You usually need to issue a proper reset/init sequence before the AD7705 
will respond – either via the RESET pin or by sending the reset command 
over SPI (32 consecutive 1s). Also double-check that your SPI mode 
matches the datasheet (CPOL/CPHA), since a wrong mode can keep BUSY 
stuck low. If BUSY never toggles, it’s often a sign the chip hasn’t left 
reset

Jordan M. wrote:
> Hello everyone,

 I'm currently testing a small data acquisition project with the
 AD7705/AD7606 Data Acquisition Module (this one:
 https://easyelecmodule.com/product/ad7705-and-ad7606-data-acquisition-module/).
 The goal is to capture precise voltages from sensors via SPI.
 Unfortunately, nothing's working so far, the module simply isn't
 responding.

 What's happening so far:
 This seems like a situation where careful signal analysis and impedance 
verification would be useful before changing the design. Small 
mismatches can significantly affect performance at higher frequencies. 
Engineers working on RF systems often evaluate options such as Buy 
QAT-4+ Attenuator(https://www.ovaga.com/products/detail/qat-4-) 
components when precise signal control and attenuation are required.
 I have a stable 3.3 V supply, all GND references are correctly
 connected, and the SPI lines are soldered. When I try to read the data,
 the BUSY pin remains permanently LOW, even though I trigger the
 conversion start (CNVST) via SPI.

 With the AD7705, I wanted to read data, but the return value always
 remains 0x0000 , apparently, the reset/init signal isn't being
 recognized correctly.


 My questions:

 Does a reset or init command have to be activated via hardware (reset
 pin) before data can be read? (For some users, this was the solution.)
 Are there any basic timing rules for SPI that are easily overlooked and
 lead to busy pin problems?

 How do you typically measure the busy pin? Do I need to reduce the load
 or add a buffer?

 I would be very grateful if anyone could help me get started or give me
 some tried-and-tested tips.

 Regards and thanks in advance,
> Jordan

A permanently LOW BUSY pin usually suggests the conversion is never 
starting, so I'd first verify the CNVST signal with an oscilloscope or 
logic analyzer rather than relying on software assumptions. Also check 
the RESET pin state and power-up sequence. For SPI, confirm clock 
polarity/phase settings match the ADC requirements, as an incorrect mode 
can easily result in all-zero readings.