Java host software - Number of Shells - circular movement

Okay, because it's the most intuitive host software around, I have been using the Java host software. But it gives me some headaches...

The "Number of Shells (0..N)" parameter does not always work as expected.

If I choose 0 as value, the software crashes on the first printing layer with an ArrayOutOfBoundsException.
Setting this to 1 does not give this problem, but doesn't work well for circular movements.

I noted, especially on the first actual printing layer, that the shell is printed around too fast. The print head does a circular movement, but the filament does not have time to stick to the raft which results in a almost straight line.


Another question: Is there any particular reason why the reprap software always prints diagonal lines? I don't really understand why X&Y motors have to be in motion all the time.

---

-------------
FiveD Firmware fork (supports Arduino Mega): [github.com]
reprapsource.com

Number of shells is the number of outlines surrounding the infill, 0 makes no sense s you object would have no smooth perimeter. I don't see the connection with this and circular motion?

If the filament doesn't stick to the raft then it needs to be hotter, or the nozzle lower, or the speed lower relative to the extrusion rate, so that the filament is not stretched as much.

By diagonal lines do you mean the infill pattern? Diagonal has the advantage of being 1.4 times faster, or the motors running 1.4 times slower.

---

[www.hydraraptor.blogspot.com]

nophead Wrote:
-------------------------------------------------------

> If the filament doesn't stick to the raft then it
> needs to be hotter, or the nozzle lower, or the
> speed lower relative to the extrusion rate, so
> that the filament is not stretched as much.

I extrude ABS plastic at 230°C at the nozzle's orifice. While extruding, the temperature is pretty much constant by +-3°C.

Both speed and nozzle height makes sense. Maybe I have to start to print lower on the first layer. But on the other hand, the infill on the first layer works fine..

I cannot find a parameter that reduces speed on the shield.

I guess I've to "learn" to use Skeinforge soon.

> By diagonal lines do you mean the infill pattern? Diagonal has the advantage of being 1.4 times faster, or the motors running 1.4 times slower.

Well, makes sense for some kinds of stepper motors. Maybe I can try Extruder0_EvenHatchDirection and Extruder0_OddHatchDirection

---

-------------
FiveD Firmware fork (supports Arduino Mega): [github.com]
reprapsource.com

Diagonal lines are done in skein forge for infill and shell layers, i don't us reprap host software that much..,

I could be wrong but it seems like it is done for strength and rotates 45deg per layer. I would recommend a shell that is at least 2 thick to add strength at both 45 and 135 deg.

I also would do Diaphragm Period every 5 layers as this adds support as well. grid separation should be 5 or 10 for this.

temp is not as important to change as people say here. In fact it may cause you headaches!

you want a temp that your machine can extrude plastic with lower viscous liquid reliably and what temp it can consistently keep. temp variation plays an important part in bridging across open space, and in warping of plastic. so for stability of your machine. yes temp is important.

I find that if a part is not sticking to the raft then bringing the build closer to the raft works best. you got to think about heat transferred from the extruded plastic to the plastic already built. the more surface area that is in contact the more heat is transferred before the surrounding are cools it. so squishing it a little causes a greater amount of heat energy to move to the plastic on the raft and further bonds the two.

think about temp of environment (room temp), heat of transferred material and of nozzle height. if you raise your nozzle temp to say 240deg, and your device can not maintain that temp, then your builds will suffer from the temp variation.

I guess i just wanted people to think about temp as not so much the extruder temp, but the fact that they set the machine up to be optimum at one temp. then they have a sticking to raft problem, and change the temp. well then all your other settings need to be changed to compensate.

changing the distance the build is from the raft, only changes that one thing, and probably will work better, and not leave you frustrated with all the mess from the temp change.


so in troubleshooting change distance of build from raft rather than mess with temp. you mess with your temp, be prepared to change all your speeds and flow settings.



hope it helps, and good luck!

After setting EvenHatchDirection and OddHatchDirection to 0 and 90 I got much better results. Think that's because my position accuracy is not good enough, yet. I currently have about 0,25mm but will upgrade it to 0,125mm at least.

I also found the parameter for the speed of the outer shell. It's OutlineSpeed. Setting this lower (I use 0.66) does make the shape looks more consistent.

After I switched from 2 to 1 foundation layers, I got an acceptable shape of the first shell. Apparently, something is messing with the foundation layer heights.

I tried some changes to the Extruder0_SeparationFraction parameter, but other values than 0.8 gave me a bad second foundation layer and/or worse results on the first shell.

I noted some additional oddness - I had to set the Z-distance twice as high as it actually moves. I double-double checked my firmware settings and they seem ok.

1,8° stepper with half steps = 400 steps
M8x1,25 threaded rod drive = 320 steps / mm

Anyone noted this too?

---

-------------
FiveD Firmware fork (supports Arduino Mega): [github.com]
reprapsource.com

Those numbers look correct.

You might want to check to make sure it is the z axis moving the wrong amount and not that the layers are too thick. This is easily accomplished by homing and then moving 50 mm and the measuring the distance from the extruder head.

I had a problem for a while with my layers being twice as high as I expected. This occurred because there was backlash in my y axis. When the head moved back and forth for the infill, the backlash caused every other segment to line up exactly with the previous segment. I had to watch very closely to verify that this is what was happening. Once I tightened up the y motor coupler things worked fine.

---

---

Darwin clone, Gen 2 electronics, Arduino Duemilanove w/ AtMega328, 5D Firmware, Pinchwheel extruder
[www.codeerrors.com]

I "fixed" the problem moving to skeinforge. The settings and hardware was all fine.

Apparently the java host was configured to run too slow and make the lines too big. After switching to skeinforge, the build quality switched literally from a Fiesta to a Ferrari.

On skeinforge, only the Nov-09 version works for me, the later versions produce G-codes with far too slow feedrates somehow.

---

-------------
FiveD Firmware fork (supports Arduino Mega): [github.com]
reprapsource.com