ATOM 3D Printer Group 造訪社團

實在太有才了◑﹏◐

可是改TMC2100, 不開大一點，回抽速度超過80就會失步了…

你的 E 軸加速度調多少? 我用 3000 回抽都開 120mm/s @@

電壓調大, 電流大, 影響的是加速度(A)

F = MA

跟回抽速度大小應該沒關係

除非小於Marlin 的 Jerk 速度, 才會沒有加速段. 用全力加速.
這時電壓多大, 馬達力量就多大.

求設定值…E加速度目前是4500, 回抽6000，電壓調至極限，計算值是0.97A

E 我用官方提供的 3000
電壓, TMC2100 大概 0.75V

下班再試試，最近下班一直泡在工作室被碎念了…

用這樣印的0.1第一層很明顯 但同時0.2就又ok Orz......

有別捲線材可以測嗎?

出料有泡泡就是因為線材濕度太高! 先前列印的經驗!

感謝樓上兩位，我正在試別卷，但是我自動退料目前也無法自己用擠出馬達退出，這樣的狀況有可能是噴嘴那邊有卡嗎？

剛已經這樣好幾次..... :(

剪掉膨脹的頭，放入用手推ok，然後再用move axis也還ok，但就是看起來有氣泡（我原來的問題），然後比如說測試0.2mm印ok，側一些其他的比如說0.1mm印或是自動退料就又會變成料卡在裡面的狀況...

之前就是因為、料濕度高 然後產生氣泡 接著就噴頭卡料 清噴頭 不斷地獄的循環 建議料件儲存要放除濕機。

感謝！一般清噴頭試如何清阿？我只有用0.4的鑽頭通過，不過目前看起來好像需要更麻煩的清理，請指教了

加熱..推看看,,也可以用小隻的六角板手幫忙推...
最慘的情形就是要分解噴頭了...

重點是去換不潮濕的料試試看、光清噴頭沒有用、清不好報消就gg了...

換色真的比較好！ 但這還是手推，現在來試試看機器

不建議用鑽頭, 會傷到噴頭

那可以用鑽頭的尾端嗎（沒有刀刃的部份）？ 目前換色好像就沒有泡泡了 測試中

建議拆噴頭, 從送料端去排除,
加熱後, 手動用料擠, 或用小隻的六腳板手推.

0.1 變好很多但路徑尾端還是不理想

0.2mm 還是很好 是因為0.1mm extrusion width 要調低嗎? 我一直用0.4

拆噴頭我很怕會把噴頭弄壞 有人有拆過的照片嗎? 我沒有「要拆到怎樣的程度-》正確的修法 」的概念 抱歉 謝謝各位回答新手問題

回覆一下也許有人之後也碰到,應該真的是水氣+我上一個東西被撞倒後我有轉到噴頭微動------->>所以再把灰色料放到防潮箱一整天之後再拿出來用+微調zoffset之後,又開始順利列印中:) 感謝各位的救援!

這...

3D渾天地動儀!

這... 應該是機構沒鎖好 @@

我上次列印中, 去搬移機器都沒事 XD

最近印了好多東西都沒遇過出料不順，就剛好地震時那一層出現擠料不足的狀況，地震搖完過一會就又正常了，沒有xyz軸失步的現象。新北市晃很大，不知道是不是地震持續搖晃輕微的拉扯線料造成，也或許今年他剛出生第一次遇見地震嚇到發抖了.....

sorry, 小弟的手機爛，若有需要，容後補圖。

你可以試放鬆一下皮帶，很難解釋緊度聲音

另外A4988的Vref有針對升級後的MineBea馬達調整過?
檔案區那個升級文件看一下有參考建議值

說到張力..
http://www.thingiverse.com/thing:33236
昨天還是請出這支來同步我三條皮帶的張力了

回家印一個來量量~

先測量一下您晶片的電壓值吧~ 藍色的晶片大概抓0.45~0.5v之間
紅色的晶片大概抓0.8~1之間跑跑看

謝謝各位的建議。我晚上會回去量量電壓。

量測結果。藍色板(ATOM 1.0)。照寶典量法。左邊是0.16，右邊是0.96。四個馬達都差不多。請問這值OK嗎? 怎麼跟各位說的差這麼多? 寶典寫: 左1.19, 右0.3

回報: 我把XYZ右邊調成0.45V之後，原來的大失步問題就消失了。試印了"300 height"和"50x50"算是成功。有趣的是，本來很奇怪的馬達聲，好像都消失了!! 請問左邊的值要調多少呢? 1.19V嗎? 另外，我的擠料不是很順(未升級前不會，升級也沒升馬達?!)。是不是擠料馬達的控制電路Vref也要調呢? 左右參考值又各是多少?

扭力不夠就往上調吧... 左邊基本上如果現在聲音愉悅也不用動... ;p

Clarence Lee 現在的聲音確實是不愉悅。請問有建議值嗎? 1.19V? 左邊這值的功用是?

http://reprap.org/wiki/G3D_driver......

感謝指引

回報: 目前看來，經過Vref調整成0.45V後，大失步的問題確認消失。基本列印已經正常。ROSC的值，我按照@ Clarence Lee 上面給的Link，把它直接逆時針轉到底，看不出和原來有什麼差異(原來的設定似乎已經是底了)。現在皮帶傳動方面就剩一個小問題了: 在噴頭平台在某個方向移動時，會有大的共振聲。聲音偏低沈、振動大。不太確定點在哪裡，像是某個軸柱上的東西的共振。不知大家的ATOM會不會和我的一樣吵? 如何解? 半夜工作，被家人complain.....

ROSC並不是只有逆時針轉到底啊..
還是有後面的微調可以看看...

不確定共振點在哪就有點難抓了..想辦法辨別看看

Po-Hsun Peter Wu 紅色的晶片是指dvr8288，藍色晶片是指v4900嗎?!

應該是在講1.x -> 2.0的使用者.. 當時都是A4988 只是有分藍色的是G3D
只有2.0出貨的才是使用DRV8825

Clarence Lee 講的沒錯！ 謝謝啦

Fan/cool
這欄就是....
目前設定的意思是0.15層高開始全速吹

是阿~可是它就是不動>_<

LCD 就是從 Temperature 那邊進入, 把 Fan 調整成 255 就是全速.

Perhaps there is a problem with GCode processing.

I tried printing 3 more test parts. When I tried the 300height part again the printer did all the preliminary moves ok, but tried printing with the extruder about 3 inches up from the build plate. I tried printi ng the 50x50 test and this time, after completing the Auto-Level stpes ok (from the part GCode, not LCD panel) the printer put the hot end right down on (touching) the build plate and waited for heating to finish. When it started to print the hot end dragged acriss the build plate and knocked off 2 magnetic arms. It also left a 1" scratch on the build plate.

After restarting I did Auto-Home and Auto-Level again from the LCD and these worked OK. Then I tried printing a new test part I created (a small cone) which I sliced with the Atom2 version of Kisslicer. This video shows the results:

https://drive.google.com/....../0B3EpEEIPcV....../view......

Under GCode control the printer goes through the Auto-Level steps and then waits to reach temperature. When it does it startsmaking strange moves about 1" above the build plate. I switched it off becuase I wasn't sure what it was going to do. All this makes me think there may be a problem interpreting/executing GCode, but again that is just a guess.

I think the official test file should be no problem.
Could you paste the gcode you slice for checking what it's doing?

For the effector, check the fan cooling your coldend. Sometimes stuck is bad cooling , so it happened after a while printing.
A photo/video focus on the effector ,fan might help to find out the issue.

The stepper Vref setting now might not good for running, sounds like ripple on slow movement. What is your 1st layer speed ?

I compared the beginnings of the GCode from the 50x50 part with the GCode I created using Kisslicer on my small pyramid part. The 2 are identical except for the warm-up temperature (50X50 was 190, cone was 200.) Here is the cone GCode that comes after the leading comments:

; *** G-code Prefix ***
;
G21 ; [mm] mode
G90 ; absolute mode
G28 ; Home
G29 Z+0.5 ;auto level
G1 F6000 Z50 ;lift nozzle
G0 X0 Y-100 Z50 ;parking+
G0 Z20 ;parking

;
; *** Main G-code ***
;
; Reset extruder pos
G92 E0
; BEGIN_LAYER_OBJECT z=0.20
;
; *** Warming Extruder 1 to 200 C ***
; Select new extruder
T0
; Warm it up, and wait till it's warm
M109 S200

;
G92 E0
;
;
; 'Prime Pillar Path', 0.2 [feed mm/s], 5.0 [head mm/s]
G1 X7.5 Y18 Z0.3 E0 F9000
G1 X7.5 Y18 Z0.2 E0 F6000
G1 E6 F7200
G1 X7.41 Y19.27 E6.0475 F300
G1 X7.12 Y20.51 E6.0951
G1 X6.66 Y21.69 E6.1426
G1 X6.02 Y22.79 E6.1901
G1 X5.23 Y23.78 E6.2377
G1 X4.3 Y24.65 E6.2852
G1 X3.25 Y25.36 E6.3328
G1 X2.11 Y25.91 E6.3803
G1 X0.89 Y26.29 E6.4278
G1 X-0.36 Y26.48 E6.4754
G

I'm not at all familiar with GCode but it looks like the code I generated is just like that from the 50x50 part so I don't understand why the printerb ehaved so differently when I tried to print the cone.

I looked at Kisslicer and it says FIrst Layer Speed is 200.

Perhaps it really is necessary to do both Auto-Home and Auto-Level from the LCD before running a GCode part. And what about the options to Restore Failsafe & Load Memory? I've seen some comments about these but it's not clear what they do.

The main fan in the end effector is running OK. I have checled this several times. And considering the fact that the printer performs Auto-Home, Auto-Level, Load Filament, and Unload Filament just fine from the LCD tells me that the steppers ad end-stops are all working OK. So I don't think there is a physical problem with the printer.

If this is true then it has to be something in the electronics or the firmware. But how to identify exactly what? I am very puzzled at the moment.

Actually we can input gcode on pronterface to check what's wrong.
G21 ; [mm] mode
G90 ; absolute mode
G28 ; Home
G29 Z+0.5 ;auto level
G1 F6000 Z50 ;lift nozzle
G0 X0 Y-100 Z50 ;parking+
G0 Z20 ;parking

If it's too fast on G1 F6000 Z50 ;lift nozzle
try to slow it down to F3000 before tuning out a better Vref

OK - I can enter those Pronterface commands now.

After a fw upgrade, if it's very difference, sometimes need to restore failesafe & store memory to apply the new setting.

1st layer 200 mm/s is not normal. it's too much.

All your GCodes ran OK
echo:SD card ok
>>> G21
SENDING:G21
>>> G90
SENDING:G90
>>> G28
SENDING:G28
>>> G29 Z+0.5
SENDING:G29 Z+0.5
-1.912 -1.912 -1.912 -2.006 -2.162 -2.162 -2.162
-1.819 -1.819 -1.887 -1.975 -2.119 -2.300 -2.300
-1.731 -1.825 -1.931 -2.031 -2.137 -2.281 -2.519
-1.725 -1.831 -1.962 -2.087 -2.200 -2.331 -2.487
-1.700 -1.844 -1.981 -2.106 -2.231 -2.394 -2.494
-1.825 -1.825 -2.006 -2.150 -2.312 -2.444 -2.444
-2.012 -2.012 -2.012 -2.219 -2.400 -2.400 -2.400
echo:Settings Stored
>>> G1 F6000 Z50
SENDING:G1 F6000 Z50
>>> G0 X0 Y-100 Z50
SENDING:G0 X0 Y-100 Z50
>>> G0Z20
SENDING:G0Z20

Above is copy from Pronterface. Printer made all moves A-OK.

I can change Kisslicer 1st Layer speed - but what is a good value? Could that explain the strange behavior of the printer?

I found the comment in the GCode that says First Layer Speed = 30, So I will put that into Kisslicer and recreate the cone GCode. Here's hoping that will work better.

30mm/s is fine, I think your profile didn't applied.
http://www.atom3dp.com/en/support/
get the profile here

Doing that now. Will again after updated firmware and next test complete.

like that, you can use the speed bar to increase the speed.
The upper limit might can adjust to 50 or 60.

Results not good: AFter firmware update ran OK I did Auto-HOme & Auto-Level from Atom2 LCD. Both ran OK. Then I started the 50X50 test print. Here is results:

Oops - hit Enter by mistake.

https://drive.google.com/....../0B3EpEEIPcV....../view......

Just a thought - after loading firmware from Arduino I did do Restore Failsafe and Store Memory, but there was no feedback from either of these operations. Should I have waited for the LCD to say something - because I did not do that.

No, there's no much response. Actually it's exactly the same as
M502 M500 on pronterface

I don't think the problem is on fw.
It's much related to the stepper vref.

Well that's an interesting conclusion for sure. I posted my current voltages above - so what values do you think I should try to set them for? The potentiometers are so sensitive it's almost impossible to set the voltages to any specific value, but I think I can come fairly close.

Note that the voltages I measure are between the potentiometer itself and the bottom-right pin as viewed from the front of the printer.

I would left it as Bato Wang previous suggest range, or try to search another value around.
I don't have the same model of the stepper motor, so I don't have much experience about it.
But it's not normal when it moving like that on the video.

What is current M503 result?

I agree that movement is bad. But the fact that all motors move OK when under led control suggests that they run OK and thus their voltage should be OK. So I still think the problem is software related.

Not really, it's on the different speed, acceleration request on gcode.
The loading and reaction on the stepper driver is different.
When the movement not normal, the stepper driver Vref setting & belt tension are usually the cause.

I sent m503 to printer from Pronterface after power-up and Auto-Home from LCD. Here is the result:
>>> m503
SENDING:M503
echo:Steps per unit:
echo: M92 X160.00 Y160.00 Z160.00 E200.00
echo:Bed Level Matrix :
-1.988 -1.988 -1.988 -2.075 -2.231 -2.231 -2.231
-1.900 -1.900 -1.963 -2.050 -2.194 -2.369 -2.369
-1.813 -1.906 -2.006 -2.100 -2.219 -2.350 -2.581
-1.813 -1.913 -2.044 -2.156 -2.269 -2.400 -2.556
-1.788 -1.931 -2.056 -2.175 -2.306 -2.463 -2.581
-1.906 -1.906 -2.081 -2.225 -2.375 -2.506 -2.506
-2.081 -2.081 -2.081 -2.288 -2.463 -2.463 -2.463
echo:Maximum feedrates (mm/s):
echo: M203 X300.00 Y300.00 Z300.00 E300.00
echo:Maximum Acceleration (mm/s2):
echo: M201 X9000 Y9000 Z9000 E10000
echo:Acceleration: S=acceleration, T=retract acceleration
echo: M204 S9000.00 T9000.00
echo:Advanced variables: S=Min feedrate (mm/s), T=Min travel feedrate (mm/s), B=minimum segment time (ms), X=maximum XY jerk (mm/s), Z=maximum Z jerk (mm/s), E=maximum E jerk (mm/s)
echo: M205 S0.00 T0.00 B20000 X20.00 Z20.00 E5.00
echo:Home offset (mm):
echo: M206 X0.00 Y0.00 Z0.00
echo:PID settings:
echo: M301 P16.54 I1.30 D52.56

Where did you get this fw? it's max acceleration is too high.
M201 X9000 Y9000 Z9000 E10000

I downloaded formware from Atom2 Support site: http://www.atom3dp.com/en/support/

maybe your restore failsafe not really applied & save
under pronterface:
M502
M500
M503
after M503 check the result on M201 line

Here are results from above:

echo:Hardcoded Default Settings Loaded
>>> m500
SENDING:M500
echo:Settings Stored
>>> m503
SENDING:M503
echo:Steps per unit:
echo: M92 X160.00 Y160.00 Z160.00 E200.00
echo:Bed Level Matrix :
0.000 0.000 0.000 0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000
echo:Maximum feedrates (mm/s):
echo: M203 X300.00 Y300.00 Z300.00 E300.00
echo:Maximum Acceleration (mm/s2):
echo: M201 X3000 Y3000 Z3000 E9000
echo:Acceleration: S=acceleration, T=retract acceleration
echo: M204 S9000.00 T9000.00
echo:Advanced variables: S=Min feedrate (mm/s), T=Min travel feedrate (mm/s), B=minimum segment time (ms), X=maximum XY jerk (mm/s), Z=maximum Z jerk (mm/s), E=maximum E jerk (mm/s)
echo: M205 S0.00 T0.00 B20000 X20.00 Z20.00 E5.00
echo:Home offset (mm):
echo: M206 X0.00 Y0.00 Z0.00
echo:PID settings:
echo: M301 P16.54 I1.30 D52.56

looks fine now

Really? I don't know what to say. I'll try the 50x50 print again. Will post results in a few minutes.

Not good. After power off/on and Auto-Home the 50x50 part completed the Auto-Level moves OK and parked OK waiting for heat-up, but as soon as 190 was reached it did the same thing as my last video - noisy vibration moving hot-end around in jerky moves slightly above the build plate.

Not sure what to do besides increase Vref to 0.4V -> it's 0.8A current limit.

notice the two yellow box , it's the Gnd & Vref touch point .

OK. I'll get the 4 steppers as close to 0.4V as possible without going over. But - what if this does not help? Actually I don't think it will help because before I had voltages like this: X=0.40, Y=0.41, Z=0.42, E=0.38.

Yes, I am using those 2 yellow box contacts to measure voltage.

Starting to check/change voltages now. But like I said - I do not have a good feeling about this.

OK - I got voltages to be this: X=0.390, Y=0.389, Z=0.399, E=0.393.

That made things worse. When trying to print the 50X50 test part the machine moved like before, but the vibration was worse. Plus the hot-end never got close to the build plate.

So as you said, it is clear that the voltages have a large effect on hot-end motion. But what to do now? My sense is to lower the voltages to around 0.35 because when I did this before the motors sounded better than they do now when they moved the carts.

I am beginning to wonder if this machine will ever work properly. Is there something unique about the machine or motherboard I have? I can't imagine an average person going through all this to get a properly functioning machine.

I set voltages to X=0.326, Y=o.331, Z=0.328, E=0.319 and tried 50X50 again. Same results as before - lots of vibration and hot-end never got close to build plate.

Can you take pictures for the following parts

The stepper motor spec is 1.4A, since it's 6 leads.
I think under 1.1A should be ok for running.
So the upper bound might be Vref=0.55.

Make sure the middle terminal of micro switch is cut off.

Here are requested photos:

Yes, I cut off center pin of all end-stops. Pronterface says all end-stops are working OK, and they appear to do so because machine does Auto-Home and Auto-Level OK.

I was talking to my son who designs chip-making machines and he suggested 2 things:

1. Try a simple GCode program that makes only a couple of simple moves - like GOTO X+10; GOTO Y+10 just to see if simple commands read from SD card will work.

2. Try making simple moves from GCode with belts removed from stepper motors to see if motors turn properly.

He also noted that stepper motors can miss/drop steps if their acceleration is too high or if the are forced to move a load that exceeds their torque capability. Since the sliders move quite easily and the end-effector is light I don't think the motors have a problem moving it. But perhaps there is a problem with the acceleration code in the GCode scripts.

I have another idea - what if I take the 4 stepper voltage control modules off the old motherboard that I have and put them on the motherboard that is in the printer.now? Those modules should have nothing to do with the end-stop short circuit in the old board, and maybe they will help resolve my current problem.

How about send the following command to the terminal and see if they work or not : G28; G0 Z10; G0 Z300; G28 (in this order)....That should give you a simple result: home, lower the extruder to about 10mm height from the bed and 300mm height from the bed and go home.

Birk Binnard Sure! Are those stepper drivers works better on previous controller board?

Well I can't say they worked "better" because the short in the motherboard prevented the printer from doing anything real, but they did allow actions like Auto-Home & Auto-Level and individual moves from Pronterface. So I'll swap them now and see what happens.

First I'll try the Pronterface commands Bato suggested. Then I'll swap the stepper controllers and do the same commands.

Unfortunately swapping the stepper controllers did not help.

The Pronterface commands Bato listed worked OK with both sets of stepper controllers . But after I swapped the controllers for the ones on the old motherboard and tried printing the 50X50 t est part I got the same results - bad vibration and incorrect end-effector position moves.

This tells me it is not the stepper motors or their controllers that is the problem - it must be something else. Since manually entered moves - entered either from the LCD or Pronterface - work OK my sense is there is something wrong with the interpretation/execution of GCode read from the SD card.

To test this I can put Bato's commands into a text file, copy that to the SD card, and try running them from there - but what else is the minimum amount of other commands I need to put in the text file?

Or do you already have a simple GCode file you can email to me or post here?

G28
G0 Z10
G0 Z300
G28

save it's to a file with extension .gcode

The other possible test is you can direct print gcode via pronterface "load file" button

Manual file ran OK both from Pronterface & SD card.

I made this manual.gcode text file:
G28
G0 Z100
G0 Z300
G28

The file ran perfectly from Pronterface; this is the Pronterface result listing:

Loading file: X:\work\manual.gcode
Loaded X:\work\manual.gcode, 4 lines
0.00mm of filament used in this print
The print goes:
- from 0.00 mm to 0.00 mm in X and is 0.00 mm wide
- from 0.00 mm to 0.00 mm in Y and is 0.00 mm deep
- from 0.00 mm to 0.00 mm in Z and is 0.00 mm high
Estimated duration: 0 layers, 0:00:00
Print started at: 09:02:16
Exception in thread Thread-4:
Traceback (most recent call last):
File "threading.pyc", line 808, in __bootstrap_inner
File "threading.pyc", line 761, in run
File "printrun\pronterface.pyc", line 1006, in statuschecker
File "printrun\pronsole.pyc", line 860, in statuschecker
File "printrun\pronterface.pyc", line 976, in statuschecker_inner
File "printrun\pronsole.pyc", line 1291, in get_eta
File "printrun\utils.pyc", line 175, in __call__
File "printrun\gcoder.pyc", line 677, in idxs
IndexError: array index out of range

I'm not sure what all the above means, but the printer behaved exactly as expected.

I put the file onto the SD card and ran it from there. It ran perfectly also.

Now I am not at all clear on what the problem could be.

Tried some other GCode models from the SD card. All had bad results:

300height.gcode performed Auto-Home & Auto-Level OK, but after heating it put the print head down hard on the print bed and seemed to want to make it go lower. I turned printer off.

I tried the small cone model I made that I sliced with Kisslicer, but I saw that it was heating to only 30 degrees, not 190 or 200. So I changed the M109 S30 command to M109 S200 and restarted it. This time it did heat to 200 but behaved like the 300height test above.

Looking at the GCode files for the test parts I see that there are a number of statements like G1 X28.84 Y29.41 Z0.35 E0 F6000.

I believe the F6000 is the feedrate/speed of moving the end-effector, so what if I change all these to F3000 based on the idea that 6000 is too fast for the steppers to move, but 3000 might be OK.

I tried F3000 in place of F6000 but that did not help. I also tried disconnecting the extruder stepper from the motherboard, thinking that perhaps there was a problem with the extruder motor itself. But this did not help either.

I believe I have identified the source of my problem. I will start a new thread explaining what it is.

下面同步輪不是可以微調位置嗎?

同步輪因按照說明書留1mm,初裝時有1pcs蠻緊,若再深需用敲的.

照理講應該手撥順順的就進去了 是不是你卡到同步輪上面的螺絲了.... 不要用敲的....

謝謝Mark大,我螺絲拔掉時也是緊.不敢敲馬達.因已費力裝上. 我可能看看軸承部分的套筒互調能不能準.

我覺得這個可以替換一顆給你不能順順裝不正常

謝謝~已經裝上要全拆再找>"<

我是會滑車,上面墮輪培林,下面的齒輪
都盡量調整到讓皮帶跟線軌平行
對照的反而不是圖上紅色畫得那兩個點,
而是去看皮帶 ＆ 線軌為主

>Clarence
上層串在一起的鋁套筒，若移一個到另一端，那皮帶上下就對得準，但皮帶就算移位約5mm。會吃到滑軌空間嗎？

吃到滑軌空間是指?
https://lh6.googleusercontent.com/....../WP_20141214......
喔..我大概知道了..退太多過去是吧?

是~

那只好維持原樣, 我是因為用的是以前1.0的套筒.
又另外使用帶邊軸承.所以另外搭墊片湊的
稍微墊高 張力調整器也許也行.. XD

原來那叫做張力調整器,學到了~

是同規格

那個是標準規格

感謝您的回覆