Menu

Glitch once per second

2020-11-22
2020-11-30
  • Steve Lupton

    Steve Lupton - 2020-11-22

    Hi

    Thanks for a great project.

    I've got a bit of a strange problem. I'm using an ESP8266 and the latest release (132).

    My symptoms are a glitch in the stepping once per second. It's exactly once a second. If the motor hold current feature is enabled, the motor twitches once per second when not moving as well.

    I've tried both an 8825 driver and an L293D shield. Have swapped the ESP module for a different one (different manufacturer), and also tried a different power supply.

    Everything else works fine - just this twitch once a second... All firmware options are disabled except the absolute minimum. I'm running in Station mode.

    Any ideas?

    Cheers,

    Steve

     
  • brownrb

    brownrb - 2020-11-23

    It is nothing to worry about. Steps are not lost. The controller is just busy updating temperature, display and other things. The controller cannot do everything at once so has to pause some things whilst doing other things.

     
  • Steve Lupton

    Steve Lupton - 2020-11-23

    Thanks for the quick response. I've solved it... Ended up being a duff motor that was drawing way too much current. Swapped for another identical motor and adjusted the current limit on the 8825 to about 600mA and it is working beautifully. It still twitches once a second when stationary but I monitored it over time and the overall position doesn't change.

    In case you're interested, this is the drive I came up with via two herringbone gears - it seems a good arrangement with no noticeable backlash.

     
  • brownrb

    brownrb - 2020-11-24

    The latest firmware does use interrupts to drive the motor so these every second blips are much reduced.

    On the ESP8266 - because of the architecture and multiple use of pins for a whole lot of things - we do get occasional blips on the motor - some of these pins are in fact multiplexed with spiffs file access, - but the motor does not really move when this happens. It is much more noticable during boot.

    Because of the limited IO ESO8266 pins actually available we were forced to use some pins like that. Not ideal of course. And the ESP8266 has other issues, like more heat, single core which also has to handle all the wifi code as well as user code - etc

    The ESP32 is much much better in this respect. All round is a much improved processor to use with less issues.

    Regards
    Robert

    The photo is great. Can I include it in the pdf?
    I see the gears are 3D printed. Did you design the gears yourself? Did you put the STL files anywhere?

     

    Last edit: brownrb 2020-11-24
  • Steve Lupton

    Steve Lupton - 2020-11-24

    Hi

    Thanks - I'll buy some ESP32s - they are so inexpensive now. I only used the 8266 because I had loads of them lying around. That said, it is now functionally perfect - just need some clear skies to actually test it!

    The gears were produced using this OpenSCAD Parametric Herringbone Gear script:-

    https://www.thingiverse.com/thing:6544

    distance between axels = 40mm
    gear shaft height = 25mm
    gear 1 teeth = 9
    gear 1 shaft diameter = 5.25
    gear 2 teeth = 13
    gear 2 shaft diameter = 4.1

    You're more than welcome to use the photo, and the STL files for my bracket and the gears are attached.

     
  • brownrb

    brownrb - 2020-11-30

    Many thanks Steve, much appreciated

     

Log in to post a comment.

Want the latest updates on software, tech news, and AI?
Get latest updates about software, tech news, and AI from SourceForge directly in your inbox once a month.