ESD and 3D printers from filament?

[richnormand]

A few days ago I came across this thread on the Prusa forum and I thought it could be of interest to folks here too.
This might apply to many other printers and might be the cause of "mysterious" resets in the middle of a long print.
My printer is a Prusa MK3S and has been working well for about two years.

https://forum.prusaprinters.org/forum/english-forum-general-discussion-announcements-and-releases/static-shocks-when-touching-the-build-sheet/
https://forum.prusaprinters.org/forum/english-forum-general-discussion-announcements-and-releases/static-shocks-when-touching-the-build-sheet/paged/2/

Although it started about ESD corrupting the display I remember other comments in various forums about a dry carpet ESD resetting the printer and in some cases having to replace the controller board. In winter the house has very low humidity and, although very good for the filament spools, it is nasty for ESD. I remember having to scratch the paint off to ensure all the parts of the printer were well bonded to the PSU ground (as well as the powerline ground) and ran a ground wire to my metal baseplate and table (as the four printer feet are isolating it). The LCD screen and rotary controls are housed in a PETG enclosure so any ESD from the display metal edge would be carried out the circuit readily.

If you get deeper in the thread I got to check to see if the build plate was grounded. It is not, nor is the all metal extruder assembly. Everything around there is made of plastic so none of the rails, stepper motors, bearing/rods are bonded to AC ground but all interconnected to the controller board eventually from floating assemblies. Easy to see how an ESD could get to the control board.

I ran a simple experiment with a Keithley 617 electrometer attached to the extruder cooling fins. There are two cartridges clamped on the extruder. The heater and the thermistor are both insulated from the extruder assembly; however it is easy to see that an ESD either from touching it or from touching the metal build plate could get to the control board jumping through the devices.

I used the Coulomb mode and then the voltage mode to monitor what happens during the process. Nothing, apart from random noise happened during the heating phase and the self leveling phase, but the moment the extrusion of the filament started charges appeared. I used the built-in first layer test pattern. It is basically a zigzag across the build plate with a simple 1x1 cm stamp at the end. The Coulomb mode went in overload in the middle. The voltage went to 50 to over 100V during the process. Even got over 150 in some runs to nearly 200-300 by using a tissue to induce a small charge on the filament prior to extrusion, probably similar to the filament cleaning things used by some people. This is for one layer.

I suspect that with a dry environment and a floating extruder head during a long print the accumulated charge could jump the thermistor and zap the controller board causing a uP reset and such. It does looks a lot like a van der Graff generator accumulating charge from an insulated belt carrying it to the globe for me. Depends on so many factors it is hard to tell.

How many other brand of 3D printers could be affected? Anyone has checked if the extruder assembly is grounded?

Thanks to the comment from Joan in the Prusa forum for seeding this inquiry!
Comments welcome.

Cheers


 

[MarkF]

I have a Creality CR-10 Mini.
There is no continuity between any of the printer assemblies where there are V-slot wheels.

No continuity between any of these:
  1)  Control box
  2)  Main frame
  3)  Build plate
  4)  X-axis gantry
  5)  Hot end

Interesting ESD problem.  My printer is in a small bedroom with carpeted floors. 
I have zapped it before but never had any issues.

[mftaylor]

Creality has a official grounding method for Ender 3 V2 to solve the layer shift issues people see.   Fixed the layer shift for me once it was installed.

[richnormand]

Just found this link where someone does make the van der Graaf comparison...
https://3dprinting.stackexchange.com/questions/15577/effects-of-static-on-3d-printers-print-quality
So... I was not the first one to think about it looks like

So the issue of shifted layers or resets in the middle of a long print might be related to ungrounded extrusion heads. There are so many factors such as the filament, its initial charge on the spool, if it was cleaned by rubbing a soft cloth on it, humidity and temperature, electronics sensitivity...

I just grounded my setup with a 100k resistor to the frame and a thin wire attached to the extruder cooling fins via the ombilical.

[xrunner]

I have an Ender 3 pro. All the rails are linear rails after I upgraded them, so they all have continuity between each other. However the aluminum plate on which the glass build plate rests is not "grounded" to the rest of the system. Also, the metal parts have no continuity to the AC line ground.

I have noticed (and this is simply an annoyance) something when I start a print and need to cancel it after a few layers. The very thin and flimsy plastic waste I need to toss has a very high static charge and I can't flick it off my tool or fingers very easily. So, I'm going to look into getting a connection between the aluminum plate and rails, and if advised, grounding the metal parts of the case and rails to actual earth ground.

[richnormand]

In my case the frame is at AC plug ground via the PSU. I remember scraping the paint off each mounting points during assembly.
The all metal extruder assembly is now connected to the frame via a 100K ohms to drain static. The resistor is in case I short one of the exposed heater wire with my wrench when changing the nozzle.

[T3sl4co1l]

Did a little project for a conveyor belt company; they said the rollers generate terrifying amounts of ESD, and they have to order their plastic parts (rollers, housings, etc.) with specified resistivity to dissipate it.

Can you get filament that's formulated the same way?  Preferably making the whole machine out of that, and yeah grounding metal parts, ESD dissipative treatment/coating, etc. would be a good idea.

I also wonder what the EMC of their boards is like.  And software design.  Like, LCD for example, often it's just written continuously, on the assumption that no bits are ever missed (or double tapped).  And as long as it stays in sync, the image looks fine, but shift it a few pixels and it gets skewed, wrapping around the side.  The drawing location can be reset every frame or few, it's not a big deal or anything, a few cycles of overhead.  But if it's not done, well, so it goes.

Like, the whole fact that CNC can be done largely open loop, is remarkable enough.  It's no surprise people get layer shifts and crap like that, there's no encoder or pulse tracker or anything in the system; and it's stupid stuff like that, that would be so obvious to fix if you could just pause the system for a moment or insert a few commands/actions as it goes, whatever; but there's no room for reaction, it's just steamroll forward and pray it works.

And unfortunately, there's no way to assess this; hardware can be tested, but doing it properly in a lab is very expensive, and it's not required for US sales.  No one ever puts it in the product description; it's just expected, assumed.  Even if they did, there's no way for the end user to verify it.

It's... and it's not even caveat emptor, the customer can't even know what they're getting, many of these things are far from evident on a finished product.

So... yeh.

Tim

[xrunner]

When I start a print and need to stop it after a few layers, I've discovered something. The extremely thin and flimsy plastic waste I need to throw away has a significant static charge, making it difficult for me to flick it off my tool or fingers. As a result, I'm going to investigate how to link the aluminium plate and rails, and if necessary, how to ground the metal components of the case and rails to the actual earth ground.

I had the same problem and added all the grounding you mentioned. It didn't help me one bit. I was putting them in a plastic box with a lid next to the printer so my cat would not try to chew on the pieces. They didn't want to release from my fingers into that box. I tried a cardboard box and it solved the problem.