I am trying to calculate how much data delay exists in the 16265 and 16384. I'm talking about the time delay between the physical gyro rotation and the data showing in the GYRO_OUT register. I couldn't find anything in the datasheets or in EngineerZone.

I'm thinking it is some negligible amount to get through the initial analog response and analog filtering. So the "gross" delay is all in the digital area. I calculate the digital delay = total taps * SMPL_PRD sample rate.

total taps is 1 + (2 * taps set by SENS_AVG). For example, if SENS_AVG[2:0] is 4, total taps is 33.

Am I correct here?

Hello Bill,

The analog and digital filters are the two key contributors to the group delay in the response. For the analog filter, we typically calculate this for each pole and then add it together, using the -45deg phase delay at the cut-off frequency (-3dB). For the ADIS16364, there are two poles: 404 and 757Hz.

T_delay_404 = (45/360) x (1/404) = ~0.31ms

T_delay_757 = (45/360) x (1/757) = ~0.17ms

For the digital filter, the delay is equal to one-half of the total number of taps. So, if you use SENS_AVG[7:0] = 0x02 and SMPL_PRD = 0x0001:

T_delay_FIR = [(2^N + 1)/2] x (1/819.2) = 5.49ms

T_delay_noFIR = [1/2] x (1/819.2) = 0.61ms

If we add them up, we get 0.31 + 0.17 + 5.49 = 5.97ms.

I hope that helps.

Best,

NevadaMark