need help! TinyG keeps stalling

Home › Forums › TinyG › TinyG Support › need help! TinyG keeps stalling

Viewing 4 posts - 1 through 4 (of 4 total)
  • Author
    Posts
  • October 1, 2015 at 6:52 am #8797
    mara
    Member

    During a job the planner buffer max’s out and the job is stopped, I have tried different sized jobs.
    I am not sure what the cause is, the only change I have made is updating firmware to 440.20. The spindle runs independent from the board so the PWM on the board is not being used however since the problem occurred the light is ON constantly even though I have entered $p1pof=0 to turn it off.

    efforts so far to try and fix;
    used avrdude to go back to 440.18.
    changing between CTS and XON with $ex settings corresponding.

    Initially I was trying to run a large file ~ 10,000 lines when this problem started and trying again repeatedly with the planner buffer maxing out and the job stopping. Could this action have damaged the board?

    When the board was running successfully the blue and red SPON lights were ON
    now the PWM light is constantly ON and the SPON light only comes on during a job.

    I am not sure what else to try so if you have any advice I would greatly appreciate it, thanks.

    October 1, 2015 at 9:08 am #8798
    mara
    Member

    Answer to problem;

    I wasn’t aware that turning off the 3D viewer helps a larger file to run so much more easily.

    Also, I overlooked the fact that I have only 2 GB of ram.

    October 1, 2015 at 10:29 am #8799
    cmcgrath5035
    Moderator

    In reflashing to 440.18, the parameter $p1pof probably reset to 0.1 (10%).
    Set $p1pof=0.0 and the PWM led will go off when spindle off.

    Back in the days of less sophisticated SPJS, there was a way to hand tune the number of lines delivered in a period of time.

    What SPJS version are you running?
    Ver 1.83 and now 1.86 are much faster (more efficient use of cpu), SPJS should not be overfilling your tinyG buffers.
    Does SPJS run a a limited resource machine?

    October 6, 2015 at 12:47 am #8810
    jlauer
    Member

    One of the main features of 1.86 is fixing stalling, so I would try that out and see how it goes. The stalling was likely a threading issue and more mutex locking was added between the inbound serial data thread and the outbound serial data thread. Early reports are showing that it 100% solved it.

    Also, the latest version of SPJS is not necessarily more CPU efficient because the feed rate override did add a bit of overhead. SPJS now has to parse each line to keep track of the current feed rate. Then when you ask for a multiplier, it knows what to multiply against. So it is constantly watching, parsing, and calculating.

  • Author
    Posts
Viewing 4 posts - 1 through 4 (of 4 total)
  • You must be logged in to reply to this topic.