Wrong Measurement fenomenal On ADIS16475

Hi,

Based on the screenshot you shared, I can see you have a 1g reading on the Z-axis, which translates to approximately 9.8 m/sec² since velocity is the integration of acceleration over the time interval (deltat), in your case, deltat = 1 / fs = 1 / 1000 = 0.001 sec.

So, if you multiply the acceleration by the time interval:

9.8 m/sec^2×0.001 sec=0.0098 m/sec

Thus, the expected delta velocity should be approximately 0.0098 m/sec.

If you read the delta velocity register correctly and apply the correct scale factor, you should get a value close to the one calculated above.

In my case, using the ADIS-FX3 Eval system (ref: EVAL-ADIS-FX3 Evaluation Board | Analog Devices), I read the following value from the delta velocity register:

Z_DELTVEL_LOW = 210799 (first value)

Using the appropriate scale factor from the datasheet (Table 73 for 32-bit data):

Delta Velocity (m/sec)=210799×0.003052​/2^16≈0.009816872 m/sec

Now, if you take the ratio of acceleration to delta velocity:

Acceleration/delta velocity​=9.8 m/sec^2/0.0098 m/sec​=1000

This ratio matches your expectations (fs)

Therefore, I recommend verifying your delta velocity scale factor and calculation to ensure you are getting the correct values.  

let me know if you still have any other question. 

thank you

Hi,
Let me explain our situation more clearly. We have three prototypes equipped with the ADIS16475-2 sensor, all featuring the same firmware and electronic PCB design. Initially, Prototype-3 performed as expected, showing the same characteristics as Prototype-1 and Prototype-2. However, after several power cycles, the output behavior of Prototype-3 changed. Prototype-1 and Prototype-2 continue to function correctly, providing output values according to the 32-bit structure and the conversion constant ((46.566128731e-9) //(100/2^31)) specified in Table 75 of the datasheet. The graphic provided belongs to Prototype-2 and shows the expected output values.

We need to understand why Prototype-3 stopped functioning correctly after the power cycle tests. Is there a limit on the number of power cycles the sensor can endure, or a limit on the number of flash memory updates? We are considering using the ADIS16475-2 sensor in our products but are currently hesitant due to this behavior.

Prototype-2:

Hi 

ADIS16575 has no limit on the number of power cycles. As you can see in the datasheet, Flash memory has the endurance of 10k cycles; again, there is no limit on reading it.  

Flash memory Footnote: Endurance is qualified as per JEDEC Standard 22, Method A117, measured at −40°C, +25°C, +85°C, and +125°C. 

Hi, I have a few more questions about your prototypes. Can you switch IMUs from one setup (prototype) to another to confirm if there is no hardware issue? Secondly, would you be able to share the data you captured for Accel and Deltavelocity? 

Hi, 
We agree with your approach. Currently, we need a laboratory with advanced equipment to remove ADIS from our prototypes, and we will replace it once we find such a facility, but it is unlikely to happen in a short time. Do you have any idea, what kind of a hardware issue can cause this? 
Attached, we are sending you the short-term test data from Prototype 3. The data is in the specified format. Additionally, during this test, we rotated the ADIS and the connected PCB around the X-axis by approximately 90 degrees. Observing the change on axes will clearly show the behavior of ADIS.

adis_data.zip

Can you help me understand how you have mounted the ADIS16475 onto your prototype

It’s placed with automatic pick and place machine. A stencil was also used when applying solder.