Welcome Guest ( Log In | Register )

4 Pages < 1 2 3 4 >Bottom

Outline · [ Standard ] · Linear+

 Repraps and DIY 3D Printing!, Open source hardware~

views
     
Drian
post Mar 22 2016, 10:13 AM

Look at all my stars!!
*******
Senior Member
4,999 posts

Joined: Jan 2003


QUOTE(izzudinhafiz @ Mar 22 2016, 08:20 AM)
the wavy lines on the X/Y axis are called ringing. Caused by lack of rigidity or a slack belt or springiness from mass acceleration. Back when i was doing machining, we get ringing too when doing heavy cuts. Its mostly due to rigidity issues.
*
Interesting was reading about it more and it seems going slower and lowering acceleration improves it.
Was reading a few more post and it was mentioned that the print bed in normal cartesian printers is also a factor. It rings as it changes direction as the print bed has quite a lot of weight to it.

https://ultimaker.com/en/community/3532-pre...r-sharp-corners

I really do think coreXY movements improves the performance of the print. With no moving motors and no moving print bed, I do think it will resolve all these issues.
Drian
post Mar 22 2016, 02:00 PM

Look at all my stars!!
*******
Senior Member
4,999 posts

Joined: Jan 2003


QUOTE(altan @ Mar 22 2016, 11:38 AM)
I think even with the corexy design it will not eliminate so called ringing or print artifacts caused by fast movement. The long length of GT2 belts used will stretch and compress with acceleration of the heavy extruder motor so ringing will still exist. There might be an optimum acceleration value but going slow is undeniably the best solution but in reality not worth the value. Another solution is to go Bowden but that will introduce even more problems and limit some materials. Take the Ultimaker 2 as an example in solving some of the issues with corexy design.

In case you didn't know the corexy was around a long time and was found in a children drawing toy called etch a sketch which uses wires instead of belts. Historically we should called it etch a sketch but corexy sounds nicer. Also, idk if anyone made a firmware for the corexy yet but I understand each energized steppers will move the end effector (the part that holds the extruder, correct me if I am wrong) diagonally and needs both steppers to move along the cartesian x and y axis.

I have been looking into this design for a long time since I saw someone made a small 3D printer with that setup. He had the platform on the xy axis and the extruder on the z axis. Anyway you should build one and tell us about it. I am building another using particle board/MDF and an incomplete Prusa i3.
*
It will not totally eliminate of course nothing is zero mass , but it will significantly reduce it. In a normal cartesian printer you have the bed mass and extruder motor weight to worry about. In a Core XY you only have extruder motor to worry about, so that's half of a factor removed. The bed mass is not exactly low, like mine I have the aluminium plate + pcb heatbed + glass weight to worry about that reduces your maximum speed you can go. You can imagine if the whole bed/aluminium/ glass moves at 150mm/s and then stops suddenly(deaccelerate),(eg:- when printing a box and reaching a corner). And it doesn't help when the whole heated bed is held by springs and screws, it sort of helps oscillation. Theoretically if you cut your moving mass by half, you can increase your acceleration by 2X and max speed by 4X.

The GT2 belt will also stretch but the amount being stretched(force) is proportional to the mass and acceleration. The lower the mass the less force it will impose on the GT2 belt and you can set higher acceleration values.

It's interesting that you mention ultimaker 2, initially i was planning on benchmarking with my office rm60k stratasys printer but looks like ultimaker2 is also a good choice to benchmark based on speed.

Why are you building an army of 3d printers?

This post has been edited by Drian: Mar 22 2016, 02:05 PM
Drian
post Mar 28 2016, 11:00 AM

Look at all my stars!!
*******
Senior Member
4,999 posts

Joined: Jan 2003


QUOTE(Prosperer @ Mar 28 2016, 12:17 AM)
user posted image
user posted image

i hope this will do it biggrin.gif
*
What is your Z resolution without microstep?

It looks like your layer height is not in multiples of your Z resolution without microstep. Do you know your lead screw specs?


Don't rely on microsteps for your resolution, the stepper will likely skip because the torque in microstep mode is way lower than the normal mode. At 1/16 microstep the torque is only 10% of your motors rated torque.

http://www.micromo.com/microstepping-myths-and-realities

The consequence is that if the load torque plus the motor’s friction and detent torque is greater than the incremental torque of a microstep successive microsteps will have to be realized until the accumulated torque exceeds the load torque plus the motor’s friction and detent torque.

Simply stated, taking a microstep does not mean the motor will actually move! And if reversing direction is desired a whopping number of microsteps may be needed before movement occurs. That’s because the motor shaft torque must be decremented from whatever positive value it has to a negative value that will have sufficient torque to cause motion in the negative direction.


This post has been edited by Drian: Mar 28 2016, 11:04 AM
Drian
post Mar 28 2016, 11:37 AM

Look at all my stars!!
*******
Senior Member
4,999 posts

Joined: Jan 2003


QUOTE(Prosperer @ Mar 28 2016, 11:22 AM)
im printing in 0.2 layer height and according to prusa calculator i can print between 0.19375, 0.2 and 0.20625000000000002
i think my lead screw got 1mm pitch
*
Can you check how many mm it moves forward/backwards for every 1 rotation. Usually it is 2mm,4mm,8mm standard for leadscrew.
The way leadscrew is calculated is not only using pitch but also "number of starts"

http://www.protoparadigm.com/news-updates/...in-3d-printers/
QUOTE
The lead is the distance that will be traveled in the course of one revolution of the leadscrew. In the case of a single-start screw, the lead is the same as the pitch.

10 threads per inch is a 1/10 inch pitch, which means one rotation takes you 1/10 inches

For a multi-start screw, the lead is the pitch multiplied by the number of starts.
for eg:- this is 8mm
http://www.aliexpress.com/item/RepRap-3D-P...99-dfa7b0d9ae1a

this one is 4mm
http://www.aliexpress.com/item/3D-Printer-...1a-2c1a64604dea
Drian
post Mar 28 2016, 11:46 AM

Look at all my stars!!
*******
Senior Member
4,999 posts

Joined: Jan 2003


QUOTE(Prosperer @ Mar 28 2016, 11:22 AM)
im printing in 0.2 layer height and according to prusa calculator i can print between 0.19375, 0.2 and 0.20625000000000002
i think my lead screw got 1mm pitch
*
Actually how you get your steps/mm?


Drian
post Mar 28 2016, 11:50 AM

Look at all my stars!!
*******
Senior Member
4,999 posts

Joined: Jan 2003


QUOTE(Prosperer @ Mar 28 2016, 11:22 AM)
im printing in 0.2 layer height and according to prusa calculator i can print between 0.19375, 0.2 and 0.20625000000000002
i think my lead screw got 1mm pitch
*
just ignore my question above ..haha I realised you will always get multiples of your resolution for 0.2mm layer height regardless of 2mm,4mm or 8mm lead screw.




Drian
post Mar 28 2016, 12:02 PM

Look at all my stars!!
*******
Senior Member
4,999 posts

Joined: Jan 2003


QUOTE(Prosperer @ Mar 28 2016, 10:33 AM)
I didnt use any spring on my bed, is this too high for z axis 5086 s/mm before it only at 2579s/mm and after i change my stepstick the s/mm also change.
*
Your steps/mm seems off.

if your using 2mm lead, 1.8 degree stepper

it should be

1/16 ustep, 1600s/mm
1/32 ustep, 3200s/mm
1/8 ustep, 800s/mm


even if you are using 1mm lead , the 5086 steps/mm seems a little bit off.


Drian
post Mar 28 2016, 12:23 PM

Look at all my stars!!
*******
Senior Member
4,999 posts

Joined: Jan 2003


QUOTE(izzudinhafiz @ Mar 28 2016, 12:11 PM)
nono he using 8mm threaded rod. which is 1.25mm pitch. 5120 is the nominal s/mm.
*
yup your value is right.


Drian
post Mar 28 2016, 12:32 PM

Look at all my stars!!
*******
Senior Member
4,999 posts

Joined: Jan 2003


So over the weekend my Z axis motor bracket cracked and failed. So the problem is I can't print a new one until I get my existing bracket fixed, so I had no choice but to use epoxy glue to glue it together to print the replacement.
Attached Image
Attached Image

The epoxy glue process isn't perfect and the bracket no longer is 90 degree to the frame.
That caused some wobble in the Z axis and it caused the Z banding that you see in the picture.
This is the first time I've seen it.

Attached Image


So if you see Z banding, check your screw, alignment, wobbling, crack in Z axis mount etc to see whether it is the problem.


Drian
post Mar 28 2016, 12:56 PM

Look at all my stars!!
*******
Senior Member
4,999 posts

Joined: Jan 2003


QUOTE(Prosperer @ Mar 28 2016, 12:32 PM)
So i use izzudinhafiz value and get 9.8 when i print 15x15x10 cube and i adjust it and get actually 10mm on the z height with this is my current value 5278s/mm

can anyone explain why my value before i change just now and my value after i change right now is still gave the same z height?
*
errors in layer height extrusion + errors/tolerance in screw pitch + errors/tolerance in nut + bed height error/tolerance + backlash error/tolerance + microstepping error/tolerance. It adds up.

The problem with calibrating it like this is that the errors might be different for different heights.

what if you print 10mm you get 9.8mm and if you print 30mm you get 29.7mm. Notice that the error is not linear.
So if you compensate for 1 you might make another height overcompensated for example.

Also you also make your layer height no longer multiples of the stepper motor core resolution.


Drian
post Mar 30 2016, 01:50 PM

Look at all my stars!!
*******
Senior Member
4,999 posts

Joined: Jan 2003




https://www.youtube.com/watch?v=DmkBi657gYM

Just finish on the XY movements.

Next, start on the Z Axis .

This post has been edited by Drian: Mar 30 2016, 02:22 PM
Drian
post Mar 30 2016, 02:23 PM

Look at all my stars!!
*******
Senior Member
4,999 posts

Joined: Jan 2003


edited smile.gif
Drian
post Mar 30 2016, 05:26 PM

Look at all my stars!!
*******
Senior Member
4,999 posts

Joined: Jan 2003


QUOTE(Prosperer @ Mar 30 2016, 04:16 PM)
Where do you home your hotend mount? mine is at the back right i want to make it home at front right but still finding which line to edit sweat.gif
*
Hmm dunno yet. Come to think about it , I have not design in the end stop mount. Probably something I need to think about after I'm done with the Z axis.
Drian
post Mar 31 2016, 10:12 AM

Look at all my stars!!
*******
Senior Member
4,999 posts

Joined: Jan 2003


QUOTE(izzudinhafiz @ Mar 30 2016, 07:57 PM)
the two motors are connected to the same belts? How do you independently move the X or Y axis?

Edit: Nvm. found the source http://corexy.com/theory.html . Actually quite smart! You force parallelism on the structure without needing high tolerance rails.

Although how about stepper misstep? like minor occasional misstep? its going to be translated in both axis...
*
Attached Image

Yes if you look at the equation motion if one of the stepper motor skips a step then it is reflected to both X and Y axis.
However the error is spread out between X and Y. 1/2 to be exact. So If the stepper motor skips one step, X will have 1/2 mm/step error and Y will also have 1/2 mm/step error.

One common problem that Core XY have is belt tension if both motors have a big difference in belt tension it will translate into X and Y error and when you print a circle it becomes slightly skewed.

This post has been edited by Drian: Mar 31 2016, 10:13 AM
Drian
post Apr 4 2016, 12:30 PM

Look at all my stars!!
*******
Senior Member
4,999 posts

Joined: Jan 2003


QUOTE(Prosperer @ Apr 1 2016, 10:08 PM)
I think he mean the curling that happened on the corner of the model the overhang area that not contact with the heated bed.

Btw i never tried taobao, how fast is the shipment? i think i want new digital caliper biggrin.gif
*
what is wrong with your existing one?

This post has been edited by Drian: Apr 4 2016, 12:31 PM
Drian
post Apr 4 2016, 05:08 PM

Look at all my stars!!
*******
Senior Member
4,999 posts

Joined: Jan 2003


QUOTE(Prosperer @ Apr 4 2016, 02:41 PM)
the one that i used right now is the cheap one that i bought from lelong and there still a gap when i closed it, maybe around 0.1mm
*
same like mine, haha but mine is 2 digit precision usually 0.01-0.04mm


Drian
post Apr 6 2016, 10:40 AM

Look at all my stars!!
*******
Senior Member
4,999 posts

Joined: Jan 2003


QUOTE(Prosperer @ Apr 4 2016, 06:28 PM)
hahaha, how your build? complete already?
*
Not yet, starting to print my parts in PETG. Encountered some issue.
Drian
post Apr 6 2016, 10:41 AM

Look at all my stars!!
*******
Senior Member
4,999 posts

Joined: Jan 2003


QUOTE(Prosperer @ Apr 5 2016, 09:48 PM)
Still waiting for parts to arrived before i can assemble all part for the frame


*
thumbup.gif
Drian
post Apr 6 2016, 10:43 AM

Look at all my stars!!
*******
Senior Member
4,999 posts

Joined: Jan 2003


QUOTE(hihi @ Apr 5 2016, 09:24 AM)
Tabao shipping time depend, because i'll request to send to one place then only send back to malaysia.
I plan to buy a capacitative sensor also. We can buy together. The digital caliper, saw a lot with different pricing, any budget in mind?
*
Can tumpang one more ?, I want a digital dial gauge to measure bed flatness.

This post has been edited by Drian: Apr 6 2016, 10:44 AM
Drian
post Apr 6 2016, 11:42 AM

Look at all my stars!!
*******
Senior Member
4,999 posts

Joined: Jan 2003


Just wanted to write a short review on a PETG Filament I bought from Ali.

Started to transition my prototype@15%infill@PLA parts to PETG@50%infill.
Here are some of my observation:-

-I printed using 245C and 80c bed temperature. No enclosure or cooling fan is required.
-PETG is more stringy than PLA. You get very fine spider web if crossing over another print.
No big deal for 3d printed parts as you can just pull them away but a consideration for things that shows off print details (figurine etc)
-Gloss surface that looks "metallic".. which is pretty nice.
-Slightly more flexible than PLA for the same infill setting.
-Warpage and shrinkage comparable to PLA. (didn't notice any difference in this). Good for big parts.
-Have to print slower. I'm now usually put 70% of my PLA print speed. (PLA print speed 40mm/s outer perimeter, 75mm/s infill)

Conclusion:-

I believe PETG can replace ABS in 95% of all use case. Certain things that ABS still do better, withstand higher temperature (105c glass transition temperature vs 80c for PETG vs 60c for PLA), less stringy for PLA.

In the process of printing this I found some issues with the RAMPS board. Will share in the next post.


Attached Image
Attached Image
Attached Image

This post has been edited by Drian: Apr 6 2016, 11:43 AM

4 Pages < 1 2 3 4 >Top
 

Change to:
| Lo-Fi Version
0.0231sec    0.47    7 queries    GZIP Disabled
Time is now: 26th November 2025 - 10:07 PM