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.

INA302: AC High Voltage Current Measurement

Part Number: INA302
Other Parts Discussed in Thread: TIDA-01590, INA260, TIDA-01065

Hi 

I am designing 115VAC switch with mosfets(back-to-back configuration) and I want to measure load current with INA302A3. Can I use mosfet’s driver isolated ground in INA302 ground ? Load current is about 20-25A, I think I can use it because it is referenced to iso_gnd

thank you,

regards

  • Hey Jamiryo,

    Thanks for posting this with the schematic. The GND pin of the INA302 can be floated as long as the voltage at the input pins (VCM) with respect to GND pin does not exceed 36V. In fact, if the INA302 GND floats, then all of the pin voltages must be evaluated with respect to GND pin. So the 5V supply must only be 5V above ISO_GND level at all times. I am not sure how this isolation gnd works for this driver or where the 5V comes from. As of now you still have the inputs of the INA302 connected to the line voltage so the GND pin, the VS pin, and all other pins will have to move in unison with the LINE voltage. The VOUT will also be generate with respect to the HV so this will have to be level shifted to be with respect to Earth GND.

    Please post back with any more questions. Here is some collateral showing our parts operating on HV rails, albeit, DC HV rails.

    40 V to 400 V Uni-directional Current/Voltage/Power Monitoring Reference Design

    http://www.ti.com/lit/ug/tidu849/tidu849.pdf

    High-voltage, high-side floating current sensing circuit using current output, current sense amplifier

    http://www.ti.com/lit/an/sboa295/sboa295.pdf

    –48-V Telecom Current, Voltage, and Power Sense With Isolation Reference Design

    Sincerely,

    Peter Iliya

  • Hey Peter,

    15V(for driver) and 5V supplies are coming from isolated DC/DC converter (coming from MCU side of the circuit, they are totally isolated from this side of circuit) and I am using this 15V for gate driver and 5V for logic and amplifier circuits. Also, after INA302, I have ADC to read current with SPI interface to send MCU via isolator IC.  So, 5V will not go above ISO_GND and 15V will used for driver 

    Peter, I do not understand totally what you said with "As of now you still have the inputs of the INA302 connected to the line voltage so the GND pin, the VS pin, and all other pins will have to move in unison with the LINE voltage. The VOUT will also be generate with respect to the HV so this will have to be level shifted to be with respect to Earth GND."

    Could you please explain it more clearly ? 

    Actually, here at TIDA-01590, the usage of GND_HV is the same of my ISO_GND, the only difference is my ISO_GND will be upper side of the shunt resistor. 

    Thank you,

  • Hey Jamiryo,

    I guess I am confused as to what absolute voltage level the iso_gnd actually is. So the 5V_ISO, 15V_ISO, and Vref_ISO are all referenced to iso_gnd, but what voltage is iso_gnd floating to? In the TIDA-01590 document, the “GND_HV” rail is at 1200 VDC. So in this reference design, the INA260’s input pins are at a common-mode voltage (VCM) of 1200V. The only reason the INA260 is not exploding is because its GND pin is also connected to GND_HV and its VS pin is 3.3V above the GND pin. So its VS pin is at an absolute voltage level of 1203.3V. Meaning the INA260 is actually operating in low-side configuration because its input voltage with respect to its GND pin is 0V. Hope this is not over-explanation.

    Anyway, as far as I can understand in your design, the INA302’s inputs are connected to AC Line voltage that is at 120VAC with respect to Earth ground.

    Will both FETs ever be turned on at the same time? If so, then INA302 inputs will be at 120VAC, correct?

    Best,

    Peter

  • Hey Peter,

    Iso_gnd is coming from DC/DC converter com pin. In TIDA-01590, the GND_HV pin is not directly connected to GND, It is connected the GND through the Solar Inverter load. So the input pin of INA260 will be around 1200V with respect to GND but thanks to the GND_HV, it will just see around shunt resistor voltage drop. In my application the input pins of INA302 will see 120VAC with respect to neutral line but if we consider iso_gnd it will see voltage between pins and iso_gnd. is that true ? Vs pin of INA302 is 5V referenced to iso_gnd, If we make the reference point is neutral, yes IC will burn. 

    Both FETs will turn on at the same time. INA302 inputs will be at 120VAC but with respect to neutral line. 

    You can also check TIDA-01065, self-isolated ac mosfet switch, It is low voltage around 24VAC but logic circuits is working with respect to gnd between mosfets.

     I just want to discuss and understand that more clearly. I think I am more confused right now :))

    Thank you for your reply, I hope we can figure it out :) 

    TIDA-01065 schematic;

  • Hey Jamiryo,

    I think I see what you're saying. If you're DCDC converter is biased to a moving Line-Neutral voltage and the INA302 GND pin also biased to this voltage, then INA302 is sensing on the Neutral voltage as well making the VCM = ~0V.

    This should be fine, but I cannot say fully if this design is OK because I'm not sure what the states the FETs are being driven to. If both are ON, then iso_gnd is Neutral and INA302 should sense the load current. If both are off, then the iso_gnd is floating and inputs are high-voltage. This should also be fine since there won't be a return path from inputs to GND pin.

    One potential issue is if there is a situation where there is backflow current from iso_gnd through the body diode of the bottom FET. If this were to happen, then the input VCM of INA302 would become a negative voltage momentarily. These voltages can exceed 0.3V and if VCM <-0.3V, then the part can be damaged and we can't be certain that it will be working properly since -0.3V is the Absolute minimum VCM voltage according to section 6.1 of datasheet.

    Best,

    Peter

  • Hey Peter, 

    Thank you for your answer, now we understand each other more clearly. Design seems okay under these circumstances except for -0.3V negative voltage. What if I put the sense resistor between the top mosfet and iso_gnd ? There will be just small voltage drop on it and it will not affect the driving. This time INA302 will not see -0.3. Do you agree ? 

    Thank you, 

    Regards,

  • Hey Jamiryo,

    I do not think this will work. If the top FET is on and the bottom is off, then current can flow through bottom FET's body diode (when LINE=-120V) and make it forward biased. Adding a resistor where you suggest will only make the VCM of INA302 more negative.

    Why not just measure current in between the FETs? This would be a true low-side sensing.

    Best,

    Peter

  • Hey Peter,

    FETs will be ON together. In the worst case, lets say the top FET is on and the bottom is off, then current can not flow through bottom FET's body diode at -120V but It can happen at 120V because of the direction of bottom FET's body diode.

    What do you mean with between the FETs ? there is a iso_gnd there. As I said before , If I put the shunt resistor and INA302 between top FET's source and iso_gnd. In the worst case, lets say

    1-If the top FET ON and bottom OFF at 120VAC, current will flow through the bottom FET's body diode and make it forward biased. at this time INA302 input pins will not see negative voltage except for shunt resistor voltage drop magnitude. at -120VAC, the current can not flow because of the direction of bottom FET's body diode.

    2-If the bottom FET ON and top OFF at -120VAC, current will flow through the top FET's body diode and make it forward biased. at this time INA302 input pins will not see negative voltage except for shunt resistor voltage drop magnitude again. at 120VAC, the current can not flow because of the direction of top FET's body diode.

    is that clear ? 

    Note: In normal circumstances, both FETs will be ON together and current will flow through them, not their body diode as long as their Rds voltage drop will not exceed the body diode forward bias voltage.

    Thank you,

    Regards

  • Hey Jamiryo,

    Yes how you explain situations 1 and 2 is clear to me and is why I recommended sensing right next to iso_nd in between the top and bottom FETs. This way you can completely avoid negative VCM voltages that could break the part. Essentially, if INA302 is sensing right next to the iso_gnd connection, then it is always a low-side configuration and VCM with respect to iso_gnd is always 0V (except for shunt voltage, but this should be small anyway).

    I also now understand that normal operation is both FETs on an conducting.

    Please let me know if you need more assistance.

    Sincerely,

    Peter Iliya

    Current Sensing Applications

  • Hey Peter, I guess I will use shunt between two mosfets, thank you for your assitance,

    Sincerely,