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.

TPS546A24A: IOUT_UC_FAULT Issue

Part Number: TPS546A24A

I have a lot of bulk cap on this rail and the operation is set to "ON". This issue happens when we use the PMBUS operation command to turn on the VR. We don't use the EN pin, or in other words the ON/OFF register is set to 0x1B.

My guess is that since there is a lot of bulk cap, and the load device draws almost no leakage current, when the regulator is already ON, our SW tries sends the OPERATION command to power on the VR even though it is already on. I am guessing that every once in a while the UC is so low that it triggers this UC fault.

I would like to know if I can set the UC lower? I don't see it as an available register.

Here is what I get when I poll for the voltage under normal operation:

OFF:

VDD Output voltage = 0.007812V
VDD Input voltage = 5.0547V
VDD Output current = -0.0964A

ON:

VDD Output voltage = 0.896484V
VDD Input voltage = 5.0312V
VDD Output current = -0.4673A

  •  

    No, the negative current limit of the TPS546A24A is not programmable, however you can change the negative slew-rate via the TOFF_FALL command, increasing the TOFF_FALL time will reduce the negative resulting from the discharge of the output voltage capacitance.

    TOFF_FALL is command 65h, a 2-byte WORD encoded with the standard LINEAR11 format (Bits 15:11) representing the 2s Compliment power of 2 exponent and bits 11:0 representing the 2s compliment mantissa.

  • Let me add more details here:

    We never turn off the VR in our use case. From the get-go the VRs are turned on over the PMBUS OPERATION command. However, we are running a script that keeps "turning on", even though it's already on, the VR over the same PMBUS OPERATION command. 

    Other questions I have are:

    1. Can I disable IOUT_UC_FAULT? 

    1. Does IOUT_UC_FAULT prevent the VR from accepting new incoming PMBUS transaction such as VOUT_COMMAND changes (i.e. voltage change)? In which case the fault has to be cleared before more new VOUT_COMMAND can be issued?

    2. What could reasonably cause this fault?

    3. I'm not clear about how TOFF_FALL helps? I don't turn off the VRs under this script test condition.

    I apologize that it seems like we power down the VR in our use-case. The slow power down issue is just for my curiosity.

  • Hi,

    Peter will feedback to you on Monday.

    Thanks,

    Lishuang

  •  

    My apologies for misunderstanding your comment.

    The TPS546A24A's negative current limit is fixed, it detects the valley current on the low-side FET during the low-side FET on-time.  If the inductor current falls below -10A, the low-side FET turns off and the inductor current is forced through the high-side FET for the remaining portion of the switching cycle.

    Negative current typically occurs when the output voltage is reduced, either due to a rapid loss of load current or a controlled decrease in the output voltage.  If you are not adjusting VOUT or shutting down the output voltage, I would suspect that it's a result of the output voltage overshooting following a rapid loss of load current and the return of the output voltage back to regulation.

    IOUT_UC_FAULT does not stop switching or interrupt VOUT_COMMAND.  The current is prevented from further decreasing but the converter continue to operate.

    Is your script adjusting VOUT? it just adjusting the load current?