Marlin's Last Hurrah?

Marlin's Last Hurrah?

Postby Mooselake » 2017-May-Thu-12-May

Still some life left in the old beast. Is this the last major release before the 32/64 bit firmware drives Marlin into irrelevance?

http://forums.reprap.org/read.php?415,765902


Kirk
  • 0

Modified KickStarter Classic Plus 7/2012
KS Thingybot Delta Pro 10/31/16
User avatar
Mooselake
My next printer is...
 
Posts: 3501
Joined: 2011-Dec-Tue-23-Dec
Location: Moose Swamp, Upper Michigan, USA
Reputation: 170

Marlin's Last Hurrah?

Sponsor

Sponsor
 

Re: Marlin's Last Hurrah?

Postby PxT » 2017-May-Thu-12-May

Well I guess it won't really be irrelevant as long as older machines are out there in the world. It's not like everyone is suddenly going to upgrade their control board to 64-bit.
Looks like a nice set of improvements anyway. Who wants to save me the trouble and build a version for the Printrbot Simple Metal v1? :)
  • 0

User avatar
PxT
Print complete...
 
Posts: 1333
Joined: 2011-Dec-Mon-16-Dec
Location: Sacramento, CA
Reputation: 96

Re: Marlin's Last Hurrah?

Postby RetireeJay » 2017-May-Thu-16-May

Hmm... That Linear Advance looks interesting to me. I don't do probing, and don't have a hardware screen attached, so most of the list is meaningless to me.
  • 0

Printrbot Plus operational January 2013
Brass threaded rods (5/16" X 18) & nuts for Z axis
GT2 belts & pulleys
Cable chain to reduce probability of fatigue failure in wires
E3D V5 Hot End, 0.4mm nozzle, also 0.8 and 0.25 in use occasionally
PB fan mount + 40mm fan -- using printed mount adapter, not the E3D supplied fan
Injection molded extruder gears
Optical Z "endstop" (custom designed and built)
Have used many pounds of T-Glase filament. Now also doing some work with Ninjaflex SemiFlex
Print on glass with Scotch Craft Stick or other glue stick
User avatar
RetireeJay
My next printer is...
 
Posts: 4767
Joined: 2013-Jan-Wed-13-Jan
Location: Greenville, SC
Reputation: 477

Re: Marlin's Last Hurrah?

Postby PxT » 2017-May-Fri-23-May

https://github.com/PxT/Marlin/tree/1.1.x/

Well, there's a pre-configured version for the original Metal Simple if anyone wants it. If your particular machine moves/homes differently you may have to tweak a few settings to make it work. It took me quite a bit of fiddling to get everything configured the right way -- those darn INVERT and HOME_DIRECTION settings get me every time.

Feel free to give it a whirl. It's configured for AUTO_BED_LEVELING_BILINEAR which uses the Z-probe and checks a grid, but you can switch that option easily enough in Configuration.h if needed.


As far as the results go - I printed a Benchy with the original "Unified" firmware, it took 1h 17m 48s. Uploaded the new firmware with all default jerk & acceleration settings and it finished the identical gcode in 1:03:19, about 20% faster. The default speeds & feeds are quite a bit faster than what I had configured in the previous version.

The surface finish on the 1.1.x firmware is noticeably better in certain places. The unified version shows some spots where the infill deformed the outside surface. It's tiny but you can feel it. The 1.1.x version is perfectly smooth in those areas.

The first few layers of the 1.1.x are very slightly lower quality than the rest of the model, which is not the case on the unified version.

Other than that the prints are almost identical. I'll try to get some photos of all this tomorrow.

M78 is kinda neat:
Code: Select all
Send: M78
Recv: Stats: Prints: 2, Finished: 1, Failed: 1
Recv: Stats: Total time: 1h 6m 16s, Longest job: 1h 3m 16s
Recv: Stats: Filament used: 4.31m
  • 0

User avatar
PxT
Print complete...
 
Posts: 1333
Joined: 2011-Dec-Mon-16-Dec
Location: Sacramento, CA
Reputation: 96

Re: Marlin's Last Hurrah?

Postby RetireeJay » 2017-May-Sat-05-May

PxT wrote:The unified version shows some spots where the infill deformed the outside surface. It's tiny but you can feel it. The 1.1.x version is perfectly smooth in those areas.

Why? I'd think that's primarily a function of the slicer, because Marlin has no knowledge of whether there's a nearby trace or not; it can't adjust for that. Or is the improvement due to that Linear Advance?
  • 0

Printrbot Plus operational January 2013
Brass threaded rods (5/16" X 18) & nuts for Z axis
GT2 belts & pulleys
Cable chain to reduce probability of fatigue failure in wires
E3D V5 Hot End, 0.4mm nozzle, also 0.8 and 0.25 in use occasionally
PB fan mount + 40mm fan -- using printed mount adapter, not the E3D supplied fan
Injection molded extruder gears
Optical Z "endstop" (custom designed and built)
Have used many pounds of T-Glase filament. Now also doing some work with Ninjaflex SemiFlex
Print on glass with Scotch Craft Stick or other glue stick
User avatar
RetireeJay
My next printer is...
 
Posts: 4767
Joined: 2013-Jan-Wed-13-Jan
Location: Greenville, SC
Reputation: 477

Re: Marlin's Last Hurrah?

Postby PxT » 2017-May-Sat-09-May

Pretty sure that's the linear advance they are talking about. I used the exact same gcode each time.

And just for the record this is something that is barely perceptible visually on the unified print, I doubt I can even get a picture of it, but you can feel it as a very tiny unevenness on the surface.
  • 0

User avatar
PxT
Print complete...
 
Posts: 1333
Joined: 2011-Dec-Mon-16-Dec
Location: Sacramento, CA
Reputation: 96

Re: Marlin's Last Hurrah?

Postby RetireeJay » 2017-May-Sat-11-May

EDIT I have uploaded the code to my printer now, and here's what I needed to go to get it to compile and have correct motor directions and endstop sensing (Under PlatformIO)

- Install the PlatformIO directory (see the big thread about that)

To keep track of mods, I try to document them near the top of the file...
- In Configuration.h:
Code: Select all
  * Basic settings such as:
 *   **Revisions by JCS 6 May 2017
 * - Type of electronics : **Printrboard Rev F
 * - Power Supply Switch:  **No switch
 * - Type of temperature sensor : **Hot end = 5; Bed = 1 (Using E3D hot end with its provided thermistor)
 * - Adjusted PID tuning for my hot end: 22.41, 1.70, 73.8
 * - Endstop polarity (?): **Set to "false" for X & Y, "true" for Z (my system uses a "switch")
 * - Steps Per Unit: ** 80, 80, 2267.7, 578.13
 * - Max Feedrate: **Reduced Z max rate to 5 from 10
 * - Max Acceleration:  **Left at high rates for now
 * - All Stepper directions should be set to "true"
 * - Printer geometry: Travel Limit for Y = 0, not 20
 *                     X_Max = 187; Y_Max = 189; Z_Max = 200
 * - Preheat fan speed **230 (90%)


- In Configuration_adv.h:
Code: Select all
* - Enabled DAC_MOTOR_CURRENT_DEFAULT, with values 60, 65, 60, 55
  • 0

Printrbot Plus operational January 2013
Brass threaded rods (5/16" X 18) & nuts for Z axis
GT2 belts & pulleys
Cable chain to reduce probability of fatigue failure in wires
E3D V5 Hot End, 0.4mm nozzle, also 0.8 and 0.25 in use occasionally
PB fan mount + 40mm fan -- using printed mount adapter, not the E3D supplied fan
Injection molded extruder gears
Optical Z "endstop" (custom designed and built)
Have used many pounds of T-Glase filament. Now also doing some work with Ninjaflex SemiFlex
Print on glass with Scotch Craft Stick or other glue stick
User avatar
RetireeJay
My next printer is...
 
Posts: 4767
Joined: 2013-Jan-Wed-13-Jan
Location: Greenville, SC
Reputation: 477

Re: Marlin's Last Hurrah?

Postby RetireeJay » 2017-May-Sat-21-May

I created a small test piece with straight lines, circles, and spaces, and created G-code with Slic3r. I ran it on my existing firmware, then again using the new Marlin 1.1. To tell the truth, I can't tell any difference visually or by touch between the two samples. I was using 20% infill, and 4 traces for the perimeter, and there's no evidence on either sample of the infill causing a disturbance on the outside. Even the dimensions of circles (inside and outside) measure the same.
  • 0

Printrbot Plus operational January 2013
Brass threaded rods (5/16" X 18) & nuts for Z axis
GT2 belts & pulleys
Cable chain to reduce probability of fatigue failure in wires
E3D V5 Hot End, 0.4mm nozzle, also 0.8 and 0.25 in use occasionally
PB fan mount + 40mm fan -- using printed mount adapter, not the E3D supplied fan
Injection molded extruder gears
Optical Z "endstop" (custom designed and built)
Have used many pounds of T-Glase filament. Now also doing some work with Ninjaflex SemiFlex
Print on glass with Scotch Craft Stick or other glue stick
User avatar
RetireeJay
My next printer is...
 
Posts: 4767
Joined: 2013-Jan-Wed-13-Jan
Location: Greenville, SC
Reputation: 477

Re: Marlin's Last Hurrah?

Postby RetireeJay » 2017-May-Sun-20-May

Hmm... discovered that the Linear Advance feature is off by default. It's at about line 620 in Configuration_adv.h
I just turned it on and compiled; it adds about 8k bytes to the code. It's getting late today; will upload to the printer and try the same G-code again tomorrow to see if there's an improvement. They say in the comments that it will work well in automatic mode with G-code created by Slic3r, but is not guaranteed to work well with code generated by Cura.
  • 0

Printrbot Plus operational January 2013
Brass threaded rods (5/16" X 18) & nuts for Z axis
GT2 belts & pulleys
Cable chain to reduce probability of fatigue failure in wires
E3D V5 Hot End, 0.4mm nozzle, also 0.8 and 0.25 in use occasionally
PB fan mount + 40mm fan -- using printed mount adapter, not the E3D supplied fan
Injection molded extruder gears
Optical Z "endstop" (custom designed and built)
Have used many pounds of T-Glase filament. Now also doing some work with Ninjaflex SemiFlex
Print on glass with Scotch Craft Stick or other glue stick
User avatar
RetireeJay
My next printer is...
 
Posts: 4767
Joined: 2013-Jan-Wed-13-Jan
Location: Greenville, SC
Reputation: 477

Re: Marlin's Last Hurrah?

Postby RetireeJay » 2017-May-Mon-09-May

Ran my test print with Linear Advance turned on. In most places, the part is identical to the two that I ran previously. But in a couple of areas, the test with Linear Advance on is slightly worse than the others; it's rougher. The differences between "original", 1.1 without Advance and 1.1 with Advance are extremely subtle; you can barely discern them in this picture (in left to right order). I took this picture with lighting and camera angle arranged to emphasize the defects; looking at these parts normally they all look pretty good. There was no difference in stringing between the three versions.
Comparison Photo.JPG

I'm going to stay with the 1.1 without Linear Advance turned on.
  • 0

You do not have the required permissions to view the files attached to this post.
Printrbot Plus operational January 2013
Brass threaded rods (5/16" X 18) & nuts for Z axis
GT2 belts & pulleys
Cable chain to reduce probability of fatigue failure in wires
E3D V5 Hot End, 0.4mm nozzle, also 0.8 and 0.25 in use occasionally
PB fan mount + 40mm fan -- using printed mount adapter, not the E3D supplied fan
Injection molded extruder gears
Optical Z "endstop" (custom designed and built)
Have used many pounds of T-Glase filament. Now also doing some work with Ninjaflex SemiFlex
Print on glass with Scotch Craft Stick or other glue stick
User avatar
RetireeJay
My next printer is...
 
Posts: 4767
Joined: 2013-Jan-Wed-13-Jan
Location: Greenville, SC
Reputation: 477

Re: Marlin's Last Hurrah?

Postby Mooselake » 2017-May-Mon-09-May

Slic3r's autospeed would seem to do pretty much the same thing, no worry about pressure changes if you're extruding at a near-constant flow rate. I fudge it a bit, perimeters at 40mm/sec for a better surface finish, otherwise 120mm/sec max.

Your middle piece looks best to me, especially the corners and edges of the slot. It looks thinner in the pic, although that might just be the camera angle. Have you tried another perimeter to smooth out the infill bumpies, or is that lighting?

Kirk
  • 0

Modified KickStarter Classic Plus 7/2012
KS Thingybot Delta Pro 10/31/16
User avatar
Mooselake
My next printer is...
 
Posts: 3501
Joined: 2011-Dec-Tue-23-Dec
Location: Moose Swamp, Upper Michigan, USA
Reputation: 170

Re: Marlin's Last Hurrah?

Postby RetireeJay » 2017-May-Mon-11-May

Mooselake wrote:Your middle piece looks best to me, especially the corners and edges of the slot.

Yup, that's why I went back to Marlin 1.1 without the Linear Advance feature turned on

Mooselake wrote:It looks thinner in the pic, although that might just be the camera angle.

Yes, just camera angle. The three pieces are exactly the same in all dimensions.

Mooselake wrote: Have you tried another perimeter to smooth out the infill bumpies, or is that lighting?

These pieces were printed with four perimeters, which should be enough to hide the corners of the infill. The material is T-Glase and it's translucent, not opaque, so some of what you see is not actually on the surface.

The direction of printing on the top surface in the picture is from the near corner toward the far corner. This is on the Y axis. I have always had a slight ringing effect on both X and Y axes immediately after a sharp corner. It fades away in about 5mm. I THINK it's caused by mechanical vibration somewhere in the 'bot structure. It gets a little better if I cut the acceleration down, but it doesn't seem to go away. It's been more than a year since I experimented with this phenomenon - I can't remember if the wavelength of the ringing as printed changes with the print speed, as might be expected if the vibration of the 'bot is at a constant resonant frequency but the trace is laid down at a different speed. I finally concluded that for the things I'm making I don't care about that slight ringing. Good enough is good enough. In forgiving lighting, the effect is invisible.

By the way, in reading Configuration.h there's a section where the authors are saying that some lines there will get printed out upon boot-up. But that's never worked for me; you can't connect Repetier to a device that's not there to see what it says when the device does appear. I guess the lines might show up on an LCD screen if attached. When I issue an M115 command, I get an ID string that is mostly produced by code in the language.h module. But I did discover a nice feature in this version of Marlin 1.1. I can uncomment a line in Configuration.h called "CUSTOM_MACHINE_NAME" and insert my own parameter. Now when I issue an M115 command, it shows PRINTER TYPE as "Jay's Printrbot Plus"!
  • 0

Printrbot Plus operational January 2013
Brass threaded rods (5/16" X 18) & nuts for Z axis
GT2 belts & pulleys
Cable chain to reduce probability of fatigue failure in wires
E3D V5 Hot End, 0.4mm nozzle, also 0.8 and 0.25 in use occasionally
PB fan mount + 40mm fan -- using printed mount adapter, not the E3D supplied fan
Injection molded extruder gears
Optical Z "endstop" (custom designed and built)
Have used many pounds of T-Glase filament. Now also doing some work with Ninjaflex SemiFlex
Print on glass with Scotch Craft Stick or other glue stick
User avatar
RetireeJay
My next printer is...
 
Posts: 4767
Joined: 2013-Jan-Wed-13-Jan
Location: Greenville, SC
Reputation: 477

Re: Marlin's Last Hurrah?

Postby PxT » 2017-May-Tue-10-May

RetireeJay wrote:Hmm... discovered that the Linear Advance feature is off by default.


Image

So I didn't even have it enabled on mine. Not sure what to make of my results now. I'm going to enable it and see what happens. Did you print a series of test cubes to calibrate the K value as they suggest on http://marlinfw.org/docs/features/lin_advance.html ?
  • 0

User avatar
PxT
Print complete...
 
Posts: 1333
Joined: 2011-Dec-Mon-16-Dec
Location: Sacramento, CA
Reputation: 96

Re: Marlin's Last Hurrah?

Postby RetireeJay » 2017-May-Tue-11-May

No, I didn't do that exercise (didn't know about it, in fact).
BUT... Is K available in EEPROM? I don't want to be re-compiling and re-flashing every time I switch to a new filament or change speeds or change nozzle size!
  • 0

Printrbot Plus operational January 2013
Brass threaded rods (5/16" X 18) & nuts for Z axis
GT2 belts & pulleys
Cable chain to reduce probability of fatigue failure in wires
E3D V5 Hot End, 0.4mm nozzle, also 0.8 and 0.25 in use occasionally
PB fan mount + 40mm fan -- using printed mount adapter, not the E3D supplied fan
Injection molded extruder gears
Optical Z "endstop" (custom designed and built)
Have used many pounds of T-Glase filament. Now also doing some work with Ninjaflex SemiFlex
Print on glass with Scotch Craft Stick or other glue stick
User avatar
RetireeJay
My next printer is...
 
Posts: 4767
Joined: 2013-Jan-Wed-13-Jan
Location: Greenville, SC
Reputation: 477

Re: Marlin's Last Hurrah?

Postby PxT » 2017-May-Tue-17-May

Sure is, they document it at that link. M905 K<value>
  • 0

User avatar
PxT
Print complete...
 
Posts: 1333
Joined: 2011-Dec-Mon-16-Dec
Location: Sacramento, CA
Reputation: 96

Re: Marlin's Last Hurrah?

Postby RetireeJay » 2017-May-Tue-21-May

Just being nitpicky here, but the K value is not stored in EEPROM. Here's what the document says:
When you have found a good value, there are two ways to make them permanent in your firmware. If you are using only one filament material, the best way is to set the K value inside Configuration_adv.h and reflash the firmware. If you are using different materials, you might want to insert a “M905 K..” line into the start code inside your slicer profile used for the corresponding material.

When they say "if you are using only one filament material" I suppose it should also say "and one nozzle size, and one standard speed for that material/nozzle combination, and one standard layer thickness too." ;)

The documentation inside Configuration_adv.h says
* Set to 0 to auto-detect the ratio based on given Gcode G1 print moves.
*
* Slic3r (including Prusa Slic3r) produces Gcode compatible with the automatic mode.
* Cura (as of this writing) may produce Gcode incompatible with the automatic mode.


Although the more I study the code the more I become confused as to the actual usage of ADVANCE, LIN_ADVANCE, LIN_ADVANCE_K, and LIN_ADVANCE_E_D_RATIO. Right now I have them all off.
  • 0

Printrbot Plus operational January 2013
Brass threaded rods (5/16" X 18) & nuts for Z axis
GT2 belts & pulleys
Cable chain to reduce probability of fatigue failure in wires
E3D V5 Hot End, 0.4mm nozzle, also 0.8 and 0.25 in use occasionally
PB fan mount + 40mm fan -- using printed mount adapter, not the E3D supplied fan
Injection molded extruder gears
Optical Z "endstop" (custom designed and built)
Have used many pounds of T-Glase filament. Now also doing some work with Ninjaflex SemiFlex
Print on glass with Scotch Craft Stick or other glue stick
User avatar
RetireeJay
My next printer is...
 
Posts: 4767
Joined: 2013-Jan-Wed-13-Jan
Location: Greenville, SC
Reputation: 477


Return to General

Who is online

Users browsing this forum: No registered users and 1 guest

cron