Cyton Daisy SD card empty file + multiple files after session Issue
Shyamal
Canada
Hello,
We've encountered challenges with the Cyton Daisy SD card files. In two separate experiments utilizing pairs of Cyton Daisy devices, data is recorded onto SD cards. One problem we've noticed is that some session files end up empty. Additionally, there are instances where a single session generates multiple files: one large file and several smaller ones containing just a few lines of data. Our devices are approximately 1.5 to 2 years old. We're seeking solutions to address these issues.
Thank You
Comments
Shyamal, hi.
Please email customer service: (contact at openbci.com). Are the failures happening on both units? If so then it seems less likely this is a hardware failure. Are you using the recommended cards?
https://docs.openbci.com/Cyton/CytonSDCard/
The smaller cards work better. I do see your previous thread. That was not emailed to customer service?
https://openbci.com/forum/index.php?p=/discussion/3614/cyton-daisy-sd-card-file-is-empty#latest
William
Yes, I noticed it on both devices.
Yes, we use recommended cards (SanDisk 16 GB/8GB).
No, we haven't contacted customer service yet. Previously, we used a USB extension cord for tapping the dongle on participants' shoulders to make this issue less frequent. However, it is now critical for us to not have this issue occur frequently.
I had also a rare (once a 50-60 recordings) issue with sd cards, you can check it here https://openbci.com/forum/index.php?p=/discussion/3775/sd-card-file-is-empty/
Thanks Max. Our devices had empty files more often, happening approximately 40% of the time. At the moment, we've opted out of using SD cards because data loss isn't something we're willing to risk. However, we're actively experimenting and exploring potential solutions to address this issue, as we would like to have higher sampling rate for our studies.
I see, the issue might be on a firmware side (especially if you modified it), sometimes even when i send 12H recording duration to board it allocates less space for sd file (which i can detect due board returns number of blocks allocated in response and i can compare it with desired duration). For now I just restart session in that case, but it seems that SD card handling on a firmware is not perfect and might be not so stable due old library (sdfatlibrary) used.
You may flash firmware i've created and see if recording becomes more stable (I have 3 cytons, 2 of which are with daisy and use them everyday with that firmware so far so good), also my firmware will allow to set higher sampling rate for sd cards for offline recordings (like 500-1000Hz). Check it out here if you want to give it a try https://openbci.com/forum/index.php?p=/discussion/comment/20480/#Comment_20480
Hi Max, Thanks for the firmware. I will update it on one of our device for testing. I was wondering if your firmware writes external trigger values as well? and personally how do you convert the sd card file after your recordings?
Hello, no i havent implemented external trigger.
After flashing firmware you can start session from my script https://github.com/roflecopter/openbci-session just open session_start.py modify parameteres (number of channels, names, sf etc) and run it. it will start session to sd without streaming and do not require dealing with GUI. When recording is finished just turn off device, pull out sd card and insert it to laptop, open sd_convert.py and run it. It will pull all parameters from session start (like channels, date of start, sf etc which stored in sqlite db file) and will convert it to BDF and CSV files with channel names and with ADC values converted to uV.
After that you can open BDF everywhere you want, i also made 2 small script for basic plots and hypnograms (yasa) - like in that message https://openbci.com/forum/index.php?p=/discussion/comment/20480/#Comment_20480