• State
  • Locked Locked
  • Replies 1 reply
  • Subscribers 51 subscribers
  • Views 419 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.

TLV2454 exposed to 12V input

Don Hotchkiss
Prodigy 45 points
Other Parts Discussed in Thread: TLV2454, OPA140, TLV2451

Hello,

We have a 12V circuit that feeds into the 5V powered TLV2454 which is being used as a comparator.

In normal operation, the signal would stay within the 5V rail, but in an abnormal condition, the input could go to 12V.

If I have a input current limiting resistor, and had to allow a fault that took the resistor to 12V, would/can the TLV2454 withstand the input current through internal path(s) to clamp to its own supply rail?

I need to stress that this is NOT a normal condition…it is a condition that a customer could encounter though, if he were to disconnect a particular connector.

Also..I cannot just re-configure the circuit.  Without going into a lot of detail…I have a 4.7V zener clamping the input, but the leakage of the zener (only about 15uA) is quite enough to cause a voltage error via the limiting resistance that feeds the OP Amp.

I would like to just remove the zener and let the input be limited by the resistor.

The wheels are rolling towards production and a minor component change is my hope.

 

Here is an excerpt from the TI OPA140 data sheet:

 

I am looking for similar info..at least some assurance that there is a clamping mechanism (like for ESD) and it certainly can be less than 10mA.

Thanks,
Don

 

  • 0
    TI__Guru* 93105 points

    Hi Don,

    Unlike the OPA140, the TLV2451 does not have ESD current steering diodes connected back to the V+ and V- rails. Those diodes are normally off in the OPA140, but can handle 10 mA should they be turned on inadvertently by an electrical overstress event (EOS). The TLV2451 data sheet makes no mention of a maximum input current like 10 mA, so there no information to tell us what a safe level is.

    I have some bench data I took at few years ago on the TLV2451 input current. What it shows is that once the non-inverting input - in a buffer configuration - is driven about +1.4 V above V+ the input current starts to climb rapidly. That is an indicator that damage might occur if the voltage is increased further. Thus, if +12 V were applied to the input when the TLV2451 V+ is +5 V the current could be very high and damage the device.

    The TLV2451 uses complementary NPN/PNP differential pairs at the inputs and we have to avoid breaking down any transistor junctions in the pairs. Should that happen the amplifier could be permanently damaged. However, if the input current is kept very low during a potential input EOS event the amplifier will likely be okay.

    Taking the set-up where the TLV2451 V+ is 5 V, and knowing the input begins to draw significantly high current when the input is taken about +1.4 V above V+, the maximum input voltage would be +6.4 V for that case. Then, with a series resistor added to the input and +12 V applied at its input end, 5.6 V would have to be dropped across it to keep the input from going above +6.4 V.

    The maximum safe input current is unknown and not something I'd be able to find documented on a 15 year old amplifier. Based on my original measurement a current of about 100 uA maximum should be okay. Taking the 6.4 V drop across the resistor, and 100 uA, the series resistor needed would be 64 kohms. That is a high resistance and you wouldn't likely want to use such a high value in an amplifier circuit because of the resistor noise. However, you mentioned the TLV2451 is being used as a comparator and the noise performance may not be nearly as critical.

    Do note that the TLV2451 has an input bias current of about 1 nA typical, and 5 nA maximum. A current of 5 nA flowing through the 64 k resistor will add, or subtract, 320 uV to/from the inherent voltage offset. Also, TI does not list a maximum input current specification. Thus TI cannot warrant the device when used in a service where it operated outside of the data sheet specifications.

    Regards, Thomas

    PA - Linear Applications Engineering