time of reading encoder position

[erhan]

Hello,

I want to ask you people where can i learn function execution times of NI 7358 board? I'm developing driver for this card on QNX 6.5. When i send read encoder position package to 7358, the first data comes to RDB register after around 10 ms for first time. Is it normal? Is there any documentation file that includes all function execution times (average)?

Best regards.

[Paul_M.]

Hi erhan,

There is documentation you can consult to look at the benchmarked execution times for all the NI-Motion functions. If you go to Start » Programs » National Instruments » NI-Motion » Documentation » NI-Motion Help. You can then search for "7350 Timing" and you one of the results should be "NI 7350 Timing Information." This will provide you with the benchmarked times.

Paul M

[erhan]

Hello Paul_M,

Thanks for your reply. I want to ask you one more question. I checked timing table in documentation. For example, if i send "read axis status" package to card after sending "start" package, this timing results are still correct? I mean board has command fifo as i see. 

As i stated my previous message, i'm developing mini card driver for QNX RTOS. To read something basically i'm sending package data to card then i'm waiting for return data on RDB. Now, i have problem with that when i send read "something" data package to board after start motion command, i get data on RDB very late according to this table. I can say this happens just after start command. If i read package while motor stopped, there is no such latency.

Regards... 

[Paul_M.]

Hi erhan,

I'm not too surprised that once the motion has started, the "Read Axis Status" function takes more time. The card should devote more resources towards managing the movement. That being said, the timing help document is only in reference to the individual function calls and as you stated the Read Axis Status alone has no latency. 

Unfortunately, besides this timing help document, there isn't other documentation that is released that discusses the execution timing. If this is a bigger concern, I would try benchmarking the process on multiple computers.

Paul M