• State TI Thinks Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 1 answer
  • Subscribers 64 subscribers
  • Views 429 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

LMR36506RREVM: Efficiency lower than expected

Goekhan Kaya
Prodigy 580 points
Part Number: LMR36506RREVM

Hi,

Our customer has been measuring LMR36506RREVM and cant get to the efficiencies advertised on the Datasheet.

Blue point below are the measured points.

What do they need to do, to be able to measure the curve in the data sheet? For them, the efficiency in the yellow marked range (10-30mA out) at 48V input voltage is particularly important.

Regards,

Gokhan

  • 0
    TI__Genius 10715 points

    Hi Gokhan, 

    Has the customer changed the default IC on the EVM to the LMR36506R5RPER? The EVM default IC is the adjustable version of the device, and the efficiency will be different. The inductor highlighted in Table 9-3 is also different than the inductor used on the default EVM, your customer will need to update the inductor to what is shown in Table 9-3 or something similar, pay particular attention to the inductor DCR as this will impact the measured efficiency.

    In addition to the schematic level changes, be sure to measure input and output voltage directly across the input/output capacitors to minimize any voltage drop. I recommend soldering a small pair of wires directly across these components and use a dedicated multimeter for each voltage measurement if possible. 

    Regards,

    Harrison Overturf 

  • 0 in reply to Harrison Overturf
    Prodigy 80 points

    Thanks for the answer.

    For testing we used the following induductor, which does have a lower DCR than the one mentioned in the datasheet. NS10165T220MNA

    We didn't change the IC to the one specified in the datasheet. Can we expect a 15% efficiency improvement by doing this? What would be the reason behind that?

    Regards,

    André Krummenacher

  • 0 in reply to André Krummenacher
    TI__Genius 10715 points

    Hello André,

    The fixed output voltage version of the device will be capable of higher efficiencies because the internal LDO will be powered by the VOUT/FB pin rather than VIN which should result in lower measured input current consumption. 

    Changing out the IC may not be the only change that is needed to measure the efficiency seen in the datasheet.

    As I mentioned before, the actual efficiency setup will also be important in measuring the device efficiency. The input and output voltages should be measured directly across input and output capacitors to ensure resistive voltage drop is taken into account. I will typically use 4 separate digital multimeters (VIN, IIN, VOUT, IOUT) to ensure that measurement results are as accurate as possible. 

    Regards,

    Harrison Overturf 

  • 0 in reply to Harrison Overturf
    Prodigy 80 points

    Hi Harrison

    Harrison Overturf said:
    The fixed output voltage version of the device will be capable of higher efficiencies because the internal LDO will be powered by the VOUT/FB pin rather than VIN which should result in lower measured input current consumption. 

    Am I right in thinking that the LDO current is only affected by the input voltage? Without a load attached(Iout=0mA) the input current I measured is 40uA. I would not consider this relevant because in the load segment that we specified the input current is magnitudes higher(see table below).

    Harrison Overturf said:
    As I mentioned before, the actual efficiency setup will also be important in measuring the device efficiency. The input and output voltages should be measured directly across input and output capacitors to ensure resistive voltage drop is taken into account. I will typically use 4 separate digital multimeters (VIN, IIN, VOUT, IOUT) to ensure that measurement results are as accurate as possible.

    I have attached pictures of our measurement setup. I don't think that the performance deviation is caused by bad measurements.

    Is there any change for us to reach the specified efficiency in a real world application?

    Best regards

    André Krummenacher

    Vin 48V   Frequency max 1MHz      
    Vout 5V Inductor 22uH  
    Device LMR36506RRPET  
      Calculated Calculated  Calculated
    Input voltage Output voltage Input current[mA] Output current[mA] Power in[mW] Power out[mW] Efficiency
       
    48,00 5,11 0,90 5,12 43,2 26,14784 0,605274074
    48,00 5,11 1,55 9,05 74,4 46,2455 0,621579301
    48,00 5,10 2,22 13,01 106,56 66,351 0,622663288
    48,00 5,10 4,73 27,95 227,04 142,545 0,627840909
    48,00 5,09 8,29 49,95 397,92 254,2455 0,638936218
    47,99 5,06 33,22 207,50 1594,2278 1049,95 0,658594713
                   
  • 0 in reply to André Krummenacher
    TI__Genius 10715 points

    Hi André,

    You're correct that the input current to the regulator will vary based on the input voltage however the input current doesn't vary only based on input voltage. 

    Take the functional block diagram for instance, it shows that the input voltage to the internal LDO will depend on whether the device is fixed or adjustable output:

    As I mentioned before, the input current to the buck will decrease when using the fixed output voltage versions of the device because the LDO will draw current from VOUT rather than VIN. 

    In terms of changes you can make to improve efficiency, the IC can be changed to the fixed 5Vout device as I mentioned earlier. Also, the voltage probing you are showing is good but could be improved by soldering a small pair of wires directly across the input and output capacitors for the input and output voltage measurements. 

    Regards,

    Harrison Overturf