• State
  • Locked Locked
  • Replies 1 reply
  • Subscribers 48 subscribers
  • Views 1309 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

OPA4171 used as comparator

Christian Carlen
Prodigy 160 points

Other Parts Discussed in Thread: OPA4172, OPA4171, OPA171, TLV2374

I use the quadruple OpAmp OPA4172 in a new circuit. One of the 4 opamps is used as a comparator: 

Pin 2: 100k to GND, 460k to Vin (voltage to be compared)

Pin3: 10k to stable 2.5V-reference, 330k to pin 1 (hystereses).

Supply-voltage: 12VDC

The comparator works correctly and has the following measured (and calculated) thresholds of Vin: Output pin1 goes low for Vin >15.6V and output goes high for Vin< 13.6V.

However, if the input Vin is 0V (or not connected), I measure 1.1V at the pin2. This corresponds to a current of roughly 10uA out of pin 2, but bias-currents are specified with some pA in the OPA4172-datasheet. 

My questions:

What is the reason for this high voltage on pin2? 

Why is the current out of pin2 higher than the bias-current specified in the datasheet?

Is the differential voltage pin2 to pin 3 too high for this opamp (I did not find a limiting specification for this in the datasheet)?

Will the OPA4171 work correctly used as comparator, as described above? We do not need speed, but accuracy.

Thank you in advance for your advice!

  • 0
    TI__Guru* 93105 points

    Hi Christian,

    The issues you are seeing when applying the OPA4172 as a comparator can be explained by the datasheet Section 8.2. functional bock diagram. 

    You will note that the Section 8.2 diagram shows back-to-back input clamp diodes across the inputs. These diodes are in place to protect the input devices from breakdown and damage that might occur if too large of a differential voltage is applied. They are fine in normal operational amplifier applications because they don't turn on, but can be a problem when applying the op amp as a comparator. What happens under some input voltage conditions is one, or the other, clamp diodes turns on and provides a path for current flow. In that case, the current will be much higher than what normally occurs as input bias current.

    I put together a PSpice simulation where an external diode has been added across the inputs of the OPA171 model; the simulation model doesn't appear to include the back-to-back input diodes. I used a 1N4148 diode but in reality the clamp diode might have different characteristics; however, it useful to illustrate the point.

    The input to the 100 k resistor has been connected to ground in the example.A current meter in the -In line indicates a current level of about 25.5 uA. About 4.6 uA flows through the 460 k resistor and the other 20.9 uA flows through R2.

    They way to resolve this issue is to use a different op amp, one that doesn't have input clamping diodes. Take a look at the TLV2374. It has similar specifications to the OPA4172 and does not have input clamping diodes. If it isn't precise enough for your application, we can try to identify another amplifier.

    Regards, Thomas

    PA - Linear Applications Engineering