ATOM 3D Printer Group 造訪社團 » Kisslicer

Birk Binnard this makes me curious. Try send me the file. I would love to look at in my 3d setup. I think you have more than one surface in the buttom.

Maybe find out the best temperature for this filament could help the problem.
http://www.thingiverse.com/thing:282520/#instructions
BTW, layer 0.15, 0.1 event 0.05mm worth a try for the beautiful detail. ;p

Andre - I'm sorry but I have deleted the STL file and have only the GCode left - so we will never know what actually happened. However, I have just finished re-defining how the part is made and I now have the inside surface joined with only the insid …… 查看更多

Andre - I forgot that I had saved the STL file from the part you looked at on a cloud drive. I also put up there my "fixed" version of the part Here is the URL for both. The fixed one is called thermos.stl:

http://1drv.ms/1GDCLSZ …… 查看更多

Finished slice this thermos.stl on Kisslicer 1.5 Beta 2.2
Peak memory usage is about 3.5GB
bottom is about 6.2mm …… 查看更多

http://www.thingiverse.com/thing:851525
remixed 245~190 version

Hi Clarence. I have the same version of Kisslicer as you do, and when I sliced thermos.stl I get strange results. My key parameters are 0.2mm layer thickness, Num loops = 3, Extrusion width = 0.38, Skin Thickness = 2 mm, and Infill = Vase.

Here are th …… 查看更多

Why infill vase or hollow? I think slicer did them right. But for this stl, it should be infill at least 10? Check the start of my gcode, you can found the slicer setting there for reference.

http://www.thingiverse.com/thing:18672
Check this stl, I think this kind of the model design would need to use infill=vase .
Suggest to design like that, a solid cylinder. …… 查看更多

Clarence = thanks for your comments. Because I started designing models before I ever got a printer I had to guess how to do it. At the time it seemed to me that a vase or bowl type part should have an inside and an outside so it would have sufficient …… 查看更多

Hi Birk, I took a quick look and it looks like there are two problems:

1. The upper flat face at the bottom of the face (not the one touching the print bed) goes through the sidewalls of the vase. Instead it should be trimmed by the inner perimeter of the vase wall

2. The very bottom face (and maybe also the same upper face as mentioned before) seem to have duplicate surfaces on the same plane. This will also confuse the slicers or 3D model programs.

1. The tiny edge peeking out of the outer side wall indicates that there is a surface with a perimeter running along it

2. The odd pattern indicates that there are 2 surfaces on the same plane

Hi Andrew - thanks for your help on this. I will make a new posting about what I did with this part to get it to print (sort of), but it's clear the key word in your comments above is "confuse".

Your observation that both of the surfaces of the bottom meet the outside surface of the part is correct. I could have trimmed the bottom's upper surface with the inside part surface, but I used the outside surface simply because that's the one I used to trim the bottom's bottom surface. I can easily use the part's inside as the trim surface; I will do this on my next print.

I did not think it mattered whether I used the inside or outside to trim the bottom's top surface because I thought slicers only cared about external surfaces, and ignored internal ones. I guess this is not the case.

I am not sure what to do about apparent duplicate surfaces. In the part design there are really only 1 of each. Each one is created as a boundary surface, where the boundary is the plane curve formed by the intersection of the outside surface with the top and bottom of a right circular cylinder that is somewhat larger in diameter than the actual part.

For my next print I will use a different method to create these surfaces. I'm not exactly sure what that method will be, but hopefully it will also resolve the problem you describe in #1 above. That one is surprising because I did a "join surface" to eliminate the naked edge at the intersection of the outside and bottom surfaces.

Thanks again for your help with this. It's clear I still have some things to learn about designing parts for 3D printing.

補上設定參數#1

補上設定參數#2

真是屋漏天逢連夜雨，前幾天是板子的5V供電故障，改以USB外接電源繼續工作，不知道是不是我幾乎24小時使用的關係，昨夜裡後吹散熱風扇罷工堵料了.....Orz 有人知道哪裡可以買得到適合的規格嗎?有沒有辦法利用另一個溫度量測散熱鰭片做第二層把關，過熱就讓機器暫停並降溫。

噴頭的散熱風扇不是12V嗎，還是2·0版的新機板有改過，散熱風扇網拍有很多找一下應該不難吧

是12V沒錯，也利用空檔上網也買到了一樣的風扇。不過在整理噴頭的時候，熱敏電阻居然自己斷了.....現在已是室內淹水狀態，停擺。atom使用的熱敏電阻B值常數是3950嗎？

http://goods.ruten.com.tw/item/show?21503337537188

昨晚去光華尋找，以為這東西應該不難買，沒想到問了三四家大間的電子材料行都沒有，只好網路下訂。

出料不順

+1可能要檢查喉管是不是有碳化物堆積造成不順喔！

好的感謝，我會找找問題!

Could you compare the first few lines of the gcode? Bypass the comment .
Looks like a little too high on first layer.
Did you use bed roughness on kisslicer?

Yes, I did compare the initialization lines from both slicers. The Craftware slicer includes an M83 command that blows the Atom2 right out of thw water! I had to reload Atom2's firmware to recover from this. So I commented out the M83 command and the GCode ran fine.

Just a power cycle reset should be ok to recover from it.

Continued......

Comparing the 2 sets of GCode shows they are pretty similar, but many of the statements are in different order. Since it is all initialization I don't this the order affects printing much, if at all.

No, I have never used bed roughness - is there any reason to, considering it is a piece of plate glass?

This is the start of the Craftware GCode:

; GCode generated by CraftWare
G28 ; Auto-Home
G29 Z+0.5 ;Auto-Level
G1 F6000 Z50 ;lift nozzle
G0 X0 Y-100 Z50 ;parking+
G0 Z20 ;parking
M201 X2000.00 Y2000.00 Z2000.00 E9000.00 ;max acceleration

M109 S200 ;set and wait head temperature
; commented out M83
GF500E5 ; startup extrude modified from GF100E30
; ------------------------------------
; Layer #0
; ------------------------------------
G21 ; mm
G90 ; abs
G92 E0 ; reset extr
M106 S0
;segType:Skirt
G0 F1800 E-9.000
G0 F3000 Z0.300
G0 F3000 X-26.191 Y26.613

And this is the start of its code for Layer 1:

; Layer #1
; ------------------------------------
G21 ; mm
G90 ; abs
G92 E0 ; reset extr
M106 S255
;segType:Perimeter
G0 F1800 E-9.000
G0 F3000 Z0.600
G0 F3000 X30.167 Y-3.193
G0 F1800 E0.500
G1 F1200 X30.671 Y-2.779 E0.5309
G1 X30.734 Y-1.895 E0.5729

So the M83 already removed from the Craftware->Slice->Gcode Script-> Header ?
Looks fine to me

I put the comment in there to take out the m83. The previous version of craft are did not have that command. I don't know why this newest version includes it.

Birk Binnard
http://www.birkbinnard.com

To solve this I used the brut-force technique - replaced all the Craftware GCode up to where it starts actual printing with corresponding Kisslicer GCode.

I also discovered that if you use a slicer that supports bed heating, and you forget to tell the slicer that you have no heated bed, the slicer will put in commands to heat and cool down the heated bed. These commands cause Atom2 to do bad vibrations that require power-off and on. I also did firmware reload just to be sure.

@@ Strange, there should be no loading at all, but might be board design difference. But without HeatBed there but request the M190, the gcode might never start the print since no Heatbed there.

Playing with craftware yesterday.
I like the "Drawing order" option. Got more choice.
My setting : atom2.cwsp
https://onedrive.live.com/redir......

It's just a start point for reference, but may not for standard ATOM2.0.
Since I have some customize tuning on my ATOM 2.0 Beta.

BTW, the 3D animation tips is cool on Craftware.
http://youtu.be/7a9eB2M1yaI

Thanks for posting your Craftware settings. I will definitely try them out. I really like Craftware's excellent visualization capabilities and hope to make it my default slicer once I come up with a good set of parameter settings. I think I'm pretty close to that now.

Clarence - a couple of comments& questions on your Craftware settings:

1. On Advanced tab I switched Source Direction to Random Position. Doing this avoids a visible exterior seam when printing something with concentric layers.

2. I've been slowly increasing my printing speed - I'm up to 45 mm/sec now. Looks like 50 mm/sec works well for you - I'll try that.

3. I had my Retract & Prime length set to 3 and that resulted in some very thin strings, so I was going to change it to 5. But I''ll go to 6 based on your settings.

4. Are you printing PLA at 225 C? That seems hot to me. So far I've done everything at 200 C, but I only have used plain PLA so far.

5. With the laters version of Craftware I've had to replace the startup and shutdown GCode with comparable lines from Kisslicer. Otherwise my Atom2 goes crazy when starting the next print and requires reloading firmware. Has hat been your experience too?

Thanks again for posting your settings.

1. random should be fine.

2. My stepper motors, controller board, stepper drivers all different from the ATOM 2.0 Release. So, may not apply the same.

3. My extruder PTFE tube is as short as 600mm, I moved the extruder stepper motor down for shorter PTFE tube. But yes, increase to 5.5/ 6 or even 6.5 still worth a try. Actually the E step speed / acceleration is important here.

4. I tested my PLA different temperature on a thin wall cylinder, to determine the best strength temperature. I would check it on every new opened PLA.

5.Strange, mine is 1.09 beta. no this issue.

https://onedrive.live.com/redir......
gcode for 4.
Please remember to uncomment the G29 & set right Z+0.x for your printer if needed.

My gcode from craftware for reference.
https://onedrive.live.com/redir......
https://onedrive.live.com/redir......

Thanks ; I'll try your gcode in the morning and post my results.

Well I've managed to (I think) get Craftbot's GCode to the point where I don't have to switch out the startup & shutdown sections, but there is something wrong with the extruder code somewhere. What happens is the extruder motor turns far faster than it should; when the print starts the extruder wheel start turning about 2 - 3 RPM which is way too fast.

Needless to say the hot end can't accept filament that fast so the extruder wheel generates lots of little filament pieces trying to force-feed the hot end.

I've tried to figure out what controls the extruder motor but haven't quite got there yet.

It looks like all that's needed for my standard Atom2 is to comment out 2 lines of GCode that Craftware generates: the M83 command and the GF100E30 command. I am printing an interesting part now with those settings and, assuming it finishes OK, I will start another thread with specifics about the settings and a copy of my Craftware parameters.

What is your retract speed? Btw, try to slice by newest cura. It's optimized to less retract.

I think this shows retract speed = 120 mm/sec

I haven't looked at Cura in a while, but I'll check it out again. Thanks.

https://www.facebook.com/groups/atom3dp/1605464206358535/
did you use this ?

Not yet but I will print it out first thing in the morning

Birk Binnard I had a somewhat similar problem when I unchecked d-string and wipe under Style. It behaved like it picked out a new starting point for each new layer. Very dtrange. The strings mucy thinner though. Strange!

It's normal if you don't enable destring ;p
BTW the starting point is control by Seam hiding options.
check the "jitter"

Thanks Andre - somehow my destring & wipe boxes got unchecked. So I was able to print 2 copies of the quick_connector_spacer with no problems. I'm running another of my goblet shapes now and so far so good.

I did run the connector STL file through the Microsoft/Netfabb Repair Service and it got a lot smaller. I'll upload it in a separate posting because this only accepts photos.

Hi Birk Binnard , we are compiling a more comprehensive ATOM manual, and I would like to use these photos of the strings as an example. If you don't mind, would you be able to send us the full resolution files of these two photos? Thanks! (email is utypedesign@gmail.com)

Unfortunately I no longer have the full-size images. I thought the Facebook software saved the actual images as uploaded so I deleted them from my phone. They never made it into Google's backup either - I looked.

But you can easily reproduce this by simply printing a hollow round object with all the de-string settings set to 0.

That's alright, thanks anyway!

Hey Andrew - I just found the photos you asked for - they got saved in my deleted email folder. So I am emailing them to you now.

Thanks! Very much appreciated

ks的時間很不準

KISSlicer的時間很準，是Marlin配RAMPS1.4系統配Delta型態的印表機，速度不準。
印外圍很慢，印中心很快...

所以滑軌是等速度的，造成噴嘴中間速度快，外圍速度慢?

那個可有可無，作用是讓小物件散熱用