- Use hardware timestamps if available to align data precisely.
- Implement continuous streaming to avoid starting/stopping latency.
- Employ hardware triggering to synchronize input and output channels.
- Utilize dedicated synchronization lines for clock alignment.
- Pre-fill the output buffer to minimize initial latency.
- Ensure consistent sampling rates and use the hardware clock for timing.
- If hardware solutions fail, use software compensation with functions like "finddelay" to align signals post-acquisition, automating this process in MATLAB if necessary.
Data Translation DT9857E variable IO latency
4 views (last 30 days)
Show older comments
Hi,
I'm using DT9857E with 1 OUT (sig gen) and 2 IN (IEPE Mics) channels to capture data into MATLAB - I am using the DT hardware Toolbox.
I have found that the system I/O latency changes from measuement to measurement making it difficult to use multiple repeats of the same measurement for data processing, as the signals are not synchronous between measurements.
I used implementation of background as well as foreground measurements from the DAQ toolbox and the result is the same - variable system latency.
I checked that by creating a loopback channel from OUT0 to another channel (IN3) and run 3 separate measurements. Then computed delay between measurements using finddelay MATLAB function. Each measurement is shifted between one another by a different (unpredictable) number of samples.
Is there any fix for that, or should I always run a finddely function and then adjust the signals so they are in sync ?
Appreciate any help,
Karol
0 Comments
Accepted Answer
Pratyush
on 8 Feb 2024
Hi Karol,
To address variable system latency in data acquisition with the DT9857E device in MATLAB, you can try the following approaches:
Hope this helps.
More Answers (0)
See Also
Categories
Find more on Analog Data Acquisition in Help Center and File Exchange
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!