Page 1 of 1
My first print that's not a test pattern.
Posted: Tue Oct 09, 2012 1:09 am
by darus67
I printed a Rocket Shot Glass
http://www.thingiverse.com/thing:9916
It looks corrugated because I don't have the acceleration figured out yet.
It accelerates and decelerates for each line segment that makes up the
circle of the body, and when it does, the extrusion gets wider. It also
makes the overall runtime longer, because it never gets up to full speed.
I'm running Teacup firmware on a regular arduino.
Dunno if I should screw around trying to tweak the acceleration on this
firmware, or bite the bullet and shell out the $$ for an Arduino Mega so
I can run a real firmware.
Re: My first print that's not a test pattern.
Posted: Tue Oct 09, 2012 8:51 am
by johnoly99
Does teacup have accel? I didnt think it did, but if it does, heck yeah, tweak on it! My accel is usually around 800-1200 for an H1. Hope that helps
Re: My first print that's not a test pattern.
Posted: Tue Oct 09, 2012 11:19 pm
by darus67
Teacup has several options for acceleration.
Here's a snippet of config.h where acceleration is configured:
Code: Select all
/** \def ACCELERATION_REPRAP
acceleration, reprap style.
Each movement starts at the speed of the previous command and accelerates or decelerates linearly to reach target speed at the end of the movement.
*/
//#define ACCELERATION_REPRAP
/** \def ACCELERATION_RAMPING
acceleration and deceleration ramping.
Each movement starts at (almost) no speed, linearly accelerates to target speed and decelerates just in time to smoothly stop at the target. alternative to ACCELERATION_REPRAP
*/
#define ACCELERATION_RAMPING
/** \def ACCELERATION
how fast to accelerate when using ACCELERATION_RAMPING.
given in mm/s^2, decimal allowed, useful range 1. to 10'000. Start with 10. for milling (high precision) or 1000. for printing
*/
#define ACCELERATION 1000.
The Rocket Shotglass was printed using ACCELERATION_RAMPING
With the numerous short moves that make up the circle, it was never getting anywhere near full printing speed. It also looks like the extrusion rate doesn't decelerate as much as the x-y motion, causing over extrusion at the beginning and end of each move. That resulted in the corrugated look of this print.
I tried RepRap style previously, but it had some behavior that caused other problems. Specifically, "Each movement starts at the speed of the previous command", even if the previous move was a rapid, non-printing move. That makes the current move start out too fast and slow down to the desired printing speed by the time it reaches the end of the move. This was causing me grief with adhesion at the beginning of a printing move that immediately followed a rapid non-printing move.
I also tried disabling acceleration entirely, but that didn't seem right either.
Looks like I need to design a controlled experiment.
Re: My first print that's not a test pattern.
Posted: Tue Oct 09, 2012 11:21 pm
by darus67
Or maybe I should dive into the code and fix the RepRap-style so it resets the acceleration appropriately after a rapid move.