RE: VFA Artifacts on X+Y Straight Edges
Same observation. 45mm/s (structural profiles) and 200mm/s (speed profiles) give the best results.
Interestingly it's the opposite on my XL. Around 120mm/s there are zero VFA's.
I have prepared VFA tests for Core One from 40 to 350 mm/s. Enjoy!
I observe significant VFAs at 80 and 90 mm/s, some at 110 mm/s and a little below 80 and at 160 and 170 mm/s. The rest is ok, especially the higher printing speeds.
Seems like the silent mode off and SPEED printing profiles are the way to go!
RE: VFA Artifacts on X+Y Straight Edges
I used this one: https://www.printables.com/model/1189256-vfa-test-prusa-core-one
And I get VFA all the way as you can see here (right is the bottom so 40mm/s and left is higher speed up to 140)
RE: VFA Artifacts on X+Y Straight Edges
I am not this kind of electrical engineer, but I was reading the data sheet as well. From what I can guess - it probably has something to do with the sine-wave look up table that is provided for microstep current. It seems like this LUT can be configured:
The TMC2130 provides a programmable look-up table for storing the microstep current wave. As a default, the table is pre-programmed with a sine wave, which is a good starting point for most stepper motors. Reprogramming the table to a motor specific wave allows drastically improved microstepping especially with low-cost motors.
But, thats mostly just a guess. It seems that there is some amount of micro stepping interpolation that can be configured as well.
From what I can tell from the implementation (mostly in phase_stepping/, precise_stepping/, and input_shaping/, in lib/Marlin/Marlin/src/feature/), that seems to be what phase stepping is doing.
RE: VFA Artifacts on X+Y Straight Edges
I've put toothed idlers on the XL and it made no notable difference.
I have some for the Core One and will put them on when I get the printer and they release the printable parts (these are needed to modify for the 1mm wider pulley), but I don't expect it to make any difference.
What are the dimensions of the toothed idlers you got, and where do you get them? I'd much rather buy from a reputable manufacturer/store, but I found some on amazon that get okay reviews (but no mechanical specifications): https://www.amazon.com/WINSINN-Aluminum-Timing-Pulley-Printer/dp/B07BPLPZWN
RE:
I've put toothed idlers on the XL and it made no notable difference.
I have some for the Core One and will put them on when I get the printer and they release the printable parts (these are needed to modify for the 1mm wider pulley), but I don't expect it to make any difference.
What are the dimensions of the toothed idlers you got, and where do you get them? I'd much rather buy from a reputable manufacturer/store, but I found some on amazon that get okay reviews (but no mechanical specifications): https://www.amazon.com/WINSINN-Aluminum-Timing-Pulley-Printer/dp/B07BPLPZWN
I bought them here.
They are 1mm wider than the stock pulley so I had to modify and reprint the core XY Gantry mounts.
Click on the picture to see the whole thing
RE: VFA Artifacts on X+Y Straight Edges
Posted by: @brian-12
I bought them here.
They are 1mm wider than the stock pulley so I had to modify and reprint the core XY Gantry mounts.
Click on the picture to see the whole thing
Thanks! I noticed that they also have a 16 tooth version [1] rather than 20 tooth. It might fit without modifications, though I wonder if wear would be a problem.
RE: VFA Artifacts on X+Y Straight Edges
How to know which one to choose? 16T or 20T?
RE: VFA Artifacts on X+Y Straight Edges
Posted by: @brian-12
I bought them here.
They are 1mm wider than the stock pulley so I had to modify and reprint the core XY Gantry mounts.
Click on the picture to see the whole thing
Thanks! I noticed that they also have a 16 tooth version [1] rather than 20 tooth. It might fit without modifications, though I wonder if wear would be a problem.
Your belt length would be affected with the smaller pulley. It's possible the length difference could be accounted for with the belt tensioners.
That being said, my suspect is that the width doesn't change, so you're not gaining anything. I know that there were some pulleys on Amazon that were the exact same dimensions as the Prusa pulley, but I was suspect of how accurate the teeth profiles would be. Since I was trying to eliminate VFA's I didn't want to risk a non precise tooth profile causing issues. So I just stuck with the gates pulleys.
It's super easy to modify the gantry parts, I just removed .5mm from each side of where the pulley mounted in cad. The hardest part of the whole thing was getting the Prusament PCCF because it's always out of stock.
Next time I may try the priline PCCF. I'm willing to bet the Prusament filament profile would work just fine for printing.
RE: VFA Artifacts on X+Y Straight Edges
Next time I may try the priline PCCF
I have had good experiences with the Iemai pccf.
I printed the parts for MK4S and MMU3 with it.
RE: VFA Artifacts on X+Y Straight Edges
Where did you get the Prusa parts from? are they already published?
RE: VFA Artifacts on X+Y Straight Edges
For the Core One they are not out yet. For the XL they are on Printables.
RE: VFA Artifacts on X+Y Straight Edges
As I pointed out up thread there is only one idler per belt that has the teeth oriented toward the idler. This particular idler moves slower on a diagonal movement compared to a pure x or y movement. Since the VFAs increase in frequency on a diagonal movement I think it rules out these particular idlers as the cause.
RE: VFA Artifacts on X+Y Straight Edges
I don't think idlers are the main cause either, many threads on many printers with limited success
RE: VFA Artifacts on X+Y Straight Edges
Toothed idler made no difference in VFAs for my mk3s+.
RE: VFA Artifacts on X+Y Straight Edges
I don't think idlers are the main cause either, many threads on many printers with limited success
Agreed, I said earlier that it made no notable difference when I changed them.
RE: VFA Artifacts on X+Y Straight Edges
I have prepared VFA tests for Core One from 40 to 350 mm/s. Enjoy!
I observe significant VFAs at 80 and 90 mm/s, some at 110 mm/s and a little below 80 and at 160 and 170 mm/s. The rest is ok, especially the higher printing speeds.
Seems like the silent mode off and SPEED printing profiles are the way to go!
Thanks for posting the models to make things a little easier.
I had to bump the temps up to +20C over your stated numbers to maintain shiny PLA (Polymaker ProPLA), with the 260-300mm/s at 290C. Had to crank up the max volumetric flow rate to get over 290mm/sec. Still doesn't sound like it gets much past 300mm/sec, certainly nowhere near 350. Of course, I can't really see precisely how fast it's going; it ain't Klipper, after all. 😛
I can replicate VFAs all the way from 30mm/sec (as slow as I could be bothered to try) up to a bit over 300mm/sec. The top-end speeds >280 showed a reduction in **some** angles (0 is slightly less obvious, 90 is much improved), and an increase in wavelength in some angles (about double at 30, with a about half that at 60 just eyeballing it), which I cannot readily explain. I also notice the 60 radial starts showing moire issues above ~250mm/sec in addition to the VFAs.
Regarding the "well, just run it at 300mm/sec, then," I've seen bandied about, that falls on its face as pressure advance and stringing artifacts show up at those speeds on even relatively sedate smooth curved corners, let alone square corners. Some filaments do NOT like to be printed that fast, which reduces the fundamental capabilities of the machine. There's a reason all the profiles don't generate 8-minute Benchys.
RE: VFA Artifacts on X+Y Straight Edges
Please all be careful about temp and PLA. Some PLA start degrading around 280-290°C, so please don’t crank temps up too high.
3D printing emits enough particules and fumes with standard temperatures, don’t try to push the limits too high!
Sorry to tell you this that way, I didn’t know how to say it in another way 😅
RE: VFA Artifacts on X+Y Straight Edges
After printing the test model linked above (at printables.com) I noticed the VFAs are not aligned on the same wall at different speeds. I would say this suggests that belt teeth / pulleys are not the cause of this.