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.

AFE3010: AFE3010: Total Reaction Time Estimate

Part Number: AFE3010

Hello,

The datasheet for the AFE3010 suggests that for high fault currents, the reaction time is approximately 10ms.

Is this 10ms measured as the time it takes from fault current occurring until the SCR gate is fired or does the 10ms also include the application circuit solenoid opening time?

Thanks!

  • Hey David,

    The 10ms time does not include the mechanical delay time for actuating solenoid, although this required time (~5ms) was considered when testing for UL943 compliance.

    Sincerely,

    Peter

  • Hello Peter, 

    Thanks, just one more thing, is the 10ms internal in the AFE3010, or is it due to the datasheet application circuit solenoid coil having a diode in series with it, so it cannot conduct during half of the VAC sine wave? (8.33ms)

    I am guessing but:

    I think the diode in series with the solenoid coil is there so that the solenoid coil won't conduct for half of the AC sine wave.  During the half period (8.33ms) where the solenoid can't conduct, the AFE3010 applies a voltage to the SCR anode.  The AFE3010 then triggers the SCR gate, turning on the SCR, dropping the SCR anode voltage.  The AFE3010 checks if this voltage drop occurs.  This is how the SCR = OPEN and SCR = SHORTED failure modes can be detected by the AFE3010.  In the half periods (8.33ms) where the solenoid coil is not blocked by the diode, the AFE3010 can monitor the voltage on the diode - SCR anode node to verify if the solenoid coil is open circuited or not.

    The reason I ask is because, I am trying to figure out how much of time budget there is.  Is the worst case reaction time, 10ms for AFE3010 + 5ms for solenoid contact opening + 8.33ms for worst case half sinewave blocked where the solenoid can't be energized = 23.33ms?

  • Hey David,

    You are correct. The worst-case is when SCR is being fired (turned on) right at the start of the negative line phase. Because of this, the SCR won't conduct current until negative phase is over and line transitions to the positive phase, which theoretically last 8.33ms. So I think this is a robust way to approximate worst-case trip time given you have an accurate actuation delay time for the solenoid.

    Real test data has shown trip times (with an additional 7ms added to account for solenoid actuation) to be easily less than 20ms for high ground faults. The reason is that really the device only needs to see one max comparator  tripped (Vout > 4.8V or Vout <0.2V) for ~256µs + register one line zero-crossing transition before it fires SCR. Thus, the typical trip time should really be 8.33ms, because the device will see this transition by then. In addition to this, if the Vout is moving and tripping these comparators with +/-5° of the zero-crossing, due to amplifier or transformer saturation, the device should immediately trip and overall bring down the average trip time for higher leakage currents.

    I hope this helps.

    Sincerely,

    Peter

  • Thanks, this was helpful.  I ran some tests with the AFE3010 eval kit and a 2.2kohm (55mA) fault current.  I saw trip times from the fault current beginning to when the SCR gate was triggered as small as 2ms and as long as 10ms.  The 10ms worst case was when the fault began on the VAC zero crossing of the negative going (non-conductive) time, so about 1.67ms for the AFE3010 and 8.33ms for the sinewave half period.  This very fast timing is helpful.