Post Go back to editing

Python DRIVER AD5687R output voltage problem

Category: Hardware

Hello,

With an  FTDI module FT4232H , I send SPI command but for code under 0x4000  Voutput  value is different from datasheet .I use CSn from SPI tie to sync\

12 bits => 4 first LSB are unused.

gain 1

Vref interne 2.5V

Vlogic 3.3V

VDD 5V

Vout (code=0xFFFF) = 2,500V

Vout (code=0xC000) = 1,875V

Vout (code=0x8000) =  1,25V

Vout (code=0x7FF0) = 2,500V

Vout (code=0x4000) =  1,875V

Vout (code=0x2220) = 1,000V

Vout (code=0x0010) =  0,002V

Someone have an idea, thanks for your help.

  • Hi  ,

    One possibility is that the GAIN pin might be misconfigured. The other possibility is that the SPI command isn’t formatted correctly. The AD5687R requires a 24‑bit frame (4 command bits, 4 address bits, and a 12‑bit data field left-shifted by 4 bits). If the data isn’t properly shifted or if you’re sending only 16 bits, lower-code values can be misinterpreted. Also, if you’re using a command that requires an LDAC pulse, make sure it’s being applied correctly.

    Could you share your complete system settings, schematic, or code snippet?

    Br,

    Den

  • Hello Dendalisay,  

    thanks you for your first answer.

    I complete my post by this light schematic

    Gain is tie to GND=0V

    Vref =2.5V out by the AD5687R

    And for SPI frame I verify it by oscilloscope with a digital decoder it correspond to   , the command send seems correct in any case, but please could you confirm the mode SPI ? 

    with the code 0x0010 (DB15:DB0)  correspond to 0x001 for 12 bit DAC and I have lower voltage output at 0,002V, it seems also correct ...

    Other point, the readback don't respond, the MISO (SDO) stay at HiZ (pulled up by R =10Kohms) when I send  0x91 (DB23 : DB16) in command. but is not important for the moment I prefere update the Voutput with good value first.

    Best regard.