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.

INA301-Q1: Problem with ALERT Pin: Goes low before ILIMIT is reached

Part Number: INA301-Q1
Other Parts Discussed in Thread: INA301, STRIKE

Hi,

We are using INA301-A1 in one of our designs. During design verification it is observed that the ALERT Pin transitions from HIGH to LOW when OUT pin voltage is lesser than LIMIT. The resistor value at LIMIT pin is 18k. The INA301 is powered by a 3.3V supply. 

Design: 

VLIMIT (typ) = 18k*80uA = 1.44V

Gain = 20

Theoretical Input voltage for ALERT to transition to low = 1.44V/20 = 72mV (Input voltage in increasing directionINA301.pdf

Practically observed Input voltage for ALERT to transition to low = 63mV

Can you please help to resolve the issue?

Best regards

Ashwin

  • Ashwin,

    Would it be possible to see a full schematic of your INA301 setup?

  • Hi Carolus, 

    In addition to the circuit diagram, 

    1. We are using 4 shunts 5 ll 5 ll 3 ll 3 = 0.9375uOhm (effective). The kelvin sensing trace is taken from the center of the resistors combination in the PCB

    2. The shunt resistor is located at a distance of 50-60mm from the amplifier. 

    3. But the ALERT transitions at a lower voltage than the set LIMIT voltage.

    Best regards

    Ashwin.

  • Ashwin,

    I'm assuming that's .937.5mohms (not uohms)?

    I think a contributing factor here is the input filter. As the device output ramps, additional bias current is drawn over the input lines (see figure 40 of the datasheet). As this bias current grows, the voltage drop you will see across the resistor grows in tandem and causes additional error from the shunt to what is actually measured at the pins of the device. Typically this causes the ALERT pin to trigger later, and not by this amount of error though.When you say you are seeing the ALERT pin pull low at 63mV, are you measuring this at the shunt, or at the pins of the device? Can you measure both to see if there is a voltage discrepancy across the input filter resistor? Also, can you probe the output pin of the device with the known input voltage to make sure the device is operating linearly. Finally, can you probe the limit pin itself, to ensure that the reference voltage being fed to the comparator is correct?

     

  • Hi Carolus

    Amplifier gain = 20 (theoretical); 20.4 (practical)

    shunt resistor: 937.5 uOhm

    Current(A) On the Shunt (Before series resistor)mV After the series resistor, on device input pins (After Filter)mV Vout - practcal(V) Vlimit - practical (V)   Vout - Theoretical
    10 11.82 9.9 0.2015 1.437   0.191352
    20 21.717 19.921 0.4082 1.442   0.382704
    30 31.362 29.926 0.6144 1.446   0.574056
    40 40.135 40.017 0.821 1.451   0.765408
    50 50.385 50.267 1.029 1.455   0.95676
    60 60.535 60.211 1.236 1.46   1.148112
    63 63.108 63.12 1.285 1.46   1.2055176
    70 72.93 70.89 1.448 1.465   1.339464

    Best regards

    Ashwin

  • Ashwin,

    A few things strike me as off here:

    - The voltage you are reading at the limit pin is changing with increasing current, which shouldn't happen. This voltage is generated via a constant current source and should remain constant regardless of input

    - The gain of the part seems linear, but a bit too high in terms of error. The part should be exhibiting a .1% max. 

    What is the application here? I noticed the filter you are placing on the input is massive (7500Hz), and then you also filter the output. What is the nature of your input signal here? Is there some switching or AC characteristic that you're trying to suppress here? Could you probe the output pin with an oscilloscope to see how stable the output signal is here? The input signal as a reference would also be helpful.

    Also, have you tried replicating this over multiple devices, or just a single INA301?