Wavy Surface

zencoffeehk

Wavy Surface
January 13, 2016 12:22AM

Registered: 9 years ago
Posts: 12

I am new to 3D printing! I built a kossel mini with RADDs board, 8825 stepper and repeiter, but the test cube surface always having wavy pattern on the XY face front and rear!
Do anyone have the same problem? Please Help!

Edited 1 time(s). Last edit at 01/13/2016 12:24AM by zencoffeehk.

Reply Quote

Attachments:
open | download - XY front.jpg (28 KB)
open | download - XY back.jpg (29.4 KB)
open | download - YZ.jpg (26.4 KB)
open | download - XZ.jpg (29.4 KB)

RRuser

Re: Wavy Surface
January 13, 2016 06:44AM

Registered: 10 years ago
Posts: 444

What side is which?

Reply Quote

nebbian

Re: Wavy Surface
January 13, 2016 08:30AM

Registered: 8 years ago
Posts: 916

That's normal when using belts around an idler bearing, instead of a belt around a toothed idler pulley (with bearing).

It's due to the belt teeth hitting the idler bearing and causing the carriage to move slightly.

To lessen the effect, slow down your external perimeters, or install toothed idler pulleys.

You can also flip the belt so that the flat part goes around the idler bearing, I did this for a while on my Kossel Mini. You do need to dremel out part of the carriage so that the belt doesn't rub on the carriage where it slowly turns 180 degrees to go around the bottom toothed pulley.

Edited 1 time(s). Last edit at 01/13/2016 08:32AM by nebbian.

Reply Quote

zencoffeehk

Re: Wavy Surface
January 13, 2016 08:35AM

Registered: 9 years ago
Posts: 12

Thanks will check it on Friday!

Reply Quote

Attachments:
open | download - IMG_20160113_213401.jpg (484.5 KB)

JamesK

Re: Wavy Surface
January 14, 2016 01:38PM

Registered: 9 years ago
Posts: 1,873

Another possible explanation is the drv8825 blanking time (missing microsteps in the low part of the voltage range).

Take a look at the excellent analysis here: [cabristor.blogspot.ca]

The short form: Put a pair of back to back diiodes in-line with each coil of each stepper motor. The original author recommends 4 diodes for each coil, but in my tests a single pair seemed to do the job.

Reply Quote

zencoffeehk

Re: Wavy Surface
January 15, 2016 02:04AM

Registered: 9 years ago
Posts: 12

Thanks! Can u provide a photo showing how u link up the diodes with the step motor please! smiling smiley

Reply Quote

JamesK

Re: Wavy Surface
January 15, 2016 05:40AM

Registered: 9 years ago
Posts: 1,873

I used an offcut of veroboard (copper stripboard for prototyping) to mount the diodes, but it's tucked away inside my electronics enclosure right now. I'll try and remember to take a photo next time I have the box open.

This diagram is the gist showing the diodes in-line with the outer two conductors of the 4-way connection between the stepper driver and the motor:

DRV8825-fix.png

Except I just used one diode in each direction, so in each outer wire I put:

   |
 /   \
+     -
D     D
-     +
 \   /
   |

Edited 1 time(s). Last edit at 01/15/2016 05:40AM by JamesK.

Reply Quote

RRuser

Re: Wavy Surface
January 16, 2016 02:35AM

Registered: 10 years ago
Posts: 444

So this is stepper ripple again, correct?

Reply Quote

zencoffeehk

Re: Wavy Surface
January 16, 2016 10:43AM

Registered: 9 years ago
Posts: 12

Guess so! Is that mean most of printer will having same problem?

Reply Quote

JamesK

Re: Wavy Surface
January 16, 2016 10:53AM

Registered: 9 years ago
Posts: 1,873

It's one of those "it depends" things. The problem affects drv8825 and not (as far as I know) the a4988, and only impacts microsteps in the lower 12% of the voltage range. Whether or not your motor is operating in that part of the range depends on the coil resistance and the supply voltage, and whether it shows up as a visible artefact depends on the mechanical arrangement of the printer. In my case it only really made much difference on the extruder which was pulsing before fitting the diodes, causing waves in the extruded filament. My understanding is that the motion geometry of delta printers is prone to showing the problem more easily than Cartesian ones.

Reply Quote

zencoffeehk

Re: Wavy Surface
January 16, 2016 12:07PM

Registered: 9 years ago
Posts: 12

But i already tried the 4988 as well, same pattern!

Reply Quote

JamesK

Re: Wavy Surface
January 16, 2016 12:13PM

Registered: 9 years ago
Posts: 1,873

Ah well, in that case it must be some other problem. Back to Nebbian's suggestion then?

Reply Quote

nebbian

Re: Wavy Surface
January 16, 2016 05:01PM

Registered: 8 years ago
Posts: 916

Have you tried simply slowing down the print?

Reply Quote

zencoffeehk

Re: Wavy Surface
January 16, 2016 08:29PM

Registered: 9 years ago
Posts: 12

Yeap i did, slow down the speed, acceleration........

I changed my control board from Mega to Radds, stepper , motor, belt.........will add the diodes and the pully soon!

Edited 1 time(s). Last edit at 01/16/2016 10:11PM by zencoffeehk.

Reply Quote

RRuser

Re: Wavy Surface
January 17, 2016 06:27AM

Registered: 10 years ago
Posts: 444

Isn't higher speed what you want? Gives higher voltage.

And/or maybe increase the drive voltage.

Reply Quote

zencoffeehk

Re: Wavy Surface
January 17, 2016 06:40AM

Registered: 9 years ago
Posts: 12

u mean vref on the stepper control?

Reply Quote

JamesK

Re: Wavy Surface
January 17, 2016 11:59AM

Registered: 9 years ago
Posts: 1,873

What microstep setting do you use?

Reply Quote

zencoffeehk

Re: Wavy Surface
January 17, 2016 10:20PM

Registered: 9 years ago
Posts: 12

U mean the Segments/s for printing (150) under Fimrware EEPROM Setting?

Reply Quote

JamesK

Re: Wavy Surface
January 17, 2016 10:51PM

Registered: 9 years ago
Posts: 1,873

No, the microstep setting on the stepper motor drivers, usually 16 for 4988 and 32 for drv8825. On RAMPS it's set with jumpers under the driver modules, I don't know what the setup is for RADDs. I just wondered if you'd decided to use full steps for some reason - that might look like the results you are getting.

Reply Quote

zencoffeehk

Re: Wavy Surface
January 18, 2016 12:52AM

Registered: 9 years ago
Posts: 12

I used the full step for 8825!

Reply Quote

Attachments:
open | download - Radds Stepper.jpg (158.1 KB)

JamesK

Re: Wavy Surface
January 18, 2016 07:51AM

Registered: 9 years ago
Posts: 1,873

Ok, so switch it over to 1/32, multiply your steps/mm settings by 32 and enjoy your new-found resolution smiling smiley

Reply Quote

angelo

Re: Wavy Surface
January 18, 2016 11:17AM

Registered: 11 years ago
Posts: 1,807

for the DRV8825 please use all on and set the steps per mm to 1/32 MS.
sample:
GT2 pulley with 20 Tooth= 40 mm / Turn
With a Nema17 with 0,9 deg = ((400 Fullsteps/Turn * 32 MS)/ 40mm /Turn) = 320 steps per mm
Please set on repetier FW a delay from 2 ms (STEPPER_HIGH_DELAY = 2).
Reason is DRV8825 has a 1,8 micor seconds to detect the signals.

Why do you do not try the Raps128 (STEPPER_HIGH_DELAY= 0)
We have develloped it for the RADDS /Arduino Due (recomanded with 24 Volt)
[doku.radds.org]

Mein Club: [hackerspace-ffm.de]
RADDS-Shield -> Commercial [max3dshop.org]

Reply Quote

zencoffeehk

Re: Wavy Surface
January 20, 2016 09:27PM

Registered: 9 years ago
Posts: 12

Problem solved! Dear all thanks for your help, i add the diode as JamesK mension! Wavy surface gone smiling smiley

Reply Quote