• State TI Thinks Resolved
  • Locked Locked
  • Replies 1 reply
  • Answers 1 answer
  • Subscribers 63 subscribers
  • Views 265 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

TPS7A96: Test configuration for Ultralow noise power density measurement from 10Hz to 20MHz and up.

Gavin Sun
Prodigy 10 points
Part Number: TPS7A96
Other Parts Discussed in Thread: TPS796

Part number: TPS7A96: Part Description: Ultra-Low Noise, Ultra-High PSRR, RF Voltage Regulator

Questions:

 

The plot shown above is taken from page 1 of the part’s datasheet. I will be performing a similar test to measure the ripple and noise power density coming out of a switcher and fed to a regulator. I would like to know how to configure my test setup for this measurement.

I heard that measuring noise power density down to 10Hz is challenging (please educate me why), and I’m concerned that this may indicate a high measurement error at 10 Hz. If this is true, how can I avoid or compensate for measurement errors?

What test instruments were used when the noise data in the picture was taken? Additionally, do you have any suggestions for test equipment for ultra-low noise (nV/srHz) from 10Hz to 20MHz (or greater) measurement? How is DC blocker implemented? Is it a built-in feature of the instrument or does it require an external attachment. 

Being new to this domain, I might be missing some crucial questions for a successful measurement. So, if you have any suggestions beyond the questions I have asked, please kindly share them. Thank you so much!

Ticket Number: CS2167801

  • 0
    TI__Expert 4035 points

    There are multiple questions in this thread that I will try to give a briefed answer for them. All of TPS7A96 device noise is measured using 4395A spectrum analyzer + in-house built noise gain amplifier. We needed to gain-up the TPS796 output noise since it is lower than the analyzer’s noise floor.

    Any spectrum analyzer has a noise floor also known as Displayed Average Noise Level

    Frequency DANL for short which represents the lowest noise such instrument can resolve (measure).

    The DANL for the 4395A or any other spectrum analyzer is usually expressed in units dBm/Hz which is the spectral noise density measured in units of power (V^2/R) with either 50-ohms or 75-ohms termination while using a 1 Hz resolution bandwidth, below is a snipped for the 4395A DANL.

    We can simply convert dBm/Hz to nV/Hz using;

    https://www.cantwellengineering.com/calculator/convert/dBm

    Z is the analyzer termination which is 50-ohms for the 4395A.

    TPS7A96 spectral noise @1kHz is ~ 1.3 nV/Hz (typ.) which translate to ~ -164.7 in dBm/Hz.  TPS7A96 spectral noise goes down to 1.1nV/Hz (typ.) beyond 10kHz which is ~ -166.2 in dBm/Hz.

    It is obvious that the 4395A analyzer cannot measure TPS7A96 low noise hence an amplifier is needed to gain-up the TPS7A96 noise to a level higher than the 4395A DANL.

    I am attaching few TI resources on noise and LDO noise measurement for more details.

    The first step to make any noise measurement is to figure-out/simulate the expected spectral noise for your device/system/application then compared against the noise floor for your analyzer. In order to get greater than 95% accurate noise measurement a 10dB ( 3.16 to 1 ratio) is needed between your device/system/application noise to the analyzer noise floor across the measurement BW.

     https://www.ti.com/lit/wp/slyy076/slyy076.pdf?ts=1705433035736&ref_url=https%253A%252F%252Fwww.google.com%252F

    https://www.ti.com/lit/an/slaa412b/slaa412b.pdf?ts=1705434274398&ref_url=https%253A%252F%252Fwww.google.com%252F

    https://www.youtube.com/watch?app=desktop&v=W2UUIUPmwdU

    Best,

    ~Haidar