Need help reducing interference

[daftscience]

Ok, so up until a few weeks ago I have had almost no issue with printing now I'm getting what I think is interference.

Here is an example:
[img]http://i.imgur.com/gB9eAGB.jpg[/img]


What I have done so far:

• Digipots 190
• Heatsinks on controllers and fan cooling them
• Disabled endstops during print in the configuration.h file
• Moved the power supply outside of the case so those wires aren't near any stepper or signal wires.
• Banged my head on my desk several times

I'm not really sure what else to do... Aside from completely rewiring everything.

[Eaglezsoar]

Could be friction in some of the joints. Have you checked the ujoints for free movement, cheapskate adjustments, etc?

[Nylocke]

Have you changed your Slic3r/KISS/Cura settings, like filament diameter or anything? Or have your steps per millimeter on your extruder changed at all? Or perhaps a temperature increase or decrease? I often get the ridges in my prints because I'm over extruding, and this could just be because you're over extruding. When you put the digipots on 190, is that an increase? I can't remember what they are set to on default. are you prating over USB or SD with the LCD? Does/Do your USB cable/LCD ribbon cables have ferrite on them? How did you wire your endstops, extruder motor, and hot end? Did you go through the center of the extrusions, or did you do what I did and wire the endstops through the extrusions, and have the extruder and hot end wired out the side panel (the unused door on the base) and up to the top and then down to the extruder and hot end?

[daftscience]

Could be friction in some of the joints. Have you checked the ujoints for free movement, cheapskate adjustments, etc?

I'm doing another print now... I'll check that when it's done.

Have you changed your Slic3r/KISS/Cura settings, like filament diameter or anything? Or have your steps per millimeter on your extruder changed at all? Or perhaps a temperature increase or decrease? I often get the ridges in my prints because I'm over extruding, and this could just be because you're over extruding. When you put the digipots on 190, is that an increase? I can't remember what they are set to on default. are you prating over USB or SD with the LCD? Does/Do your USB cable/LCD ribbon cables have ferrite on them? How did you wire your endstops, extruder motor, and hot end? Did you go through the center of the extrusions, or did you do what I did and wire the endstops through the extrusions, and have the extruder and hot end wired out the side panel (the unused door on the base) and up to the top and then down to the extruder and hot end?

Lots here...
I should have specified what's happening. There is a shift of the print and then a shift back. It seems to be random, given the same print has different shifts (so I had ruled out over extrusion.)

Digipots have gone from 175 -> 190
Forgot to mention I removed my LCD Panel
Endstops run through the center of the extrusion (But again, they are disabled through firmware)
Motors are wired through the center
Extruder is going out the door up and over the top then down through the center.

[Nylocke]

I would be suspicious of the endstops disabled in the firmware, unless you actually found the code that disables them, or you wrote that code. I guess you could test to make sure its working by pressing them a bunch during a print and seeing if it affects the print in any way just to be sure. If not the its probably (and hopefully) your U-joints or cheapskates, etc

[daftscience]

I would be suspicious of the endstops disabled in the firmware, unless you actually found the code that disables them, or you wrote that code. I guess you could test to make sure its working by pressing them a bunch during a print and seeing if it affects the print in any way just to be sure. If not the its probably (and hopefully) your U-joints or cheapskates, etc

Code: Select all

// You can disable endstop checking for print moves. This is needed, if you get sometimes
// false signals from your endstops. If your endstops don't give false signals, you
// can set it on for safety.
#define ALWAYS_CHECK_ENDSTOPS false

I did test it and holding endstops while printing does nothing. But it homes like normal.

And I have the magnetic arm upgrade.... :/ I think I'm going to end up rewiring it :/ :/

[Polygonhell]

I think it's unlikely the issue is electrical, if it were I wouldn't expect the shift to recover. The only concept of position in the code is the counted steps it's sent, it has no way to correct for extraneous ones.

I would suggest printing simple test shapes, it can make diagnosing a lot easier, tall cylinders are probably a good starting point, you want to print a single perimeter one with no infill, and a multi perimeter one with infil. If the single wall print doesn't exhibit the issue, you are likely over extruding, if it does, the issue is most likely medhancal IMO

[int2str]

I think it's unlikely the issue is electrical, if it were I wouldn't expect the shift to recover. The only concept of position in the code is the counted steps it's sent, it has no way to correct for extraneous ones.

+1

Especially the end-stops can hardly cause any issues. If you suspect that interference woudl cause a false trigger, then that's non-recoverable. Once a stepper comes out of sync with the others, it cannot "snap back". If you think that checking the end-stop CAUSES interference, I can assure you it does not.

[AlexBorro]

I'm having a similar issue in my just born Rostock Max. Although I have good knowledge on cartesian printers and slicers, I'm noob on deltabots.
Sorry low quality pictures from my phone, but the issue is clear.
As an electrical engineer I don't believe on electrical interference/issues.. maybe some config or mechanical issue..

Info:
- Stock Rostock Max bought on Oct/2013
- ABS at 225ºC
- 0.35 nozzle, 0.2mm layer height, 0.35mm extrusion width
- 3 perimeters (about 20mm/s external and 100mm/s internal)
- 20% infill (100mm/s)
- print from SD Card
- Repetier 0.83
- Cooler on RAMBo to cool the drivers.

The extruder seems to be calibrated, since the top infill looks good.. need some tinkering but I don't think it will cause such layer misalignment.

My aluminum U-joints have a little bit play, but just a little bit.. I mean, about 0.1mm.. I figured out quite impossible to let it friction free and no play AT ALL.
By the way, I think this is the huge drawback of this machine.. they should ship that Trick Laser arms instead of this highly-human-dependent solution.

Cheers.

Alex.

[Nylocke]

From your pictures I would suspect its over extruding, I'd double check your Extruder calibration.

[AlexBorro]

From your pictures I would suspect its over extruding, I'd double check your Extruder calibration.

But over extruding only on random layers ?!?!? That's weird... Anyway, next step will be fine tuning the extruder.

I don't have the print on hands right now, but I guess the layer is shifted, I mean, one side (in the picture) is outwards and the opposite side is inwards. I need to check that..

[Nylocke]

The effects of over extruding are misaligned layers. It's not on random layers, it's effects just are more prominent on different layers. It's constantly too much material, and it just moves to the most open space, creating random ridge patterns in the print.

[daftscience]

I've been pulling 70 hour weeks at work, so I haven't had much time to play around and wont until the new year. However, I did another print, this time dropping the flow rate, and didn't notice a difference in the shifting.

Just to elaborate, the reason I am skeptical of over extrusion is that the layers that are protruding out of one side are accompanied with an equal indentation on the opposite side. See below:
[img]http://i.imgur.com/gvnvYHv.png[/img]

I'm not sure if I explained that right to begin with, or if it's obvious from the pictures posted so far.

[Nylocke]

It was a suggestion, I get what you're saying. Its probably mechanical, lash or friction would be my bet.

[Eaglezsoar]

Disconnect the belts and check the cheapskates for free movement without any play.
While you have the belts off inspect them and the drive gears. Sometimes the gears
also develop play from the screws working loose.

[AlexBorro]

That's exactly the issue I have.

I'm wondering, if we have some kind of slop/play in the belts/motor/cheapskate, this issue will happens in all layers.. and probably the effector will not travel flat, I mean, parallel to the bed. Otherwise some play in the effector will keep the travel flat and cause similar issue.... need to check... but since daftscience is using magnetic arms, It is not usual play/slop.

[img]http://i.imgur.com/gvnvYHv.png[/img]

[daftscience]

It was a suggestion, I get what you're saying. Its probably mechanical, lash or friction would be my bet.

No worries, it was a good one.. the picture in my first post looks almost identical to a print that was over extruded.

Disconnect the belts and check the cheapskates for free movement without any play.
While you have the belts off inspect them and the drive gears. Sometimes the gears
also develop play from the screws working loose.

This is probably going to be the next thing I try when I get a chance. I disconnected the motors and ran them slowely up and down. There was one cheapskate that was considerably tighter than the others. I loosened it, but no luck.

That's exactly the issue I have.

I'm wondering, if we have some kind of slop/play in the belts/motor/cheapskate, this issue will happens in all layers.. and probably the effector will not travel flat, I mean, parallel to the bed. Otherwise some play in the effector will keep the travel flat and cause similar issue.... need to check... but since daftscience is using magnetic arms, It is not usual play/slop.

I know I don't have much play in mine. But since a lot of people are leaning toward mechanical. I might try looking for stuff stuck in the teeth of the belts when I get home.

[Eric]

Check for any debris in the magnetic joints. Something sticky, or perhaps a metal filing sticking to a magnet? Or perhaps something became worn enough that one wiggles in use?