Home General

Reading the SD card

edited November -1 in General
So I can get a mac to talk, most of the time, to the Log4.USB and save the results. But I can log without my Mac, so then how do I read it? I looked at the SD card, and it doesn't show up (weird format for a PC?). I'd like to set this up and retrieve the log at a later time.


  • edited November -1
    Logfile saved to SD card is just normal textfile. If you can't open it, that is probably because newlines are not those used by Mac and your text editor doesn't understand it.

    You could try using another text editor which understands different newlines. (I don't use Mac so I can't recommend any text editor.)
  • edited November -1
    Yes but... Column 1 is what? Column 2 looks like current in uA, column 3 is voltage in mV. When column1 doesn't increment, does that mean what?
  • edited October 2017
    Column 1 is Unix Timestamp (time elapsed since 1st Jan 1970) but in milliseconds. I've never seen it to not increment.

    btw, I've created a small web-page to convert log to other formats (only works with relatively short snippets, what can be copy-pasted to webpage):
  • edited November -1
    For column 1 to work properly, you need to connect the logger to software on PC/Mac at least once, to set correct time. Otherwise I'd expect column 1 to just count time in milliseconds but not be proper Unix Timestamp.
  • edited November -1
    Thanks for the info!
  • edited November -1
    My trace has almost all increment of column 1, but there are in the list differences of 0 and of more than 1 as well. example below:

    1506699492522 113500 4992
    1506699492523 114300 4993
    1506699492524 113400 4991
    1506699492526 113400 4991
    1506699492526 112700 4991
    1506699492535 112000 4991
    1506699492536 112800 4991
    1506699492537 115400 4991
    1506699492538 115400 4991
    1506699492539 118700 4990
  • edited November -1
    I havn't really used 1ms interval, I usually use 10ms.

    I guess that with 1ms measurement interval and 1ms timestamps, small timing errors could cause duplicate values (24,26,26 instead of 24,25,26).

    But gap from 26 to 35 seems strange, perhaps a bug where logger is unable to get 1000 measurements per second?
  • Hello,

    The 1ms interval on SD card doesn't work reliably on all SD cards dependent on the display mode of the logger. We have found that 5ms appear to be stable and more recently we have updated the firmware build to allow for slightly faster SD card access again.

    We are continuing to review the DMA implementation of the SD Card read/write and will release another patch if we can further improve the speed of SD access.

    The higher read speeds tend to choke up the Desktop software visualization system below 5ms dependent on the PC you run the analyser on.

    For now it is recommended to run at 5ms until the next firmware update is pushed and then trial at 2ms.

    Kind Regards,
    Tektyte Admin

  • Hello,

    We have since updated the firmware to improve 1ms stability. While it isn't a default option the Log4.USB and Log4.PoE now include a CUSTOM sampling rate option to set the logging rate below 5ms.

    The improvements to timing performance within a large number of firmware sub-systems made it possible to achieve this lower sampling rate but unless you have a good SD Card or a very fast computer live streaming at these rates are not recommended.

    NOTE: One part of the firmware update which assisted with the faster logging is the move to BINARY FILE FORMAT for rates below 5ms.

    The new desktop software now includes a converter from BINARY -> CSV for the export to other analysis tools. This means you can record reliably at faster sampling rates and then convert this file format with ease.

    Kind Regards,
    Tektyte Admin

Sign In or Register to comment.