SoftFever / OrcaSlicer

G-code generator for 3D printers (Bambu, Prusa, Voron, VzBot, RatRig, Creality, etc.)
https://discord.gg/P4VE9UY9gJ
GNU Affero General Public License v3.0
7.4k stars 883 forks source link

Max Flow rate being ignored #7307

Open convicte opened 3 weeks ago

convicte commented 3 weeks ago

Is there an existing issue for this problem?

OrcaSlicer Version

2.1.1.

Operating System (OS)

Windows

OS Version

Windows 11

Additional system information

Ryzen 5950X 32GB DDR4

Printer

Anycubic Vyper

How to reproduce

Any max flow rate set is being completely ignored, allowing excessive speeds for the combination of layer height and deposition width. The slicer indicates maintaining set max flow rate, while exceeding it at time 2 times.

Since it's been an issue reported for several years, across several tickets, auto-closed and unresolved, I hope it's not really necessary to provide much more detailed reproduction steps, but should this not be the case, I'd be happy to do so.

The interesting part however is that the old bug report would see the UI display incorrect flow rates:

1216 and so would #2033 at first, but for 2.1.X this UI got fixed but the flow remains off as a result of speed vs. line width calculation.

The issue reported is similar to #7037 but the reporter indicates his problem being UI, while it most likely isn't.

Actual results

image

image

https://makeit-3d.de/en/tools/maximum-printing-speed.html#:~:text=Determine%20maximum%20print%20speed,calculate%20the%20maximum%20printing%20speed. image

The flow rate required to maintain 163mm/s at 0.25 layer height and 0.62 line width is roughly 25.4mm3/s image I've seen this discrepancy get even more excesive

Expected results

Max flow rate being considered when slicing

Project file & Debug log uploads

Built-in Orca Cube

Checklist of files to include

Anything else?

No response

Zogar89 commented 3 weeks ago

I disagree. Some factors can change the real flow like Flow Ratio, arachne with variable line width etc..

convicte commented 3 weeks ago

I disagree. Some factors can change the real flow like Flow Ratio, arachne with variable line width etc..

By over 25% of the set value... and at time over 50%...

You are absolutely entitled to an opinion, though all my settings should decrease the expected flow rate, not increasing it. Both the calculations and my print effects (which I systematically tested with 5 Orca cubes) show the printer is attempting to push more material than the max flow set and what the device is capable of.

So TL:DR, the max flow is not enforced.

20241030_201924.jpg

ShaneDelmore commented 3 weeks ago

I use max flow rate to control all of my speeds daily and it works great. Which is not to say you don't have an issue, only that it's not completely broken so there must be some set of circumstances where it breaks. I spent all morning today doing pressure advance testing at various flow rates and accels and orca estimate matched what klipper (mainsail UI) displayed within 5%. I wonder what is triggering the behavior you are running into.

convicte commented 3 weeks ago

If I curtail nothing but the speed, to prevent the printer from exceeding the flow cap when running calculations, I end up with the same cube printed very well.

I'd say heading towards perfection, not using ironing and ignoring the VFAs, which are unavoidable at 160-180mm/s on a bed slinger: 20241031_090322.jpg 20241031_090607.jpg

All the flow fine tuning and linear advances seem perfectly reasonable, so I arrive at max flow control as the only culprit and the UI seems to confirm it.

convicte commented 3 weeks ago

I've decided to use the VFA Orca test as a quick and simple constant flow vs. constant speed test.

With 60-200 mm/s range I get no issues maintaining the print at speed, but in corners, things go completely pear shaped: image

20241031_093153.jpg

20241031_093222.jpg

20241031_093139.jpg

I've fine-tuned the max flow through line thickness and layer height, to prevent it from going outside what my hot end can manage: image

Not sure if it's more informative or more confusing, honestly... With your recent efforts on Pressure advance and the like, @ShaneDelmore would you venture a guess what I may be dealing with here.