• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 1 answer
  • Subscribers 62 subscribers
  • Views 695 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.

Original question:

TPS706: TPS706 Iq is very high when the input voltage is lower than the fix output voltage

TPS706: TPS706 reverse current blocking threshold

Mark Pieper
Expert 4430 points
Part Number: TPS706

Hi. If Vout is reinforced by another regulator, and Vin is allowed to fall below the Vout voltage, I expect that the reverse current blocking will cause the LDO to turn off.

Other posts have suggested that this isn't the case. Consider this situation:

TPS70633 operating from USB input with EN tied to VIN. The 3.3V output is also reinforced by another regulator. When the USB is disconnected, the load on VIN is very low, maybe <1mA.

As VIN falls from 5V to ~3.3V, the regulator goes into dropout and Vin~=Vout, connected by a FET that is ON. Small current will feed from Vout to Vin, reinforcing the input voltage and the EN signal. In this case, will the regulator detect the "reverse bias" across the Vout-Vin and turn off?

What is the reverse voltage threshold (unspecified)?

Thanks.

  • 0
    TI__Guru 55940 points

    Hi Mark,

    You are correct that the reverse current protection will be active in your example situation until the output falls below 1.8 V.

    Keep in mind that this reverse current protection is only active when the voltage on the OUT pin exceeds the voltage on the IN pin.  The leakage current from the output to the input will be typically 10 nA. 

    If the voltage on IN does not exceed the voltage on OUT, there is not a reverse current situation.  If the voltage at the EN pin is above the enable threshold, the LDO will be enabled and operate in dropout.

    Very Respectfully,

    Ryan

  • 0 in reply to Ryan Eslinger
    TI__Expert 4430 points

    Hi Ryan,

    Thank you for the answer, but I don't think it answers my question. The linked thread implies that with Vin slightly less than Vout, the TPS706 does not always enter into a low Iq mode. In my case, Vin will disappear and decay very slowly. at the same time Vout will be held up to a voltage greater than 1.8V. The LDO will approach dropout, the pass element will saturate, and there will be a resistive connection between Vout and Vin allowing reverse current to flow. The internal circuit appears to try to detect that reverse voltage and turn off the pass element. My concern is that the very small reverse current (~100uA) that is allowed by my circuit will not cause enough delta-V to cause this circuit to trip.

    What is the reverse trip (Vout-Vin) threshold? That parameter isn't documented in the data sheet.

    Thanks,

    Mark

  • 0 in reply to Mark Pieper
    TI__Guru 55940 points

    Hi Mark,

    The referenced post is discussing the increase in current going to GND through the LDO rather than current flowing out of the IN pin.

    Reverse current can only flow when Vout exceeds Vin in any LDO.  As Vin approaches Vout, Ignd will increase as the LDO goes into dropout because the LDO requires more power to drive the pass element fully on.  

    Once Vin drops slightly below Vout, the path for reverse current through a PMOS LDO without dedicated reverse current protection is through the body diode of the pass FET.  I drew the body diode below in red.  The TPS706 reverse current protection is set so that it will detect that Vout is less than Vin before the forward voltage of the body diode is reached.  Therefore, for TPS706 reverse current will not flow because either the switch is open or the forward voltage of the body diode on the pass FET has not been reached.

    As a result, in your application the 100 uA will not flow out of the IN pin so long as Vout remains above 1.8 V.  The 1.8 V requirement is so that the internal amp can be powered.  This will take ~90 nA which is why 100 nA will flow into the OUT pin but only 10 nA will flow out of the IN pin while the reverse current protection is active.

    Very Respectfully,

    Ryan

  • 0 in reply to Ryan Eslinger
    TI__Expert 4430 points

    Hi Ryan,

    I've completed testing the TPS706 regulator. With Vout reinforced by an external voltage at levels slightly below the regulation setpoint, when Vin is removed the device DOES NOT turn off. Please see the below scope capture.

    What is the reverse bias turn off threshold voltage?

    Below upper trace is VIN, lower trace is VOUT (3.3V), There is a 10K resistor between VIN and GND.VOUT is reinforced at 3.0V from another power supply (diode coupled). You can see that VIN does not fall, and that it is held up by the output voltage.

  • 0 in reply to Mark Pieper
    TI__Guru 55940 points

    Hi Mark,

    Unfortunately the reverse bias turn off threshold is not a tested parameter.  The reverse current protection is designed to protect the LDO from reverse current; therefore, it is designed to be enabled before the body diode of the main pass element turns on.  As such it should be less than the body diodes forward voltage.  As this is an untested parameter, it is the responsibility of the user to perform sufficient qualification testing to ensure reliable operation in their specific application.

    If the output is externally biased with a voltage greater than the regulated voltage when Vin is removed, the reverse current protection will engage disconnecting the main pass element from OUT.  When this occurs reverse current flowing in the OUT pin and out the IN pin will be limited to 10 nA (typ). For example if the regulated Vout is 3.3 V but biased to 3.9 V, the reverse current protection will engage.

    If the output is externally biased to less than the Vout set point, it is possible that additional current may flow into the OUT pin; however, it is important to note that some of this current will still be flowing to GND as it is powering the reverse current protection block.  Depending on the Vout to Vin differentiation, additional reverse current (current flowing out the IN pin) could flow holding Vin up at the output's biased voltage.  This is what appears to have occurred in your scope shot.  If you are able to increase the bias voltage on Vin, you should be able to avoid this condition.

    Very Respectfully,

    Ryan