search

MarlinFirmware / Marlin

Marlin is an optimized firmware for RepRap 3D printers based on the Arduino platform. Many commercial 3D printers come with Marlin installed. Check with your vendor if you need source code for your specific machine.
https://marlinfw.org
GNU General Public License v3.0
16.17k stars 19.21k forks source link

Max speed #1563

Closed campe89 closed 9 years ago

campe89 commented 9 years ago

Hi, I have a ramps 1.4 with drv8825 (32 microstep) mounted on a home made printer for laser engraving. I have repetier-host 1.0.6 on windows 7 x64 and the gcode file is about 80MB. With 10000mm/s² of acceleration, I can go up 2000mm/min for a fluid movements; at more speed the printer goes at intermittent movements. Can be a mechanical problem? firmware? host software?

nophead commented 9 years ago

How many steps per mm do you have? That speed is slow for belts but fast for lead screws.

campe89 commented 9 years ago

The speed is referred at the X Y movements with belts and t20 pulley; the microstep is 160.

nophead commented 9 years ago

There are un-resolved issues with DRV8825, see https://github.com/MarlinFirmware/Marlin/issues/975. Not sure if that is your problem though.

campe89 commented 9 years ago

I can try with the old a4988

boelle commented 9 years ago

is this one still relevant or can we close it?

campe89 commented 9 years ago

I haven't resolved the problem; but you can close.

boelle commented 9 years ago

we should only close if either:

1) its not a bug in firmware 2) it has been fixed...

as @nophead said there are some un-resolved issues in #975 and in there at least one other issue is mentioned... but those issues went dead and we opted to close them as there was no response.

boelle commented 9 years ago

i have now opted to mark it as a verified bug on the grounds that there are still issues..

Wurstnase commented 9 years ago

My guess is, that the gcode has a very high resolution. With that, the planner has a lot to do (a lot of expensive devisions) and this can make this issue.

campe89 commented 9 years ago

@Wurstnase I think the same. Maybe Arduino Due can resolve the problem?

boelle commented 9 years ago

that would also mean its not a bug in marlin

Wurstnase commented 9 years ago

@boelle yes, I think so @campe89 well, the Due is much faster (~10 times) but the planner is a realy expensive part of Marlin.

Just 2 numbers: With the old 32bit Marlin from bobc (with only some tweaks) you can reach 46k steps/s. And there is just one division in stepper.cpp. With the lookuptable the possible steps/s are doubled to ~96k steps/s.

The only thing which could help is that the buffer can be bigger than in the AVR.

boelle commented 9 years ago

will close this one...

its not a bug with marlin, but rather just a limit on the speed that the hardware can process gcode at

thinkyhead commented 9 years ago

Issues like this can be exacerbated by having a graphical LCD which uses up time, and by printing over USB, which is not as fast and error-free as printing from SD. It's not just limited to the planner. Even if there is some planner optimization, it probably wouldn't be enough to completely eliminate stuttering when resolution is very high and when printing from a host. But when we hit limits like this it's good to know! We can now at least document that such speeds are close to the limits of USB printing for a basic RAMPS (with Mega 2560?).

github-actions[bot] commented 2 years ago

This issue has been automatically locked since there has not been any recent activity after it was closed. Please open a new issue for related bugs.