ATOM 3D Printer Group 造訪社團 » CES

[Post process]

最近有點忙，本業的事情開始多了～～～所以最近比較少在印東西。

另一方面也是我發現這樣 slice & try 的模式有點問題。

因為買了一台好機器，卻因為切圖程式的優化程度有限，讓印出來的東西品質變化極大。問題該算在使用者對 FDM 的概念不夠純熟嗎？其實好像也不盡然。

所以，我最近越來越不期待切圖軟體產生的 gcode 就一定會多完美，也越來越想自己寫程式來分析路線，然後修正整個列印的程序和節奏。

同時，有一些整合的東西也得開始有計劃的進行，讓整個工作能更加順利圓滿的進行，這樣實在很令人感到興奮與期待。

不過我不會也懶得再學 python 了，所以灌了 php 來用。

以後 gcode 生成後都要自動 review & optimize 一下了。

Very nice accessory

http://www.thingiverse.com/thing:951909

I printed a set of these in a little over 2 hours and they work perfectly. The handles make the hex tools much easier to hold and control.

翻譯年糕

Big and Little

Photo 1 shows the largest part (Big) I have printed so far. It is a little more than 200 mm high and used about half a reel of PLA filament. I designed the part and have tried to print it several times, but up to now I have only been able to print small versions of it becuase the slicers I've used would not slice the full size part. Yesterday I purchased the Simply3D slicer and it allowed me to print the full size part successfully.

Photo 2 is a smaller version (Little) I printed a few days ago. I used Kisslicer to prepare the GCode for printing.

Photo 3 shows Big on the printer at about 92% complete after 19 hours of printing.

Photo 4 shows the part's 2-layer design; it is hard to see the infill that is about 15%.

Photo 5 is Big & Little together. After looking at the tool path for Big I decided to change the design a bit; Big has fewer ribs than Little to make the hills and valleys more evenly spaced, and there a a bit more twist at the top which gives the part a more consistent look.

Based on this result I'd say the combination of the Atom2 printer and the Simplify3D slicer makes one of the best 3D printing solutions available today.

翻譯年糕

0.3mm layer height: OK but not great

Just as an experiment I set layer height to 0.3mm for this part. Doing that reduces overall print time, which is nice, but the resulting part does not have nearly as nice a finish as one sliced with 0.2mm layer height.

翻譯年糕

Hello!
my Atom 2.0 is not printing in level, the back part (closer tot the Z axes) is lover (about 0.3 mm) than in the front, is there any solution for this? I've tried to put some tape under the glass but it didn't help after auto level it is printing the same way. I've checked the aluminium extrusions and the corner pieces there are almost no gaps (maybe under 0.1 but I can't push tighter). Thanks

翻譯年糕

Kon-nichiwa! Thank you for approval .

My name is Dr. Ishikita a designer of VapoJect 3D printable anesthesia device.

Unlike the operating room anesthesia devices weighing over 700 pounds. VapoJect device put the core function on this machine in palm of your hand!! (US PATENTED)

I believe VapoJect will help Mars mission, because astronauts doesn't have anesthesia device yet.

Please spread and support our aerospace medical project to help millions of lives!!

翻譯年糕

Translucent PLA filament

Here are 2 goblet-shaped parts I made using translucent red filament. I bought a reel of this to test it and see what happens when you make items and try lighting them.

The first 2 photos show the 2 parts with normal lighting. The second 2 are with strong backlight. The patterns on the last photo are due to triangle fill material printed between the part's inside and outside surfaces. There are no such patterns in the 3rd photo because I made the inside and outside surfaces of that part closer together (thinner walls) and there is no fill material.

Even thought the filament was described as red it appears orange in normal light. I think lighter colors - like yellow - would have more transparency and not require such strong lighting to show off their translucent characterisitcs.

翻譯年糕

An interesting problem: Here are 2 pics from an aborted attempt to print the Eiffel Tower model. I scaled the model up to 270mm height and used a layer height of 0.200 mm. As you can see I used no support structure which may have been my fatal mistake.

The first photo shows the results after about 5 1/2 hours of printing. The second shows a close-up of the results about an hour later. I had to stop the print because the Z-Min adjusting screw got caught in some un-evenness and knocked off a couple of pieces of the tower's support beams. When this happened 2 of the magnetic arms came off their knobs and I of course had to stop the print. Fortunately I was close by when this happened; I had planned to let the print continue all night. I'm not sure would have happened if I had not been there to stop the printer.

I don't know if I am the only one who has encountered this problem, but it could be easily fixed by simply using a shorter screw for the adjuster. It wouldn't have to be much shorter - just a few mm would have prevented this problem from happening.

翻譯年糕

New design - nice results

I changed the design of my ribbed vase so it has only one outside surface and one bottom surface. I had only a small amount of brown filament left so I printed a small test version of the part. I set Kisslicer's Skin Thickness to 4 mm, but it looks like the sides of the part are about 2 mm. This may be because I scaled it down to a small size due to lack of filament. (The last pic shows how much I had left when the print ended.)

I'm not sure why the outside surface looks so much better than the inside - see pics # 2 & 3. This may be a quirk of Kisslicer. But it does seem that a single-surface part can print OK; this means I do not have to be concerned about both outside and inside surfaces. Thanks to Clarence Lee for pointing that out.

翻譯年糕

Good news and Bad news: the good news is I was able to print (almost) the ribbed vase I was having trouble slicing. To get it to slice I doubled the bottom's thickness from 3.1 mm to 6.2 mm. And when I printed the part I scaled it up from 110 mm high to 120 mm high.

With these changes Kisslicer seemed to slice it OK. However, Craftware still would not slice it properly. So I used Kisslicer's GCode to start printing. The estimated total print time was a little under 14 hours.

The print looked absolutely superb up to the time I went to sleep for the night. You can see how smooth the inside and outside surfaces are in the first image below. Without a doubt this is the highest quality I have printed so far. Setting Layer Height = 0.2 mm might be the key to that. I was hoping for a really great finished part when I got up the nest morning.

Unfortunately this was not the case. When I got up and turned on the light (it was just before 5AM) the printer was at about the 90% mark and doing just fine. I turned on my PC to get ready to post this message, and very shortly after that the printer stopped printing. I looked at the LCD and it had gone back to the main Info screen where it showed 0 for the heated temperature and the actual temperature dropping fast. So I had no choice but to stop the print and do an Auto-Home.

The printer is powered from the same power control center as my PC. This unit has 4 independently switchable power outlets and is itself connected to a UPS box. My guess is that there was a slight voltage drop caused by turning on my PC, and this somehow affected the Atom2's circuits.

After I did Auto-Home I checked the motherboard connection of the 2 yellow thermistor wires and that was fine. Then I tried Preheat to 200 and that worked OK. So I do think the printer is physically OK.

The second photo below shows the other problem I had: The bottom came off the part when I tried removing it from the print bed. In fact, the bottom did not print as designed - most likely as a result of what Andrew Lee discovered when he looked at the part's STL file. The bottom was supposed to be 6.2 mm thick, but in fact it printed a bottom only 0.6 mm thick (0.6 mm = 3 times 0.2mm layer thickness).

Even though this print failed I am very pleasantly surprised at the quality the Atom2 is capable of printing. I am going to change the way the bottom surfaces are created and try this print again - probably in a smaller version first. Hopefully that will yield an even more impressive result.

翻譯年糕