Setting Control law frequency

[aimthiazz]

Hello everyone,

I am validating the control law/loop frequency.

I am setting my task frequency to be 25kHz (as shown below)

and SYS_TICK_FREQ as 25kHz (shown below)

I am using the following syntax to log the control law frequency

I expect a control law time of 40us (1/25kHz). However, I noticed that the control law time is ~25us (shown below).

Any Idea why I notice the difference between the set control law frequency and the observed control law frequency?

[npetersen2]

Can you try a few other control intervals and see if they work? For example, 20kHz and 30kHz? I believe Bharat also saw rather odd timings for certain control intervals. On Aug 24, 2022, at 3:06 PM, Imthiaz Ahmed ***@***.***> wrote:  Hello everyone, I am validating the control law/loop frequency. I am setting my task frequency to be 25kHz (as shown below) [image]<https://user-images.githubusercontent.com/14706814/186424602-ab8c662b-16de-4ea4-b5f0-9575190bd7c3.png> and SYS_TICK_FREQ as 25kHz (shown below) [image]<https://user-images.githubusercontent.com/14706814/186424824-5bca3582-65eb-4530-b9c4-68910ea547a1.png> I am using the following syntax to log the control law frequency [image]<https://user-images.githubusercontent.com/14706814/186424401-6ea4175f-5e7b-45d8-9b31-9f3d5d147c77.png> I expect a control law time of 40us (1/25kHz). However, I noticed that the control law time is ~25us (shown below). [image]<https://user-images.githubusercontent.com/14706814/186425857-ddc30f16-ab2b-4026-8edb-f8b8aadb4869.png> Any Idea why I notice the difference between the set control law frequency and the observed control law frequency? — Reply to this email directly, view it on GitHub<#283>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/AEZ4CHT77MQERMOV2LMCGXTV2YM5XANCNFSM57PIMN6Q>. You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

[aimthiazz]

Thanks, Nathan, I changed the control law to a few different frequencies (10kHz - 25kHz), looks like consistently the expected vs actual time is scaled by ~1.5 times.

Additionally, I also notice a significant variation in the logged loop time at high frequency (~25kHz)