ADE7880 evaluation board - maximum line voltage

Hello,

I have a question about the ADE7880 evaluation board. I am measuring 230VAC input line voltages. However the resistors at the system inputs (R26, R27, R28) are only designed to support 150V normally (ERJ-6ENF1004V).

Is this a design error? Why is it still good in my measurement? Do I have to change them?

Thank you in advance,

Maxime

Parents
  • +1
    •  Analog Employees 
    on Jan 13, 2021 1:26 PM 1 month ago

    The 7880 I think has a single 1M resistor. This board was designed for bench eval with an option to work with 240V. The single resistors should have been a min of 2 (500K) resistors . In the actual application 3 to 5 resistors in series are common.  One other issue with single resistor is the capacitive leakage around the resistor at 50Hz is a percentage of the 1Meg resistor and measurement errors are possible. 

    1 more resistors reduce parasitic capacitance 

    2 more resistors decrease power dissipation per resistor

    3 more resistors increase the series spacing between the pads this increases the creepage and clearance in the series resisters and reduces the possibility of arcing around the voltage dividers resistors during EMC transient events.

    Design error. yes.

    Why is this still good in my measurement. Depends on how much heat is dissipated to the board and traces. Will fail eventually

    Do I have to change them. Yes use multiple resistors for the voltage divider or a larger footprint / wattage resistor that meets the continuous voltage requirement. 

    Dave

Reply
  • +1
    •  Analog Employees 
    on Jan 13, 2021 1:26 PM 1 month ago

    The 7880 I think has a single 1M resistor. This board was designed for bench eval with an option to work with 240V. The single resistors should have been a min of 2 (500K) resistors . In the actual application 3 to 5 resistors in series are common.  One other issue with single resistor is the capacitive leakage around the resistor at 50Hz is a percentage of the 1Meg resistor and measurement errors are possible. 

    1 more resistors reduce parasitic capacitance 

    2 more resistors decrease power dissipation per resistor

    3 more resistors increase the series spacing between the pads this increases the creepage and clearance in the series resisters and reduces the possibility of arcing around the voltage dividers resistors during EMC transient events.

    Design error. yes.

    Why is this still good in my measurement. Depends on how much heat is dissipated to the board and traces. Will fail eventually

    Do I have to change them. Yes use multiple resistors for the voltage divider or a larger footprint / wattage resistor that meets the continuous voltage requirement. 

    Dave

Children
  • Thank you Dave for your very clear and complete answer.

    => Design error. yes.

    Okay, that's what I was afraid of.

    => possibility of arcing around the voltage dividers resistors during EMC transient events.

    I broke the evaluation board just like that: I had a connected three-phase motor (230VAC) as a load, during a transient, an electric arc was created and short-circuited the 1MOhm resistor for a short time. The ADE7880 did not hold the shock.

    I am not happy and I find it a pity to break the card like that because of this error, given the price of the card!  (386,49 €). So I will now have to reorder one, change the input level (and add protections: TVS diodes/MOV).

  • 0
    •  Analog Employees 
    on Jan 14, 2021 3:06 PM 1 month ago in reply to ElectroMax

    Normally TVS or MOV manage the surge and keep the transients low enough not to cause any issue. If you are using a motor I suggest you add tvs or mov across the phases to keep this from happening. This board has no Transient protection. 

    If you can solder the ADE7880 the parts are cheaper than the board. 

    Dave

  • Yes, I will use a TVS in parallel with the MOV (higher value of the TVS). Unfortunately, I do not have the necessary equipment to solder a new ADE7880.