Pro2 Calibration ...

User avatar
Korni
Posts: 222
Joined: Wed Apr 11, 2018 8:11 am

Pro2 Calibration ...

Postby Korni » Mon Jul 30, 2018 4:57 pm

Surface, dimensions?
What do you think?
OK or better? If so, where and how to Raise3D Pro2 +
0.1 Layer
https://youtu.be/Kf3BTh1-NJg



Thank you
Pro2+ Image

Jetguy
Posts: 2700
Joined: Tue Mar 22, 2016 1:40 am
Location: In a van, down by the river

Re: Pro2 Calibration ...

Postby Jetguy » Mon Jul 30, 2018 5:43 pm

Example ringing.jpg

Lower your jerk value to reduce the ringing.
2 ways of performing that
Temporary for testing do it at the slicer and this puts the value into the gcode. The idea being try different prints with different values until you are happy.
Ideamaker Jerk setting.jpg


Once you find the ideal Jerk value, you can set it in the printer so that all prints even files you made before have the new Jerk setting applied.
Jerk setting was enabled in Raise Touch screen software on the printer at version 0.9.7

You can define the Max Jerk for X/Y, Z and E on touchscreen under Setting -> Machine -> More Settings -> Hardware -> Jerk.
For example:
Max Jerk - X/Y = 12 is default, so change this value to what you determined via testing above
Max Jerk - Z = 0.4
Max Jerk - E = 5

Jetguy
Posts: 2700
Joined: Tue Mar 22, 2016 1:40 am
Location: In a van, down by the river

Re: Pro2 Calibration ...

Postby Jetguy » Mon Jul 30, 2018 5:47 pm

The firmware setting of Jerk value.
Home screen.jpg

Select settings menu.jpg

Select hardware.jpg

Firmware settings.jpg

Jerk Settings.jpg

Jetguy
Posts: 2700
Joined: Tue Mar 22, 2016 1:40 am
Location: In a van, down by the river

Re: Pro2 Calibration ...

Postby Jetguy » Mon Jul 30, 2018 5:56 pm

Detailed explanation http://3digitalcooks.com/2014/12/marlin-planner-101/
Jerk max constants (aka DEFAULT_XYJERK, DEFAULT_ZJERK, DEFAULT_EJERK [mm/s]). Represent the instant speed increment (infinite acceleration) that an axis can handle without skipping steps.

User avatar
Korni
Posts: 222
Joined: Wed Apr 11, 2018 8:11 am

Re: Pro2 Calibration ...

Postby Korni » Mon Jul 30, 2018 6:34 pm

Hello,

these are my pro2 "jerk" settings
IMG_7454.JPG


After watching my video do you think I should set (only) "Max Schritt X / Y" to 8-10?
Max Schritt Z & E are OK?
What is E?

I do not understand the settings in the Slicer at all, since I do not recognize a connection with the attitudes in the printer settings itself:

How to fit Max Schritt X / Y, Max Schritt Z, Max Schritt E (in the printer settings) to the settings in the Slicer ????
Print Jerk?
Inner Shell Jerk?
Outer Shell Jerk?
Infill jerk?
Bottom Solid fill Jerk?
Top Solid Fill Jerk?
Travel Jerk?
What do you think would be the best start / test settings in the slicer?
EVERYTHING under jerk settings to 8-10 mm / s ????
Pro2+ Image

Jetguy
Posts: 2700
Joined: Tue Mar 22, 2016 1:40 am
Location: In a van, down by the river

Re: Pro2 Calibration ...

Postby Jetguy » Mon Jul 30, 2018 6:49 pm

I'd hope it was clear, but understand that there may be a language barrier.
This answer is for the firmware setting at the printer.
I would only adjust XY Jerk in firmware (hence why I wrote next to it and only circled it in red)
Jerk Settings.jpg

I would try 10mm/s run the same exact test print file, and see the result.
Then try 8mm/s again, same print file and see the result.
Lower jerk value slightly slows the print down because segments that before had no acceleration and deceleration subtracted now will be slower, but that's the whole point, those moves are the ones creating the ringing and ghosting.

Note, this answer is around the question of the settings in Ideamaker as a slicer creating the gcode.
If you do this at the slicer level, it then just appends the new jerk value for that given print feature on a layer.
I have to say, while someone thinks this is smart and sure adjust for even some tiny feature of a print layer, reality is, the motion planner needs time for a setting to act upon the values right? I look at the firmware, and I look at the gcode, and think it won't play back that gcode the way you intended. I'd love to be incorrect about this but doubt it works the way this is written.

So, that leads to this, if you change one, change them all.
"EVERYTHING under jerk settings to 8-10 mm / s ????"
Yes.
Ideamaker Jerk setting.jpg

I'll also say this, if you change it in firmware, disable the Ideamaker slicing setting of appending jerk rates in gcode.
Yes, in theory, a gcode based change overrides a firmware setting, but like I said, based on what I know, the gcode is the less preferred method.
Dynamically changing a setting in the planner- seriously, go back and read how the planner works, would cause mayhem.
Last edited by Jetguy on Mon Jul 30, 2018 6:58 pm, edited 3 times in total.

Jetguy
Posts: 2700
Joined: Tue Mar 22, 2016 1:40 am
Location: In a van, down by the river

Re: Pro2 Calibration ...

Postby Jetguy » Mon Jul 30, 2018 6:50 pm

Also, E is extrusion, the feeder motor pushing the plastic wire into the nozzle.
Again, no reason to adjust that value.

User avatar
Korni
Posts: 222
Joined: Wed Apr 11, 2018 8:11 am

Re: Pro2 Calibration ...

Postby Korni » Mon Jul 30, 2018 7:08 pm

In summary:
BETTER in the firmware "Max step X / Y in 8 or 10 change (and each print a test print cube) and NOT necessarily in the slicer?
DO NOT change "Max Step E"?
What about "Max Step Z" (0.5mm / s)?

MANY thanks for your effort and patience
Pro2+ Image

Jetguy
Posts: 2700
Joined: Tue Mar 22, 2016 1:40 am
Location: In a van, down by the river

Re: Pro2 Calibration ...

Postby Jetguy » Mon Jul 30, 2018 7:14 pm

Korni wrote:In summary:
BETTER in the firmware "Max step X / Y in 8 or 10 change (and each print a test print cube) and NOT necessarily in the slicer?

Yes, actually better in firmware even though I started off saying you could test from the slicer, the more I look at how it is structured, firmware is just the better choice. That said, I always said do it in firmware as the last step.

DO NOT change "Max Step E" or "Max Step Z" (0.5mm / s).
Correct, no reason to change either of those.

User avatar
Korni
Posts: 222
Joined: Wed Apr 11, 2018 8:11 am

Re: Pro2 Calibration ...

Postby Korni » Tue Jul 31, 2018 7:31 pm

Have now tested with X / Y Jerk (firmware) "10mm / s", SOMETHING better .... will tomorrow X / Y Jerk "8mm / s" test.
But measurements .... X = 19.94-19.96, Y = 19.87-19.99, Z = 20.16-20.20
How can I correct THAT?


THX
Pro2+ Image

Jetguy
Posts: 2700
Joined: Tue Mar 22, 2016 1:40 am
Location: In a van, down by the river

Re: Pro2 Calibration ...

Postby Jetguy » Tue Jul 31, 2018 8:01 pm

I would not correct that "But measurements .... X = 19.94-19.96, Y = 19.87-19.99, Z = 20.16-20.20"

First, understand that a 20mm test cube on a 300mm build area, you do not tune size based on a test cube that is showing a 0.05mm error.
Instead, you print something 300mm long on an axis, then see what it is- then make a decision.
Second- Z is a combination of error. Z is and always will be user and process error prone. The first massive assumption is the first layer to bed surface gap is the same as the software is expecting. Since this is a user adjustment (Z homing position) not to mention bed leveling and surface variation, some amount of error happens just on first layer. You will fight this as long as you do 3D printing, true in every known format- even SLA printing because you the user determine the Z zero starting point. Then the Z error is cumulative as each layer builds upon the previous layer. Last, unlike XY motion, you have a binary layer. It either is a layer or no layer sliced. That means depending on your chosen layer height and all settings and details like first layer height, then may result in a rounding error since again, there either is a layer or is not layer- but no %layer to make it exact.

Jetguy
Posts: 2700
Joined: Tue Mar 22, 2016 1:40 am
Location: In a van, down by the river

Re: Pro2 Calibration ...

Postby Jetguy » Tue Jul 31, 2018 8:19 pm

Again, just stop now. You can tune if you want to, but first make smart decisions.
Do not adjust steps per mm based on a 20mm test cube. This is just asking for trouble. The test sample size is too small.
Next, all plastic has some shrinkage. This is not and should not be adjusted for in steps per mm. You should consider this as a known variable or limitation in printing- not unlike injection molding.

Flow rate has some impact on extrusion width (result), remember, difference between theoretical and reality. The gocde path of where the nozzle goes is the slicer extrusion width. The actual noodle of plastic out of the nozzle is a result of nozzle actual size, actual nozzle height to the previous layer or first layer, incoming filament actual diameter VS value entered into the slicer, flowrate override, actual extruder feeder steps per mm accuracy and grip for a given filament.
How do you properly calibrate all this?
#1 measure filament diameter for several places over a meter or so of filament and take the average. Enter that value into the slicer
#2 Calibrate steps per mm, but ensure flowrate is set 100% on the LCD before this test. Using a marker, mark a line on filament 100mm above the extruder. Using the control panel to preheat the extruder, now use the jog menu and command 100mm of filament to extrude (the jog distance is 10mm per button press- so 10 presses will also accumulate 100mm increase in distance extruded) Validate now that the 100mm mark is now at the entrance of the extruder. Be sure to note if your ruler when measuring this has the zero point away from the edge of the ruler on top. For best accuracy, do this test for 200-300, or even more distance to better understand the error.
Those 2 things now mean the filament loaded, and the extruder steps per mm are accurate for extrusion volume.
However, with all things in life, you now adjust the slicing profile to give correct 100% infill and further examine extrusion volume.
So, you print a test cube (20mm is fine for this test) because you print with 100% infill and also ensure, all flow rate values in the profile and filament settings are 100%. Print the test cube and observe the nozzle placing the layers- especially starting at 1/2 total height until the print is finished. Since we are using a lot of filament in a small space, we see exactly does the physical extruder output volume exactly match the theoretical slice path volume. It's then here you can choose to adjust the filament or profile extrusion flowrate modifier value to fine dial this in. A common mistake I have seen is people then use calipers or other method to see how flat the top surface of the 20mm test cube is. This is NOT measuring for any distance. I cannot stress this enough, people try to combine calibration concepts and measure this 20mm test cube for distance. That is wrong. We are doing this test to fine tune in extrusion volume, which while there is some phsycial aspect, the cumulative nature of 100% infill layer building on top of layer, you can see as little as a 1% error because it's cumulative.
#3 then and only then, after 1, 2, and 3 are in place, and now you can begin XY distance testing, again, using the largest object you can to use the entire 300mm distance, and then see the error.

All that said, again, on your numbers, I would not adjust anything. Not based on a single test cube and that low of an error.

Jetguy
Posts: 2700
Joined: Tue Mar 22, 2016 1:40 am
Location: In a van, down by the river

Re: Pro2 Calibration ...

Postby Jetguy » Tue Jul 31, 2018 8:30 pm

Again, a huge concept is extrusion volume calibration.
The first half of the test is basic mechanics, does the extruder push the linear distance of filament as commanded? This is a variable because understand that the very teeth gripping into the side of different filament, that filament has a different hardness. That determines the effective diameter of the gear driving the filament (it's not tooth tip diameter and not root tooth diameter, but somewhere in between those values with various filaments). The idea thought is to get this close. Important that we do not have any firmware based modifiers (LCD screen flowrate must be at 100% during the test) when commanding the 100mm extrusion distance.
So then the idea is we have a machine that follows the very gcode we send it.

However, we have more variables- that is actual filament diameter. This matters because think of the filament like and infinite length cylinder. You tell the slicer the diameter, and for a given segment of gcode exrtrusion, it uses that filament diameter to put the E extrusion length on the line of gcode. That's why you measure filament diameter as it's not all exactly 1.75mm. The slicer has to know what you are feeding.
Then, if all that is right, and the nozzle diameter and the layer height and all that is right, then the slicer is logically making a path to fill between the lines on a given layer. That path has a length, height, and width, thus a volume and again, the slicer is using that given filament diameter to reverse the math and say how much filament do I push to equal the same logical volume. If all was right in the world, and every measurement was accurate and true, then flowrate would be 100%, the machine would follow the gcode exactly to the mm, the filament would extrude and your object would be perfect.-- oops, before it shrank from thermal properties of the plastic.

Jetguy
Posts: 2700
Joined: Tue Mar 22, 2016 1:40 am
Location: In a van, down by the river

Re: Pro2 Calibration ...

Postby Jetguy » Tue Jul 31, 2018 8:34 pm

The alternate method you might find floating around the internet on extrusion volume calibration uses a single wall print, and you measure the wall. Here is why I do not agree with that method.
#1 the test sample of total extrusion length of filament is small. A fraction of what a 100% infill test cube is.
#2 The error of measuring a thin wall, and knowing that extrusion width is a function of how much plastic the nozzle squirts out and becomes oval based on the nozzle smashing it to the previous layer. All kinds of room for error here, and you are using this single test to determine extrusion volume.

Some people love and swear by this method of thin wall extrusion test. I've explained why and how, you can use whatever you feel best.

Jetguy
Posts: 2700
Joined: Tue Mar 22, 2016 1:40 am
Location: In a van, down by the river

Re: Pro2 Calibration ...

Postby Jetguy » Tue Jul 31, 2018 8:55 pm

This is where the 100% infill cube calibration method was documented http://www.sailfishfirmware.com/doc/tun ... 8-730005.1


Return to “Pro2 Series”

Who is online

Users browsing this forum: No registered users and 1 guest