Because of the holidays, TI E2E™ design support forum responses will be delayed from Dec. 25 through Jan. 2. Thank you for your patience.

  • State TI Thinks Resolved
  • Replies 3 replies
  • Answers 2 answers
  • Subscribers 46 subscribers
  • Views 61 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

AMC3336: AMC3336QDWERQ1 - As an isolated power supply

Kosta Kotsis
Prodigy 90 points
Part Number: AMC3336
Other Parts Discussed in Thread: ISOTMP35-Q1,

Tool/software:

Hello all,

I am using AMC3336QDWERQ1 in my system in order to do DC-Link voltage measurement, and phase voltage measurement.

At the same time I am planning to use a couple of temperature sensors on my HV side (ISOTMP35-Q1), and since these sensors are on the HV side, I wanted to use an isolated power supply in order to supply them.
Since I want to use as few components as possible, I was wondering if I could use this component(AMC3336) as an isolated power supply for my sensors. 
And so, I was thinking to connect these sensors between DCDC_OUT & DCDC_HGND.

Here I am also attaching the schematic of this system:

The sensors have a current consumption of a couple uA, so referring to this value, AMC3336 should be able to supply them.

However I was wondering if there is any other implication I am not taking into account. 

Any advice or feedback would be appreciated. 

Thank you in advance.

Best regards,
Kosta Kotsis

  • 0
    TI__Guru 50981 points

    Hi Kosta,

    I think this should be OK from the AMC3336 perspective however it may introduce additional radiated emissions or other EMC interactions. 

    https://www.ti.com/lit/an/sbaa515a/sbaa515a.pdf?ts=1734354482017

    From the TMP perspective (you can also create another thread addressing this PN to get that teams feedback) is the MCU that will be reading the temperature output VOUT referred to the same HV potential as the high side (HGND) of the AMC3336? 

    It may be helpful to share a higher level block diagram of what you are trying to do to make sure you are correctly spanning the isolation barriers. 

  • 0 in reply to Alexander Smith
    Prodigy 90 points

    Hello Alexander,

    Thank you for your feedback.
    Below I am attaching a higher level Overview of the system I am trying to design.

    As an explanation:
    I am on the process of designing the evaluation board for a full bridge DC-DC converter.
    On this converter, and related to my question posted here, I am interested in measuring the DC-Link voltage, Load voltage, and temperature of the transistors I am using.

    For the temperature measurement, I am considering the ISO_TMP35.
    For the DC-Link, and Load voltage measurement, I am considering the AMC3336.

    As I mentioned before, one of my objectives is to use as few components as I can. This is why I considered supplying the two ISOTMP35s from the output of one of the AMC3336s (the one that i will use for DC-Link voltage measurement).
    I am considering to place these temperature sensors on the HVDC- side of the transistors.
    I am doing this because, if I am correct, this should allow me to use one isolated power supply for both of these sensors. The isolated power output of the AMC3336 is referred to HGND, which at the same time is on the same node as the HVDC-. These implications make me consider using one of the AMC3336 as an isolated power supply for the temperature sensors.

    At the same time, to answer your question, the MCU that will be reading the temperature VOUT is NOT referred to the same HV potential as the high side (HGND).

    I hope this gives a better overview of what i am trying to do. Hopefully there is more feedback I can get from you or from anyone else.
    At the same I will create another thread from the TMP perspective.

    Thank you in advance.

    Best regards,
    Kosta Kotsis High_Level_Schematic_Overview.pdf

  • 0 in reply to Kosta Kotsis
    TI__Guru 50981 points

    Hi Kosta,

    Yes indeed, thank you very much for the confirmation. 

    I was not sure how you planned to send the TMP data back across the ISO barrier. I see now. 

    I think this should be OK, my main concern is now ensuring that there is a low inductance (or protected) connection between the TMP, AMC and gate driver. 

    If the TMP sees a common-mode potential bounce, it could pull on the supply from the AMC or OV stress the Gate driver ADC. 

    P.S. the AMC33xx will not require the SN65 driver.