Makelangelo/Adafruit MSV2 – Drawing Stalls or goes TOO FAST

[Anonymous]

Hi all –
I just started pulling together my Makelangelo, using:
Arduino Uno w/Adafruit Motor Shield V2
Two NEMA 17 motors
Servo
Gondola I printed from Thingiverse

Makelangelo V5 (I know it’s older than 7.1.0, but for some reason v5 works better)

I have downloaded the firmware to the Arduino, configured Makelangelo, set my machine and page dimensions, adjusted the tool parameters. I can move the gondola around just fine using the manual controls, now it’s time to start a print!

I tried the hilbert pattern to start, and when I attempt to start from gcode statement 0, it hangs on that statement (I haven’t been able to decipher it, will include it in a follow-up post). When I move past that statement, it draws the pattern, but does it in like 42 seconds!! Breaking a land speed record no doubt, but making for a very erratic looking print. The estimate was 8 min on the pattern.

I have been messing with feedrates on the adjust tool page, to no avail. I have even adjusted the gcode itself, no change. If I had to guess, it looks like these commands are being rapid fired at the arduino, with no waiting for the completion of each.

These two problems are probably related. Any guess on what is going on?

BTW, I have messed with 7.1.0 a little bit, and have other Win issues with that version, will cover in a separate thread.

Thanks in advance for the help,
Rob

[Anonymous]

Well…. the AMS2 isn’t officially supported. There could be some weirdness going on with that specific board.
Are your motors 200 steps-per-turn or 400? It makes a big difference.

[Anonymous]

These are the motors – they are 200 steps/turn:

https://www.adafruit.com/products/324

The AMS2 does use 16 microsteps, but that is in the firmware already.

What is interesting is that manual operation of the motors works fine.

[Anonymous]

In active tool settings, try lowering your feed rate. it’s probably something like 7000 steps/min. Try 2000 instead.

[Anonymous]

I dug into the code. Looks like the AMS1 is using 1/16th microstepping and AMS2 is NOT, which would make AMS2 run 16x faster?

There’s also this line in setup():

//TWBR = ((F_CPU /400000l) – 16) / 2;

which will 4x the speed of the motors on AMS2 only.

I’ve just made a commit to the github repo that might fix the issue. Let me know?

[Author]

Posts