search

gnea / grbl

An open source, embedded, high performance g-code-parser and CNC milling controller written in optimized C that will run on a straight Arduino
https://github.com/gnea/grbl/wiki
Other
4.02k stars 1.6k forks source link

Loosing Steps #516

Open algo1980 opened 5 years ago

algo1980 commented 5 years ago

Hi everyone, I am ending up in a very confusing situation. When I jog the machine while the spindle is turned off it works well but as soon as I kick in the spindle the machine starts to lose steps. For the electronics, they are placed in a metal enclosure which is grounded.

romychs commented 5 years ago

I think the spindle eats up all the power of the power source, if it is common for him and stepper motors. Lack current to stepper motors to hold the position. If you think this is interference, try soldering the ceramic capacitor 0.1uF-1uF between the motor terminals.

algo1980 commented 5 years ago

Romychs I have connected the spindle into a different socket, it's totally isolated from the driver and controller box which is grounded. to let you know more about the situation it was working well before it ended up in this bug. How can a spindle interfere with other electronics when it is not even connected to the main electronics board "I mean completely wired up separately"? Not to criticize your point of view but this will allow me to learn more.

I am going to tryout your proposed solution but just to be sure did you mean I should connect a ceramic .1-1uf cap on both the terminals of steppers which are A+ A- and B+B-? and this should go for all axis?

swarfer commented 5 years ago

please read this https://github.com/gnea/grbl/wiki/Wiring-Limit-Switches and this http://www.mycncuk.com/threads/9504-Why-star-ground

separate wiring does not ensure the interference does not happen because the interference is transmitted through the air just like a radio signal. if your spindle has brushes then it makes sparks, and sparks make a lot of radio noise (the first radio transmitters were just spark gaps!). any wire that has varying electricity in it will transmit, and any other wire will pick up some of that transmitted energy.

romychs commented 5 years ago

Algo1980, I had in mind a capacitor between of spindle motor leads. Swafer well described the cause of possible interference. Re-read the question again, the steps are lost when you kick the spindle (by hands, for example), do not turn it on or off? Then, all I said above will not solve the problem. Try adjusting the hold current of stepper motors. Try reducing the half-step mode, for example, from 1/16 to 1/8 for testing only.

zeevy commented 5 years ago

This may depend on how you wired and routed the spindle wires. you should avoid routing spindle wires along with other signal wires

wmarkow commented 4 years ago

as soon as I kick in the spindle the machine starts to lose steps

When you manipulate spindle speed you need to wait a bit of time by adding some dwell commands. For me G4 P0.5 did the job. See also here.