Post Go back to editing

AD4134 SPI Data Read Issue & Minimum SPI Mode Clarification

Category: Datasheet/Specs
Product Number: AD4134

Controller: STM32H7A3ZIT6Q
ADC: AD4134
Communication: SPI (4-wire mode)
Data Size: 8-bit
MSB First
SPI Mode: CPOL = 0, CPHA = 0 (Mode 0)
Frequency: 4 MHz

PIN/SPI - HIGH

RESET -  HIGH

1.Issue: Unexpected Data on SDO When Power is Off

  • Hardware is configured according to Table 22 in the AD4134 datasheet.

  • Clock Type: CMOS.

  • Problem:

    • Observed unexpected data on the SDO line using a CRO when the unit is powered off.

    • However, reading Vendor ID & Scratch Pad ,INTERFACE_CONFIG_A  max all registers works correctly.

    • Verified all power supplies, and they are functioning properly.

    • Still unable to identify the root cause of this issue.

Point right Question:

  • What could be causing data to appear on SDO while the unit is powered off?


2. Minimum SPI Mode – ADC Data Read Process

  • I want to confirm the correct procedure for reading ADC data in Minimum SPI Mode.

  • Based on my understanding, after toggling the ODR pin, we should read 24-bit ADC data, and this process repeats for every sample.

Point right Question:

  • Is this the correct method for reading ADC data in Minimum SPI Mode?


3.ADC Data Frame Format – Multi-Channel Output on SDO

  • When reading 4-channel ADC data on the SDO pin, how is the data structured?

    • Is it interleaved (mixing samples from different channels sequentially)?

    • Or is it channel-wise, where all samples from one channel are read first before moving to the next?

Point right Question:

  • What is the correct frame format when reading 4-channel data on SDO?

4.

Question: SPI Locking & Unlocking Pattern for Minimum I/O Mode

  • What is the SPI locking/unlocking sequence for AD4134 in Minimum I/O Mode?

  • Is there any reference or documentation explaining how to implement this correctly?

  • Hi  ,

    Please refer to the initial answers below. I will confirm other details and get back.

    1. Is there connection between AD4134 and STM32 when you measured it? I will look into this more and get back.
    2. Yes. With sufficient SCLK, you should be able to read data over SDO, configure the ADC as page 69 and check. From what I know, SPI of STM does not support the this data format over SPI.
    3. The data of the 4 channel will be interleaved like the figure below

    1. I will look into this further and get back.

    Thanks,
    Janine

  • Hi  ,

    On your question about SPI locking & unlocking, there is only that section on the datasheet.

    Writing 24 consecutive 1s will lock SPI (read or write commands will be ignored). To unlock, write 23 1s and one 0.

    Thanks,
    Janine

  • HI jestayo

    thanks for response .

    due to pull resistor on cs , i cant access registers ,but its working and capturing data some more question i have please help me. .

    Currently accessing AD4134 using two SPI interfaces:

    • One SPI is connected to SDI, SDO, SCLK, and CS.

    • Another SPI is connected to DOUT1(SDO) and DCLK.(SCLK)

    AD4134 is in ASRC slave mode with the following configuration:

    • ODR and DCLK are configured as inputs.

    • Gated DCLK is used.

    • Clock type: CMOS.

    • XCLKOUT_EN is enabled, and I observed 48MHz at both the input and output, which is working correctly.

    • Operating ODR as gpio output.maintaing ODR high time as per figure 2.

    Issues faced:

    1. Unable to access all registers. Some write operations (e.g., power-down register) are not successful. How can I ensure all registers are accessible?

    2. Chip error observed. The DIG_FILTER_OFUF register shows 0xF. What could be the reason for this error?

    3. Issue with data capture speed:

      • Set DIGITAL_INTERFACE_CONFIG to configured FORMAT = 0, meaning data from all four ADC channels should be output on DOUT0.

      •  Set frame 24-bit data only.

      • At SPI clock = 3MHz, data from all four channels is captured correctly.

      • At SPI clock > 6MHz, only two channels work correctly 48clocks , while trying to capture 4 channels generating 96 clocks getting constant data.

      • Why does increasing the SPI clock affect data capture for all channels?

    4. Channel selection clarification:

      • Based on my understanding, if I want data from only channel 1 and channel 4, I should power down channel 2 and channel 3. Is this correct?

    5. How can I achieve high-speed data acquisition for all channels?