Regards, Thomas

    PA - Linear Applications Engineering

  • Dear Thomas

    Please find the attached presentation showing the output at R32 when connected to uC and when it is not connected to uC. All the output are diffeerent and connected to the Differnet Diodes and output of diode is connected togehter with PA0

    Answers to your questions as below

    1) Yes we Have three boards and all are behaving in same manner.

    2) Yes we bought is from Digikey

    3) U15 is the TLV3404IPWR

    4) I have diode which are connected to resistor and they are making them isolate when i remove the diodes. Please see slide # 4 of the presentation.

    Please let me know if you need more information. I will capture the wave form for the output of U15 and the +3.3V_CONTROL and send you soon.


    Krunal ShComaprator Output.pptxah

  • Hello Krunal,

    Thank you for all of the good information and answers to my questions. I gather from your explanation that the output voltage levels goes unexpectedly low when a load connection is added to either the output U17, or U15 outputs. The only normal circuit occurance that could allow that to happen is if a relatively low resistance load were connected from an output to ground when the comparator output is in high state. The TLV3401 (TLV3404) has an open-drain output and depends on a pull-up resistor to pull the drain up the the high level which is the Vcc level.

    If a load is connected to the comparator output when it is in high stae the current required by the load will flow through the pull-up resistor. In the case of U15 the pull-up is 200 k-Ohms. You mention in that case the output voltage drops from 3.3 V, down to 0.4 V, when the D7/wake-up load pin is connected. That would indicate that a load current of 14.5 uA is being drawn through the 220 k-Ohm pull-up resistor. That sounds like a high current for an input current, but would be the equivalent current had by adding a 27.5 k-Ohm resistor from the comparator output to grounf with the 220 k-Ohm pull-up in place. It is only when the comparator output goes low that drain current flows and causes the drain voltage level to move close to 0 V. The current throught he 220 k-Ohm resistor would be about 16 uA in that case.

    You might try taking one of the U15 outputs, without the microcontroller input connected, and add a 27 k-Ohm resistor from it to ground. If the output drops from 3.3 V to about 0.4 V, then that indicates that the microcontroller input is pulling the comparator output down.

    You show that in the case of U17 that the output pull-up R71 is 47 k-Ohm, and the high output level drops from 3.3 V to 2.1 when the Schottky diode, plus microcontroller input, are connected to its output. That would indicate that the load is drawing about 26 uA. Reducing R71 from 47 k-Ohm, to 22 k-Ohm, should allow the output high-state to move up to about 2.7 V if the current stays the same. The main effect will be more current through the pull-up resistor when the U12 output goes low.

    Here are a couple of thoughts;

    The TLV3401 and TLV3404 have pin-for-pin equivalents that have push-pull outputs, the TLV3701 and TLV3704, repectively. Their output(s) can sink and source current, whereas the TLV3401 and TLV3404 outputs can only sink current. The output will actively pull up and it will source current if any current is being required by the load. I am not suggesting this as a permamant change for now, but it would help answer the question if the low output levels are due to the external load pulling down the output. The pull-up resistors could be left in, or removed, from the circuit for this test.

    I am not familiar with the microcontoller that you are using. Is it possible that it has programmable pull-downs at the inputs? Some of TI's MSP430 microcontrollers have this feature.

    Regards, Thomas

    PA - Linear Applications Engineering

  • Dear Thomas

    Thank you so much for details, We have checked in detail with Micro controller's Reference Manual and found that when we configure PA0 as as wake-up in it will work as input pull down pin and internal pull up/pull down resistor is typically 40kOhm.

    So when we connect our Pull up resistor R71/R77/R79/R83 to it via diode it becomes voltage divider at PA0/WAKE-UP which is causing to drop the voltage to 2V.

    Now to wake up the Micro-controller we need 1.8 or greater voltage so we made R71/R77/R79/R83 as 10k so our output at PA0 becomes 2.64V which is enough to wake-up the Controller.

    So no Problem from TLV3401 or Micro controller its pull down resistor that was creating this problem.

    Thank you so much for your help.


    Krunal Shah
  • Hello Krunal,

    Thank you for the update. It is good to know the TLV3401/TLV3404 were not causing the issue.

    A suggestion is you make sure the TLV3401/TLV3404 does pull low enough when the is in the low state. Reducing the resistance of the pull-up resistors will require the output MOSFET to conduct more current when the output is low. Datasheet Figure. 3, LOW-LEVEL OUTPUT VOLTAGE vs LOW-LEVEL OUTPUT CURRENT, should provide a pretty good idea how close the output approaches 0 V when it is sinking current. The 2.7 V Vcc used in the graph is only 0.6 V lower than the 3.3 V you are using. The sink current capability increases with increased Vcc which works in your favor.

    If you are finding the MCU is recognizing the input levels without problems, then they are within its acceptable VIH and VIL ranges.

    Regards, Thomas

    PA - Linear Applications Engineering