Forum

 

Notifications
Clear all

Layer Shift, Ender 3 V2

Page 22 / 25

tosio3D
(@tosio3d)
Eminent Member Registered
Joined: 1 year ago
Posts: 26
 

Hi @tundrwd @kevin

I also had emails back and fort with Creality about ESD. That could be really an issue with the plastic belts with no connection to a functional earth like the chassis. I'm not sure if the glas bed could also have an influence I do not see any things kept through static power to the bed.

It is not allowed (at least in EU) to connect housing part directly to protective earth of the power plug. So @kevin, I hope no connection point of the proposal from Creality to attach this additional wire is directly connected to protective earth (it would need a 1 MOhm resistance in between)...yes I see the CE mark on that printer....huh

Also my point, if ESD would be cause of layer shift, I would expect that ESD would destroy some electronic parts right from the beginning and not produce such "reproducible" print failures after weeks. Also a new board (4.2.7) can not solve this with the same connectivity and wires only if they had put some additional ESD protection (like protection capacitors) to each signal wire that comes to the board to functional ground of the chassis, housing. This is a normal mitigation in such PCBA design with a good EMC strategy

I reduced printing speed to 40mm/s, opened the housing under the mainboard fan and let the part cooler fan always be on 100% speed (this cooler is connected in parallel to the mainboard cooler what I have learned from Creality, strange...). I have a 4.2.2 board and can print 22h without issues since I got the printer in January...I think I have the real Trinamic stepper drivers ;.)


ReplyQuote
kevin
(@kevin)
Noble Member Registered
Joined: 1 year ago
Posts: 1045
 

@tundrwd You already have mentioned that the V4.2.7 is MUCH better at dealing with static discharges than the 4.2.2 board. I think you are right.

Actually, these are some individual cases that the electrostatic protection of board did not works well. I hate to say the core issue is the quality issues of mainboard, but in fact that seems to be the case.

So, this is not a solution that targeting the root cause, and not all Ender 3V2 layer shift issue caused by electrostatic interference, but at least have a try if you want the layer shift gone.


ReplyQuote
kevin
(@kevin)
Noble Member Registered
Joined: 1 year ago
Posts: 1045
 

@tosio3d Ok, looks like that is not a universal method. ? ? And the "reproducible" print failures are really weird. If you print some other models, will the layer shift happen at the same height?


ReplyQuote
YvonH
(@yvonh)
Active Member Registered
Joined: 1 year ago
Posts: 19
 

Grounding stepper motors is not a bad idea no matter what.  The motors are isolated and with the movements back and forth, there is static electricity accumulating and could make its way back to the controller board.  The Trinamic datasheet recommends it also.  There are some discussions on web about it like this one:

https://forum.duet3d.com/topic/3200/grounding-steppers

I grounded all motors and running a test tonight.

 


ReplyQuote
kevin
(@kevin)
Noble Member Registered
Joined: 1 year ago
Posts: 1045
 

@yvonh Look forward to your result ? 


ReplyQuote
tundrwd
(@tundrwd)
Eminent Member Registered
Joined: 1 year ago
Posts: 42
 
Posted by: @yvonh

Grounding stepper motors is not a bad idea no matter what.  The motors are isolated and with the movements back and forth, there is static electricity accumulating and could make its way back to the controller board.  The Trinamic datasheet recommends it also.  There are some discussions on web about it like this one:

https://forum.duet3d.com/topic/3200/grounding-steppers

I grounded all motors and running a test tonight.

 

I agree - grounding to remove static is necessary, and honestly should have been done by Creality.  But grounding does not appear to be a complete solution.  That still seems to require getting the 4.2.7 motherboard.

I'm going to ground mine as well, but honestly, unless the grounding screw that they have on the power supply actually goes to ground - this isn't going to do a lot except make "everything" the same level of static.  It's got to bleed the static to actual ground.

 


ReplyQuote
YvonH
(@yvonh)
Active Member Registered
Joined: 1 year ago
Posts: 19
 

Hi all,

I got some interesting results.

First here is my setup for this test:

X: 1.24V
Y: 1.26V
Z: 1.24V
Extruder: 1.08V

VX-Jerk: 7
VX-Jerk: 7
VZ-Jerk: +000.30
Ve-Jerk: +005.00

I did not include the extra code for controlling the acceleration and jerk like I did on the last test.

I grounded all the motors relative to the power supply enclosure.  Keep in mind that the control board ground is not connected to the frame ground.

I printed 2 large pieces last night.  When I say large, I mean X / Y large but only 4mm thick.  Both took about 4 hours each.  They both came out clean, no apparent layer shifting (see attached photos).

Today I printed my 10 hour piece that I tried before and had clear layer shifts.  This time, there is still a little bit of layer shift but not as bad as before.  See attached report.

Grounding the motors seem to have a positive effect.  If the motors are not grounded, charges will accumulate on them and eventually could jump on the motors' wires and made its way to the motor drivers.  From what I understand from the Trinamic motor driver datasheet, this could reset the driver and could explain some of the layer shifting.

I only printed 3 pieces.  This is not a lot but results are promising.

This post was modified 1 year ago 4 times by YvonH

ReplyQuote
tundrwd
(@tundrwd)
Eminent Member Registered
Joined: 1 year ago
Posts: 42
 

Ok - I've about had it with this forum software.  Not sure what software - but it really screws things up, and I lose posts when halfway done.  So here it is again:

I checked continuity on a US POWER CORD, with a VM/Ohmmeter, and there is continuity between the ground plug (ON A US POWER CORD), and the screw the document mentioned in the PDF in post https://forums.creality3dofficial.com/community/postid/3502/ , has continuity.  Again, this is on the grounded US power cord supplied with the unit.

Checking around, other screws also have continuity (the motherboard enclosure) with the ground plug on a US power cord.  NOT NEUTRAL - the ground plug.

So if you have created the ground harness and are still having problems I can only see one of 3 issues:

1.  Your grounded outlet is NOT grounded.  Get a 3-wire tester from a hardware store, make sure the socket is wired correctly.  You'd be surprised how many aren't - especially when homeowners attempt to create an extra socket/plugin somewhere.  This will create a floating ground and you aren't really getting much done with that.

2.  You didn't do a good job on the harness.  Personally, I'd only rely on soldering the connectors to the wire - not just crimping.  Also - the document doesn't specify the gauge of wire to be used - but it appears to be 16 or 18 gauge.  Don't use 20 or 22 gauge.  

3.  Despite helping - your current motherboard is damaged at this point, and you'll probably need a new motherboard to fix the problem.

Even if you have already popped in a 4.2.7 mobo and everything is working - I would STILL do the wiring harness as per the document.  It should only help, and might keep your upgraded mobo from suffering the same fate as the 4.2.2 board.

I do NOT know about EU or international standards and plugs - but read up on your locality and see how to test a proper ground between a unit and your plug end (DO NOT PLUG IT INTO POWER! - just test it) - get a cheap multimeter/ohmmeter, and test to make sure it is grounded to GROUND, so the static will bleed off.


ReplyQuote
YvonH
(@yvonh)
Active Member Registered
Joined: 1 year ago
Posts: 19
 

Yesterday, after my last test, I took apart the control board and package it, ready to be shipped.  My supplier will be sending me a new board.  It will probably be another 4.2.2, I'm guessing.

I agree that it might already be damaged before I grounded the motors.  The motors will stay grounded from now on.  I will post more tests with the new board.


ReplyQuote
hantuuk
(@hantuuk)
Eminent Member Registered
Joined: 1 year ago
Posts: 40
 

@tundrwd Thanks for the heads up on how to ground the motors effectively. I do notice that the motors enclosures seem to have a coating on the case which throws me a bit when I thought I haven’t grounded it properly. I had to scrape the metal bit to make sure the motor case is properly grounded using the mounting screws. Phew.

anyway, my machine is an ender 5 pro with the very old v1.1.5 board. I complained to creality and after lots of emails, videos and photos they said they would replace the board with a 4.2.7 board, but they sent me a 4.2.2 board instead. After a few more emails to and fro they said they will me a 4.2.7 board. Oh well we’ll see. 

cheers

Creality Ender 5 Pro


ReplyQuote
tosio3D
(@tosio3d)
Eminent Member Registered
Joined: 1 year ago
Posts: 26
 
Posted by: @yvonh

Keep in mind that the control board ground is not connected to the frame ground.

Hi @yvonh

thank you very much for the information. I want to do the same and fully agree with your opinion about the trinamic driver.

Board ground (digital ground) should be normally connected with an RC network to a functional earth/protective earth, but never directly. I'm not sure what kind of instrument type class Creality is claiming under the CE (it is their manufacturer declaration), but I think it would be not allowed to connect housing parts together with screws for a FE/PE...so a good protection wiring is more than appreciated in such a case...


YvonH liked
ReplyQuote
Sbern87
(@sbern87)
New Member Registered
Joined: 1 year ago
Posts: 3
 

I wanted to add in that I am having the same issue. Small prints come out perfectly, but larger prints have layer shifts, often starting after the 3 or 4th layer and accompanied by knocking noise from the steeper motor. I contacted the amazon reseller, amazon and creality this morning. I will follow up when I hear back.

It shouldn't be our responsibility to buy the working 4.2.7 board to have a working 3D printer. I bought my 3V2 in late February of 2021.


ReplyQuote
YvonH
(@yvonh)
Active Member Registered
Joined: 1 year ago
Posts: 19
 

If you hear knocking noise, this might means there is something blocking the movement.  Make sure that each axis runs smoothly when not powered by slowly pushing them with your hand. My Y-axis was running rough when I slowly moved it. It was one of the fixed wheels that was screwed too tight.


ReplyQuote
joel52334
(@joel52334)
Active Member Registered
Joined: 1 year ago
Posts: 5
 

I've been printing 4 layer rafts. Adhesion to the bed is what I've found is one issue to this inquiry.  Is your printer "level"?  Is it placed in a "high traffic" place in your house?  Vibrations could cause layer shift. 

My 5 "Go to" folks are Joel, Thomas, Steve & Angus (on youtube & their q&a for patrons)  I'm a patron of Tom & Gina (the primary on the OctoPrint application).   

These folks really work on problem solving.  They seem to be the ones that get "almost all" 3d printers to review & test! 

I've been using glue sticks, hair spray (Aquanet), and just got some water-based "professional" adhesion material "Prevailent T-80" it is ~$12 on Amazon.   I've not been in the mood to soak the goo/glue off with 91% ETOH  4-5 paper towels in a 1gal Ziploc bag, I let soak for at least 30 minutes.  Then I scrape off the goo...  I have a variety of putty knives.  Plastic, Creality, 2" Flexible, 2" Stiff.  Each serves a purpose. 


ReplyQuote
Sbern87
(@sbern87)
New Member Registered
Joined: 1 year ago
Posts: 3
 

@yvonh Thanks! I think you just saved me the hassle of going through the warranty. I had one nut that was too tight! I thought I had checked everything and jumped to conclusions. I still have the motherboard cover open just incase, but I had a successful large print last night that went 5.5 hours and have another one going right now with no issues. My design isn't complicated, just a 4"x5" empty box I will be using for mold making. The box goes with the base I am printing right now.


YvonH liked
ReplyQuote
Page 22 / 25