• State Resolved
  • Locked Locked
  • Replies 3 replies
  • Subscribers 48 subscribers
  • Views 257 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.

OPA2810: Maximum differentiual input voltage/ input current

ArTzy
Genius 15599 points
Part Number: OPA2810
Other Parts Discussed in Thread: NE5534

Hi Experts,

Seeking your assistance:

I build an transimpedance amplifier circuit with the OPA2810. In a special condition, the inverting input pin is (quasi) connected to the positive supply voltage of 12V. The non-inverting input pin is connected to ground over a 2kOhm resistor (see picture of the simplified schematic).

I assumed (based on datasheet section 8.3.1 (see below)) that the differential input voltage is clamped to 7V, so the non-inverting input pin should have a potential of about 5V (=12V-7V). But the measured voltage (real measurement, no simulation) of the non-inverting input pin was only 0.44V (I_In = 235µA).

So on one hand the input currents are far less than the maximum rating but on the other hand the differential input voltage is greater than the maximum rating. Will this situation damage the chip or is it already damaged?

OPA2810 Datasheet statings:

Absolute maximum ratings:
- Differential input voltage: +-7V
- Continuous input current: +-10mA
Section 8.3.1:

“The OPA2810 also includes input clamps which enable maximum input differential voltage of upto 7 V (lower of 7 V and total supply voltage). […]

The input bias currents are also clamped to maximum 300 μA, as Figure 57 shows, which does not load the previous driver stage or require current-limiting resistors (except limiting current through the input ESD diodes when input common-mode voltages are greater than the supply voltages). 

Thank you in advance for your response. 

Regards,
Archie A.

  • +1
    Guru 140200 points

    Hi Archie,

    the +/-10mA only refer to the ESD clamps which play no role in the way the OPA2810 is tortured here. And the datasheet doesn't say that the differential input voltage is clamped to +/-7V, but only that the internal clamp circuit survives a dfferential voltage of up to +/-7V. Applying 12V across the inputs clearly violates the absolute maximum ratings and may result in a damage.

    Kai

  • 0 in reply to kai klaas69
    TI__Genius 15599 points

    Thank you, Kai.

    So what is the 300µA input bias current clamp really for? Is it just to reduce the load of any previous stage if the inputs are within the supply voltages and the differential input voltage is between 4V and 7V or -4V to -7V?

    I mean if the differential input voltage is lower, the input current is lower anyway and if the differential input voltage is higher, the OPA2810 can be damaged.

    Regards,
    Archie A.

  • 0 in reply to ArTzy
    Guru 140200 points

    Hi Archie,

    section 8.3.1 of datasheet explains very nicely the benefits of this protection scheme.

    Let me add my thoughts:

    When a fast HF-OPAmp shall be able to work with high supply voltages, the inputs have to be protected against too high differential input voltages in order to protect  the vulnerable FETs of the input stage. Usually, a diode clamp is used between the inputs. It looks like the OPA2810 contains two diodes in series to limit the differential input capacitance added by the diodes' junction capacitances. To further limit the junction capacitances, the protection diodes must have rather small pn-junctions. And this will only allow rather low maximum currents through these protection diodes. From other OPAmps like the NE5534 for instance we know that currents up to 10mA are allowed. But the OPA2810 may only allow somewhat lower currents, maybe 2...3mA.

    The usual way to protect the diode clamp against too high currents is the add of a current limiting resistor in series with the inputs. But this technique has two disadvantages. It erodes the phase margin (especially unwanted in a HF-OPAmp !) and adds noise. According to figure 57 of datasheet, a current limiting resistor of R = (7V - 1.5V) / 2mA = 2k75 would be needed to limit the input current to 2mA at an differential input voltage of 7V. But 2k75 would add a voltage noise of 6.7nV / SQRT(Hz), which is more than the OPA2810 produces itself (6nV/SQRT(Hz)) !

    So, adding a current limiting resistor is not a good method to limit the current through the diode clamp of a low noise HF-OPAmp, because it adds noise and ruins the phase margin. Also, you don't want to have too much heat dissipation in the input stage caused by the diode clamps, because this may effect the symmetry of input stage and by this degrade the input offset voltage.

    TI has found another method which does not need big current limiting resistors. I guess that a current limiting circuit with FETs is used. Keep in mind, that a resistor of R = (7V - 1.5V) / 300µA = 18k would be the equivalent of a 300µA current limting...

    Kai