Cura profile for Selpic Star A

[foo_bar 1]

I post a recommended profile for Selpic Star A.11.curaprofileSettingManual.pdf

Thanks for Selpic's support.

[foo_bar 1]

At retraction test, The speed had to be set to 30 mm / sec.

Original site: https://www.thingiverse.com/thing:4825787

 

towers.gcode

[foo_bar 1]

I made fillenium malcon.

Parameters

model: fillenium malcon https://www.thingiverse.com/thing:919475 thanks!

Layer: 0.1mm

height:112.8mm

infill: 10%

Print speed:60mm/s

Travel speed:120mm/s

Print Temp:190℃

Bed Temp:60℃

Bed leveling: auto https://www.thingiverse.com/thing:4831530

time:48hour

 

robin_nano_cfg.txt

>DEFAULT_ACCELERATION        11    #X, Y, Z and E acceleration for printing moves
>DEFAULT_XJERK             8    #Default Axis-X Jerk (mm/s)
>DEFAULT_YJERK             8    #Default Axis-Y Jerk (mm/s)
 

At first glance, the speed seems to be too fast, but in reality the acceleration and jerk are so small that it does not reach the maximum speed. Acceleration and jerk values were too small. I should have set it to a higher value.

[foo_bar 1]

[nursereese 0]

Thanks for the profile and gcode. Works well for me. I'll have to try the cfg edits later today.

Edited May 9, 2021 by nursereese

[mikeramos 0]

On 4/12/2021 at 2:30 PM, foo_bar said:

I post a recommended profile for Selpic Star A.11.curaprofileSettingManual.pdf

Thanks for Selpic's support.

Thank you! It got it up and running easily with this. Much appreciated.

[eduo 0]

On 4/17/2021 at 9:40 AM, foo_bar said:

At retraction test, The speed had to be set to 30 mm / sec.

Original site: https://www.thingiverse.com/thing:4825787

 

towers.gcode 4.54 MB · 10 downloads

 

Hi. I couldn't help notice you got the heated bed installed.

 

I got mine as well as the LCD but I just can't find how they're supposed to be installed. Do you have a link to the instructions or a quick summary of them?

[Anto777 0]

hi, i downloaded the profile but cura said it is an invalid or broken file. It has a unknow format

[eduo 0]

On 5/30/2021 at 8:50 PM, Anto777 said:

hi, i downloaded the profile but cura said it is an invalid or broken file. It has a unknow format

Rename the file so it has .cura extension.

 

Alternatively, use the manufaturer's: https://drive.google.com/drive/folders/1JNS2ilYK1HiQA0XcJVHyMyV6n_wgWoYF?usp=sharing

 

Extract the corresponding zip contents and place in the root of the SD card.

[eduo 0]

Found how to update the config file. Printer is astoundly slow and noisy. The latter seems to be because there's a thick wire in the path between the fan and the heater. I wonder if that could be modded out.

 

At any rate, fan speed seems to be hardcoded at 100%. Can't be reduced via the config file or onscreen, as far as I can see

Edited June 2, 2021 by eduo
EDITED due to stupid question that got resolved immediately.

[davidbarros 0]

Hey guys! 

I'm using Cura, and it's previewed correctly, but when printing it's not correct.

https://imgur.com/a/Bmdy2aK

 

 

Am I doing something wrong? I've followed the SettingManual on first post, but this is the result.

I've tried to convert the STL to gcode with Cura and realized that with the file from foo_bar has ~169.000 lines, and the one I'm exporting form Cura only has ~46.000 lines. 

 

Can anyone help me? I'm stuck with the test files :(

 

[gr5 2,170]

That is severe underextrusion.  Something isn't set up right.  For example you may have cura slicing for 3mm filament yet you may be using 1.75mm filament (that's a common error right there).

 

What kind of printer is it?

[davidbarros 0]

4 hours ago, gr5 said:

That is severe underextrusion.  Something isn't set up right.  For example you may have cura slicing for 3mm filament yet you may be using 1.75mm filament (that's a common error right there).

 

What kind of printer is it?

Thanks for the answer, It's a Selpic Star A, I should set that under "machine settings" and "Extruder 1" right? It's getting 2.85, but before the print I set it to 1.75.  

[gr5 2,170]

You have to tell cura what kinds of filament the printer will accept - that's in machine settings.

 

Then the filament itself has settings.  That's more critical.  The filament selected has to match what you actually use.

 

If you think you did it right then you can look at any two successive gcodes that include X,Y,E values.  Knowing the layer height multiply line width X layer height X distance (use distance formula knowing the from and to point in X,Y distance).  That gets you the *volume* of filament you want to extrude in cubic mm. 

 

To translate to E value for 1.75mm filament you divide that volume just calculated by the cross sectional area of the filament (pi r squared) or 2.4 square mm.   So you could test that Cura is doing it right before trying to print the gcode.

 

I mean there are over 500 parameters in cura and many could be messing you up.  Such as "flow" settings.

 

Or it's possible the E values are perfect and your printer has something wrong with it.  Or you are just printing too cold and too fast.

 

[davidbarros 0]

Hey  GR5, 

I don't believe that the problem is in the printer, I'm doing something wrong during the exportation from CURA. The test models, and the towers that are on this thread print perfectly (foo_bar post towers.gcode).

 

For example, can you provide me a gcode extracted from CURA to test? Thx 

[GregValiant 1,250]

@gr5 pointed out the most popular missed setting (filament diameter) but there is a 2nd most popular missed setting.

When Cura calculates the amount of filament needed to produce a given length of extrusion, it uses the Filament Diameter of the Material.  Generally, that results in the Gcode E values in "MM" of filament. 

If that printer is running Marlin firmware then it has an option under "Control | Filament" of "E in mm3".  Cura and the printer need to be talking Apples and Apples or Cura might be saying "MM" but the printer is reading "mm³".  So make sure that in the printer "E in mm3" is turned off, and that if the printer is running Marlin firmware then check the Machine Settings and make sure the Gcode Flavor is set to Marlin and NOT "Marlin/Volumetric".  Whether Volumetric is turned ON or OFF - Cura and the printer need to match.  If Cura says E1.0 and meant 1mm of filament, that would be 2.4mm³.  If the printer is set to Volumetric then the same E1.0 would be read as 1.0mm³ of filament and so you can see it would be under-extruding by about 58%.

Edited June 4, 2021 by GregValiant

[davidbarros 0]

Hey Greg, I've just confirmed and it's both in mm.

Thanks anyway . 

 

Can anyone with a Star A printer pass me a gcode file exported from cura that works ? 

Thanks in advance

[gr5 2,170]

Did you check that the *filament* settings in cura are set to 1.75mm.  Not the machine settings.  Not the profile settings.  The *filament* settings.  In cura.

[Ihorschd 2]

Received my printer two weeks ago, everything in the package (touchscreen, heatbed). Assembled it, did the leveling stuff and printed the rocket. One thing I noticed was the power plug. It’s not very tight and depending on the rotation the printer turns off, but I got used to this. 
As everybody said it is quite slow but the print of the rocket was good at all. 
I updated the software described by @foo bar.
I thought a good idea is to print the wide known calibration cube (Thingiverse) to check the dimensions. Very soon i noticed a big layer shift.
I thightened the belts to fix it, the result was better but by far not as good as the rocket.
The next thing I did was printing two models from Thingiverse. A small catapult and another rocket. Guess what: the layer shift was greater than ever before. Checking the belts again didn’t fix the problem. After looking all the comments I found the hint with the  robin_nano_cfg.txt
> DEFAULT_ACCELERATION 11
> DEFAULT_XJERK 8
> DEFAULT_YJERK 8
This helped a lot. The models came out ok. Not perfect but I think we can’t expect too much.
I’m planning to change the values to find the best results because the speed is again slower. Maybe I will upgrade to the other firmware from @foo bar?!
One thing to say:
After just 2-3 times inserting and removing the SD card, a small metal piece fell out. The result was that the card will not ‘click’ in any more.
Because if this I contacted the support. The response was quick. they wanted a video of the problem, fine. They told me that the main board has to be replaced, but I will get a complete new printer. 
Because of the long delay, the bad communications I was a little skeptical. Surprise! Today (after 5 days) I got another Amazon package with a brand new printer.
Sorry for the long post but maybe it’s interesting for somebody of you.

[GregValiant 1,250]

Thanks for the report.  Sometimes we wonder what happened as people get their problems resolved but don't report back.

 

The all-new printer may not have the problems with layer shifting.  It is always a mechanical problem and can be the belts, loose screws on the drive pulleys, insufficient voltage reference from the mainboard, etc.  It can be tough to chase down.

 

The cantilever design makes for a compact unit but the starts and stops of the print head can shake the machine.  The fix is to keep Acceleration control setting very low which results in slow prints.  I wrote a little application for calibrating Accel and Jerk.  With a 120 x 120 build plate and the Accel at 11 and Jerk at 8, the max speed you can get is 24mm/sec.  You can set Print Speed higher in Cura, but at an Accel of 11, 24mm/sec is the best it will do (16mm/sec around a circle).  Trying to speed up the machine by increasing the Accel results in more shaking and directly affects print quality.  I would think you could go to 100 though and that would result in a possible speed of 55mm/sec. My Ender runs very well at an Accel of 500.  At 3000 it's pretty shaky and the quality on curves and corners definitely suffers.  An industrial strength machine like an Ultimaker can take a lot more.

 

If you  get one of these SD to Micro SD adapters, you can use an SD card instead of fooling with the little micro SD.  A second benefit is that it stays in the printer all the time and doesn't wear-out the memory card slot.

[eduo 0]

On 6/4/2021 at 5:40 PM, davidbarros said:

Hey Greg, I've just confirmed and it's both in mm.

Thanks anyway . 

 

Can anyone with a Star A printer pass me a gcode file exported from cura that works ? 

Thanks in advance

I'm in this same situation. If I print the buit-in gcodes (Vase and Rocket) they come up perfectly.

 

Even trying the profile posted and all tweaks, all my printgs are seriously underextruded. Changing filament width didn't help. 

[GregValiant 1,250]

Have you calibrated the E-steps?  (I don't actually know if it's even possible on that printer.)

 

When Cura calculates the E-value for a particular extrusion it's "E = LineWidth x LayerHeight x ExtrusionLength / (PI x FilamentRadius^2) and that gets defined as 100% Flow.  So when your printer sees an E value it must move from the previous E-location to the new E-location and provide the correct number of E-Steps so the extruder pushes the correct millimeters of filament.  It is the E-Steps/mm that insures the correct volume of filament was supplied by the extruder.

The variables then are the Filament Diameter, the E-Steps/mm, and the Flow % (which is a fudge factor).

[eduo 0]

1 hour ago, GregValiant said:

Have you calibrated the E-steps?  (I don't actually know if it's even possible on that printer.)

 

When Cura calculates the E-value for a particular extrusion it's "E = LineWidth x LayerHeight x ExtrusionLength / (PI x FilamentRadius^2) and that gets defined as 100% Flow.  So when your printer sees an E value it must move from the previous E-location to the new E-location and provide the correct number of E-Steps so the extruder pushes the correct millimeters of filament.  It is the E-Steps/mm that insures the correct volume of filament was supplied by the extruder.

The variables then are the Filament Diameter, the E-Steps/mm, and the Flow % (which is a fudge factor).

 

I tried first with the profile for the printer linked above, which has a default value supposedly ad-hoc to the printer. I then tried tweaking it. I'll try contacting the manufacturer, it's so far from ideal there must be something else in play.

[ais80 0]

Hello all, any advice would be great

I have the Star-A printer. Upgraded to the official firmware posted in May by Foobar. Originally I had Cura setup as RepRap as per the instructions set by the company. Prints weren't perfect but ok. After the first firmware update, my filament started blobbing (coming out thick and thin, not consistent, gear slipping as it was pushing out too hard. Next firmware update in May, updated, looked at the new instruction posted above and downloaded the cura profile. Instructions now state Marlin. Made sure Cura matched. Went to do a print and...(see the pink prints in pic 1)

 

Ive seen layer shift, but this is a whole new level. the left was take 1 with jerk and acceleration control. Right hand one shifted after about 1mm and started printing about 20mm off where it should. Both the RepRep and Marlin have extreme layer shifting. I'm confused as to what the cause could be, I've tightened all the belts and screws. the two Calicats that completed were printed out on the first firmware. Ever since then its been downhill and i cant figure out why.

 

Support has stopped relying to any of my emails. I'm hesitant to install the custom Marlin fork that Foobar has kindly compiled as i also have a laser engraver (not yet used but want to). My start G-code is:

;Sliced at: {day} {date} {time}

;Basic settings: Layer height: {layer_height} Walls: {wall_thickness} Fill: {fill_density}

;Print time: {print_time}

;Filament used: {filament_amount}m {filament_weight}g

;Filament cost: {filament_cost}

M140 S{material_bed_temperature_layer_0} ;Set bed temperature

M104 S{material_print_temperature_layer_0} ;Set hotend temperature

M190 S{material_bed_temperature_layer_0} ;Wait for bed temperature

M109 S{material_print_temperature_layer_0} ;Wait for hotend temperature

;Next 3 values were set in the provided gcode profiles from selpic

G21 ;metric values

G90 ;absolute positioning

M82 ;set extruder to absolute mode

M107 ;start with the fan off

G28 X0 Y0 ;move X/Y to min endstops

G28 Z0 ;move Z to min endstops

G1 Z15.0 F{travel_speed} ;move the platform down 15mm

G92 E0 ;zero the extruded length

G1 F200 E3 ;extrude 3mm of feed stock

G92 E0 ;zero the extruded length again

G1 F{travel_speed}

;Put printing message on LCD screen

M117 Printing...

 

 

[GregValiant 1,250]

That's the basic vanilla Start Gcode.  Nothing problematic there.  The Acceleration recommended by the Selpic people is really low.

Insert the following lines right after the M107 line in you start gcode:

M201 X11 Y11

M203 X50 Y50

The M201 sets the max Accel and the M203 sets the max speed.  Those figures will be the limits that the printer imposes on itself when it calculates the moves.  If you get a decent print then you can start moving the Accel up in increments of about 10.  At a low Accel value like that (Ultimakers may be able to take 10000) it will have a huge effect on the speed so moving the speed up won't really affect anything.