How to print Gcode from Host

Enrique,
Thanks for the update. I have tried oozebane using the November 20 version and find there is a problem when there are several short steps near the end of the extrusion. I would be pleased to offer a code correction but cannot see how it is intended to spread the slow-down over existing steps. The gcode below is from the diamond.stl example starting at line 108.

With oozebane:
G1 X-0.554 Y-8.309 Z-5.947 F600.0
G1 X8.059 Y0.304 Z-5.947 F600.0
M103
G1 X4.609 Y3.754 Z-5.947 F500.0
G1 X2.487 Y5.876 Z-5.947 F300.0
G1 X0.366 Y7.997 Z-5.947 F100.0
G1 X-28.156 Y43.473 Z-5.947 F450.0
G1 X-27.78 Y43.006 Z-5.947 F600.0
G1 X6.826 Y-0.038 Z-5.947 F600.0
M101
G1 X-0.188 Y6.976 Z-5.947 F600.0
------------------
Without oozebane:
G1 X-0.554 Y-8.309 Z-5.947 F600.0
G1 X8.059 Y0.304 Z-5.947 F600.0
G1 X0.366 Y7.997 Z-5.947 F600.0
M103
G1 X6.826 Y-0.038 Z-5.947 F600.0
M101
G1 X-0.188 Y6.976 Z-5.947 F600.0

I hope this helps.
Jon

Hi Enrique

I now understand the temperature change and have made my first part.

There are still a couple of puzzles in the set up. I am getting raft at F480 but the job sets the feed to F960 which is way to fast on my machine. I had to edit the Gcode to get a result. What sets the feed rates? I have been through all the preferences but I am obviously missing something.

Also the help file for slice mentions extrusion fill density ratio which is no longer in the slice preferences.

Lastly what controls the raft spacing which I think I need to make bigger than the spacing I am getting at the moment.

TIA

---

Regards

Ian

Hi Enrique

Within minutes of my last post I realised what was controlling the feedrate. I still don't see what setting makes the Raft so much slower. With the material I am using CAPA it seems the raft and object should be at a similar speed.

---

Regards

Ian

Hi Ian,

The base feedrate is inversely proportional to the square of the "Base Layer Height over Extrusion Height" ratio. The feedrate is slowed down so that more plastic is deposited in an area, making the thread higher and wider. The default setting is 2, given a base feedrate of one quarter the feedrate preference.

Cheers,
E

Hi Enrique

Thanks again, I am now controlling feed rate OK.

I am slowly getting better & better results as I begin to understand the various functions. I am trying to produce the mug. I noticed one strange thing. I was getting 2 extrusion paths round the mug as I expected and then I got one path round the mug with the extruder off. What puzzles me is the fact that the path with the extruder off is lower than the extruded path so it pushes the plastic out of shape. By deactivating different functions I found that it was comb that produced the extra path. Should it be lower? Or have I got the comb parameters set up screwy. As far as I know I am using the default settings.

---

Regards

Ian

Support material has been added. I don't know how easily the support material can be removed from the object or if the object will stick correctly to itself because it will have cooled while the support is being added, but at least it's there. If anyone uses support and finds temperature which work well, please send em to me so I can make em the defaults.

Thanks to Jon, Marius and Philip a bunch of bugs in oozebane have been found and fixed, the update is attached and in subversion. Jon also spotted a do not cool when tower is on bug which has been fixed. Ian found a comb bug into the object bug which has also been fixed.

The option of adding arrows to skeinview has been added, the default is on.

Hi Enrique

Sorry to advise that the new Raft Preferences window is too long for my screen and I can't reach the save buttons. For the moment I am editing the file. Any chance of making it shorter & wider to accommodate the new controls.

---

Regards

Ian

Hey Ian, I get that too sometimes. In Ubuntu, you can alt-drag the window up with your mouse, then right click the border and re-size it to larger than your screen, so you can get the up-down buttons back.

Also, you can adjust the scaling of the Skeinview window smaller; I don't remember the exact parameter, but it's in there.

Wade

In raft the temperature change times for all kinds of layers are all the same. Would anyone have an objection if there was only one temperature change time?

If there was only one time, the raft window would be smaller and would fit on more screens.

Enrique

Hi Enrique

I find the change time after the raft is built needs to be longer than the build layers. The delay between each build layers I always have the same.

---

Regards

Ian

Hi Ian,

So for the preferences:
Temperature Change Time of Raft
Temperature Change Time of First Layer Outline
Temperature Change Time of First Layer Within
Temperature Change Time of Next Layers
Temperature Change Time of Support Layers
Temperature Change Time of Supported Layers

do you mean that the change time of the First Layer Outline, First Layer Within and Next Layers can be the same?

Enrique

Hi Enrique

I have change time of Raft 0 because my machine does not move until extruder reaches temperature.

I have first layer outline set to say 120 to allow raft to cool as it is done at a higher temperature.

Then first layer within and next layers can be the same maybe 30 seconds only so all build (object) layers (not raft) have the same delay between them.

Support layers I have not used yet so I dont know what will be needed.

Temperature Change Time of Raft is really Temperature Change Time BEFORE Raft &
Temperature Change Time of First Layer Outline is Temperature Change Time BEFORE First Layer Outline etc. as far as I can see. This confused me for a while.

---

Regards

Ian

I always cool my rafts to room temp with a couple of minutes of fan blowing on them. Otherwise the temperature depends on how big it is and so the first layer bonding is very variable. The only way to get the whole raft at a consistent temp is to wait a relatively long time after building it.

When getting the raft to peel easily but still prevent warping, temperature is critical. Certainly 10C makes all the difference and having the raft 10C hotter is the same as the filament being 10C hotter. As the raft cools to room temp with an exponential decay, the last 10C takes a long time.

---

[www.hydraraptor.blogspot.com]

Would anyone have an objection if the parameters Temperature Change Time of First Layer Within and Temperature Change Time of Next Layers were replaced with Temperature Change Time before Next Threads?

Enrique

Sorry for jumping in sideways,

Wade said something that made me think

Quote

I actually get better accuracy running at 8 mm/s and slowing down my extruder, but I usually run it at 16 mm/s just to keep the production rate up.

So is there a way to set different speeds for outline and infill?
And if not, any chance you can add that?

'sid

sid Wrote:
> So is there a way to set different speeds for
> outline and infill?
> And if not, any chance you can add that?
>

This is on my wishlist too

Re. the dialog size discussion:
It feels like the feedrate and flowrate settings should be in another tool than raft. Such a split might also reducethe complexity of this component.

Great work!

---

~/= Marius

--
We are Elektropeople for a better living.
[reprap.soup.io]
[www.metalab.at]

Hi Enrique

It would not affect me at the moment if there was only 1 Temperature Change Time before Next Threads. However in view of the suggestion of having selective feed rate and temperatures in another tool maybe that would be better if it is possible.

One thing has always puzzled me. Why do you use movements without the extruder for delays rather than G4 Pnn where nn is delay usually in seconds. If this was an option it would produce shorter files and you could quickly edit the delays when you first start with new feed stock and need to tweek to get a result.

---

Regards

Ian

A bug in support in raft was fixed, unpause, gcode_step & speed have been added, and the update is attached and in subversion.

Unpause is a skeinforge tool to speed up the feedrate to partially compensate for pausing. It is based on Shane's delay compensation in the gcode firmware, as described at:
[shane.willowrise.com]

The default is off, the delay should only be compensated with unpause or Shane's firmware patch, not both.

Gcode_step is an export plugin which converts the floating point gcode lines into integer step lines. The resulting gcode is still readable, although the scale is off. If the Arduino was changed to accept the step values directly, instead of converting the floating point values to integer steps on its own, the delay would be smaller and so there would be less segment pausing.

Speed is a tool to set the feedrate and flowrate parameters which were set in raft.

Having a different feedrate for the perimeter and infill is a great idea, but since that would interact with the finicky oozebane it will take a while to implement.

Ian,

I use movements without the extruder for delays rather than pausing because when the extruder pauses ooze could accumulate in one place. If you want to shorten the travel, set the "Orbital Feedrate over Operating Feedrate" in speed.py to a low value, like 0.1. Setting it to zero is very bad, a really small number is ok, but not zero.

Thanks for pointing out that there is no "extrusion fill density ratio" in the preferences. It is actually a variable printed out to the console, and I've updated the slice_shape documentation so there's no confusion in future.

Cheers,
Enrique

Hi Enrique

Thanks again for the changes, I am slowly improving quality of the object produced.

I am using hop as it seems to help with my extruder. One thing I have noticed. When you stop extruding and go into the delay loops for cooling the movement there does not retract vertically (I have Hop angle 90) as hop does.The code has the Z with the first X & Y movement which takes the extruder across the Job. The reason I use hop is it stops the extruder dragging over the surface so a vertical movement before the XY would help me. Could hop be added there as well (If it is activated).

---

Regards

Ian

Hi Enrique

I just realised that I am not getting temperature time delay between layers. Should there be a delay regardless of temperature settings? If this may be caused by the options I have activated let me know and I will send a copy of my preferences.

---

Regards

Ian

Hi Ian,

I tried hop out and it's raising the temperature change loop. Is your "Hop over Extrusion Height" greater than zero? The default is one.

There is only a temperature delay for when the temperature changes, which only happens on the raft, the first layer and any layers which have support.

If this still doesn't work, email me your preferences and object to:
perez_enrique@yahoo.com

E

I am trying to produce the shape in the image below but slice_shape throws the following error. Do I need to avoid the 'island' for skeinforge to work? The gts file is attached.


File "D:\gbs\reprap\skeinforgeNov26\skeinforge_tools\slice_shape.py", line 397
, in getSliceGcode
skein.parseTriangleMesh( slicePreferences, triangleMesh )
File "D:\gbs\reprap\skeinforgeNov26\skeinforge_tools\slice_shape.py", line 818
, in parseTriangleMesh
z = self.getZAddExtruderPaths( z )
File "D:\gbs\reprap\skeinforgeNov26\skeinforge_tools\slice_shape.py", line 759
, in getZAddExtruderPaths
loops = self.getLoopsFromMesh( z + zAround )
File "D:\gbs\reprap\skeinforgeNov26\skeinforge_tools\slice_shape.py", line 728
, in getLoopsFromMesh
if len( originalLoops ) < 1:
TypeError: object of type 'NoneType' has no len()

Hi Jon,

Thanks for posting your bug and shape. The updated skeinforge is attached and at:
[members.axion.net]

The shape has holes in it, so it triggered the skeinforge hole spanning code, which I didn't check after a long ago update. If the shape was made in Art of Illusion and you used joins, you could try making the shape with subtraction, which seems to work better. However, although slice_shape throws up lots of warnings, the slice output does seem to be good enough so you can print the shape anyways.

Also the temperature change times in raft have been renamed Temperature Change Time Before ... from Temperature Change Time of ..., so anyone who changed the preferences from the defaults will have to reenter them. The Temperature Change Time before Within Infill and Next Layer were combined into "Temperature Change Time Before Next Threads". The temperature change description in raft is reprinted below.

Cheers,
Enrique



The extruder will orbit for at least "Temperature Change Time Before Raft" seconds before extruding the raft. It will orbit for at least "Temperature Change Time Before First Layer Outline" seconds before extruding the outline of the first layer of the shape. It will orbit for at least "Temperature Change Time Before Next Threads" seconds before extruding within the outline of the first layer of the shape and before extruding the next layers of the shape. It will orbit for at least "Temperature Change Time Before Support Layers" seconds before extruding the support layers. It will orbit for at least "Temperature Change Time Before Supported Layers" seconds before extruding the layer of the shape above the support layer. If a time is zero, it will not orbit.

Enrique,

Thanks for the quick response. I printed a first attempt last night and although my 3D design needs some adjustments the shape was produced despite the warnings.

I was intrigued by the complicated fill pattern on the bottom layers which was quite different to that which I have been seeing in earlier products and the extra switching off and on of the extruder increased the ooze problem. Is the different fill pattern likely to be due to the process used to create the shape in Art of Illusion? Does anyone have any specific advice about the different results from using addition or subtraction?

Jon

Jon,
You need to read Adrian Bowyer's article about AOI here: [reprap.org]

In short you can add or subtract booleans but if there are any coincident faces you create non manifold objects. Since when adding you generally want some concident faces it is easier to make shapes by subtracting.

---

[www.hydraraptor.blogspot.com]

Jon,

I updated the code a while back to fix the almost overlap bug. It created new not fill enough bugs, then I fixed those in the latest version. I suggest using the latest version, and setting the infill pattern in fill.py to the default "Line". If all this don't work, please email me your shape and zipped preferences.

E

Hi Enrique

I am still not getting hop where I expect it after the raft so I have sent the files requested direct to your email. It maybe my expectation is the problem!

---

Regards

Ian

Multiply has been added, which can multiply a shape into a table of duplicated shapes. Skeinforge has been sped up, missing thread bug has been fixed, a z bug has been fixed and the update is attached, in subversion, and at:
[members.axion.net]

Thanks to Ian, a bug in hop was found and fixed.

E

Edited 4 time(s). Last edit at 12/19/2008 01:14AM by Enrique.

I am getting a Skeinforge physical errer on the outputed g-Code. I am coding a ships funnel which has a curved surface with two sloping bands - see photo.

If you look at the capture there is a notch band in the middle of each of these funnel bands. You can see them on the photo and in the screen capture of the G-code visual output from Skeinforge.


The STL file is clean.


Any thoughts?

I may be missing something obvious.

I have the Skeinforge Art of Illusion scripts and the program installed.

How on earth do you import the scripts. the script manager only imports from the web libraries and cannot be told to look locally.

Help please.