Post Go back to editing

Powering AD5941 with Buck-Boost

Thread Summary

The user is designing a battery-powered wearable device with a single-cell Li-ion battery and USB Type-C interface, using the BQ24075 for power-path management and the AD5941 ADC. The initial plan to use the ADM7155 LDO for powering the ADC was reconsidered due to limited headroom and dropout voltage. The user opted for the ADP2504 buck-boost converter to maintain stable 3.3V output across the battery discharge range. The final design includes a 3.255V AVDD and 3.299V DVDD/IOVDD, with ripple analysis showing 38mV (AVDD) and 41.5mV (DVDD) ripple. The engineer confirmed that the ripple is acceptable if it remains below 27.77mV, suggesting increasing output capacitance to reduce ripple further.
AI Generated Content
Category: Hardware
Product Number: AD5941

Basic Explanation of my System:

The proposed design is a battery-powered wearable device that integrates advanced sensing and power management features. The system operates primarily from a single-cell Li-ion battery (3.7 V nominal) and includes a USB Type-C interface for charging and external power supply.

The device uses a battery charger with power-path management ( BQ24075) to ensure seamless operation:

  • When USB power is connected, the system is powered directly from the USB input while simultaneously charging the battery.
  • When USB is disconnected, the system automatically switches to battery power without interruption.

fuel gauge IC monitors battery health parameters such as voltage, state of charge (SOC), and remaining capacity, reporting these to the microcontroller via I²C.

The system includes an electrochemical front end (AD5941) with a 16-bit ADC for high-precision sensor measurements.

Sensor signals are routed through a multiplexer (MUX) to the AD5941, which interfaces with an STM32 microcontroller development board for data acquisition and processing.

My Question

Initially, I considered using the ADM7155 low-noise LDO to power the ADC. However, given the system’s 3.7 V Li-ion battery, the limited headroom and dropout voltage of the ADM7155 present a challenge, especially as the battery discharges.

To address this, I decided to use a buck-boost converter, specifically the ADP2504, which can maintain a stable output voltage across the full battery discharge range and during USB power input. This ensures reliable operation for both the ADC and the microcontroller.

For powering the AD5941, my plan is to supply DVDD directly from the ADP2504 output and derive AVDD from DVDD using a ferrite bead for isolation, along with appropriate decoupling capacitors. This approach minimizes noise coupling between digital and analog domains while maintaining a compact and efficient power architecture.

Could you confirm if this is an optimal solution for powering the AD5941 in a wearable application.

Or may I know you have any suggestions for better circuit /IC options

  • Hi,

     

    What voltage do you plan on powering your ADC with? The AD5941 will work with a supply voltage down to 2.8V. Depending on how low you want your battery to discharge, you could likely still just use the ADM7155 and power the ADC at 2.8V, because the dropout of the ADM7155 is about 120mV (and is lower for smaller load currents), so it would work when the battery is around 2.9V. 

     

    You could also look at LT3073, which has a dropout voltage of 45mV at max load, and lower as the load current decreases.

     

    Thanks!

  • Hi Karmijo,

    My entire board is working at 3.3V supply.

    Regards

    HARI

  • Hi,

     

    You would get good noise performance by using a buck-boost converter to regulate the voltage to about 3.7V and then for any noise sensitive chips, using a part like the LT3045, LT3042, LT3046, or any other current source reference LDO (like those from the LT304X family) to provide your ADC with 3.3V.

     

    Alternatively, I think a better solution would be that you could power the rest of your system via the 3.3V buck-boost, and then power the ADC directly from the battery, since the ADC will operate down to 2.8V and generates its own reference. Batteries have better noise performance than LDOs (except for the LDOs from our LT304X family), and when the battery drops below the 2.8V minimum to power the ADC, the battery should already be about 95-97% discharged. Finally, discharging the battery down below 2.8V will likely reduce the life of the battery since the more often and deeper it is discharged, the more likely it is to become damaged and the life reduced.

     

    Thanks!

     

    Thanks!

  • Hi Karmijo,

    My Power tree is given below.Could you please check this.

    I designed the below circuit using for powering the ADC and remaining circuitry using ADP2503. My simulation results and circuit diagram are given below.The ferrite bead P/N

    is MPZ1608S601ATA00. The simulation results are given below.

    Can I go ahead with this design. The voltage rails are respectively  3.2551427 V (AVDD) , 3.2995143 V (DVDD & IOVDD) ?

    If I am using internal reference, will the ripple on the supply line is an issue?

    Regards

    HARI

  • Hi,

     

    I do not see any issue with this power tree. You can go ahead with the design.

     

    As for using the internal reference, the PSRR of the internal references is about 60dB, so if this is acceptable for your system based on your 3.3V converter's output voltage ripple, then yes, you can go ahead and power straight from the 3.3V source. Otherwise, I would power directly from the battery. For the 16-bit ADC with a 1.82V reference, 1 LSB of noise would be about 28mV. 

     

    Thanks!

  • Hi,

    "As for using the internal reference, the PSRR of the internal references is about 60dB, so if this is acceptable for your system ".

    May I know how do I check this,I mean how to do the analysis and make sure that it is fine for my system

    May I know you have any application notes or examples to approach this.

    Regards

    HARI

  • Hi,

     

    For the internal reference of 1.82V, you can figure out what 1 LSB of noise would be. Since the ADC is 16 bits, 1 LSB corresponds to LSB = 1.82V/2^16, or 27.77uV. 

     

    Thus, if you want to have 27.77uV or less of noise, your power supply for a 60dB PSRR can have no more than 27.77mV of ripple. 

     

    This comes from the formula PSRR = 20log(output_ripple/input ripple). Solving for input_ripple given worst case output ripple is input_ripple = output_ripple/(10^(PSRR/20)), or for this case,  input_ripple = 27.77uV/(10^(-60/20)), or 27.77uV/10^-3 = 27.77mV.

     

    Thanks!

  • Hi ,

    Thank you very much.I am learning new things.Thanks for your support.

    May I know how to check theoretically the power supply cannot have ripples  more than 27.77mV.

    Regards

    HARI

  • Hi,

     

    No worries! Glad I could help. 

     

    To find out the theoretical ripple of your power supply, you could run an LTSpice transient simulation of your design with the ADP2504. Make sure to include the parasitics of the passive power components to make sure you get an accurate value.

     

    Thanks!

  • Hi Karmijo,

    I simulated a buck boost for directly powering this ADC. Please see the details below.

    Please let me know my thought process is correct or not.

    I am getting a ripple of 41.5mV (Peak to Peak) in digital rail(DVDD & IOVDD) and 38.01mV Peak to peak in AVDD. In the LTspice simulation  the ESR of the output capacitor (4 mΩ) and

    the  series resistance of the inductor (210 mΩ), so these losses and their effect on ripple are represented in the result. PCB parasitics are not considered.If you have any suggestion for including that in the simulation I am ready to redo the simulation.

    For a 16-bit ADC at 3.3 V, 1 LSB is approximately 50 µV.  The  ADC’s  internal reference has an AC PSRR of about 60 dB. ​This shows that the allowed ripple at AVDD = 50 μV × 1000=50 mV (Please correct me if I am wrong).

    • Ripple at ADC after PSRR:
    • FOR AVDD:
    • 38 mV / 1000 = 38 uV
    • For DVDD:
    • 41.5 mV /1000 = 41.5uV
    • 16-bit ADC LSB at 3.3 V ≈ 50 µV → ripple error ≈ 0.8 LSB (slightly below 1 LSB).
    • Please correct me if the above calculation approach is wrong.
    • May I know Can I go with BUCK-BOOST solution
    •  Or Do I need to go with BUCK-BOOST + LDO Solution.
    • Schematic:
    • DVDD RAIL:

    AVDD RAIL:

    Regards

    HARI