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.

TL431 Reference Voltage

Other Parts Discussed in Thread: TL431

Hello all

Please could you advise on this.

We use a TL431 (TI part TL431AIDBZT, not sure what the “T” represents) as a voltage sensor.  There is a 33k resistor (also considering 27k, actual value T.B.A.) between pin 2 (Reference) and pin 3 (anode, also ground in our circuit).  There is a 1M resistor from pin 2 to a voltage rail whose level we need to detect which is over 70V (it can rise to 350V or so, hence the high resistor value to minimise power dissipation).

According to the data sheet, the “AI” part has a reference voltage of 2.47V to 2.52V, and there is a thermal drift of 34mV for the SOT23 package which we use.  Thus the lowest I expect the device to switch on is 2.47-0.034=2.436V.  Yet at room temperature (23°) the device switches on with 2.05V on the reference pin (measured with an oscilloscope).  This is around 16% lower than expected.

Please could someone advise urgently what I have missed.

Thank you.

  • Might be useful to know as well that the Cathode is pulled to +15V with a 10k resistor which provides 1.5mA which is above the 1mA minimum Cathode current required.  Although the problem manifests before any Cathode current flows.

    In addition, there are no capacitors on this circuit and specifically the load capacitance across Cathode-Anode is parasitic only so is in the left side stable region indicated on the data sheet, but I plan to place a small (nF range) capacitor between Reference and Anode to filter the incoming signal.

  • Brian,

    The TL431 has no VCC pin. So cathode is like a VCC pin and Reference pin is a start up control.

    For this chart consider Vref as X axis and Y axis is still cathode current.

    It is normal for cathode to draw current before 2.5V.

    When you said the device was on with 2.05V on reference. was the output near 2V (1.3mA)?

  • Thank you, Ron

    The Cathode is not connected to the Reference, neither directly nor via a potential divider.

    The Reference pin is driven via a potential divider from a higher DC voltage which we need to sense and determine whether or not it's above a certain value.

    The Cathode is pulled with a 10k resistor to +15V and drives a FET gate.

    When the Reference voltage attains ~2.05V, the Cathode switches on and reduces its voltage from +15V to ~0V.

    The circuit works in principle, but at the incorrect voltage.  It switches at ~2.05V rather than nominally 2.495V.

  • Brian,

    Here are the expected result for a similar setup. Note that VKa starts drops from 15V a little early. The final low voltage is near 2V (not 0V).

    If the output drops to near zero, it was not the TL431 that made that happen. 

  • Thank you, Ron

    It is not the Cathode voltage that bothers me.

     In your circuit the blue line drops when the green line attains ~2.50V which is as per the data sheet.

    In my circuit the blue line drops when the green line attains ~2.05V, which is unexpected.  That is about 18% below where it should be, which is cause for concern.

    An obvious possible explanation is a damaged IC, but I've eliminated that option through replacement of the IC with due respect for precautions against static damage, and the problem persists reliably.

  • Brian,

    I built the circuit with 1M, 33k input divider and 10k to 15V.
    It worked as I predicted.

    There is something in your circuitry that isn't in my test that pulls cathode voltage to near zero. The TL431 can't do it. A dead 15V or something else on the node is responsible.

    Sweep.xls