Hi All,
I would appreciate some help.
I have installed the 0.4 obxidian HF nozzle on one of my X1Cs, it works in that it is producing clean benchys.
I have installed the profile from the bambu website but when I select it slice times are no different from a standard nozzle. I can see the Max volumetric speed has increased to 29 and that the speeds have been adjusted (see screenshot) so why do the slice times not show it as being quicker on 17 hour print?
The slicer’s core algorithm for estimating time is notoriously inaccurate. There have been countless posts about this, and after chasing the root cause for a while, I’ve simply given up hope.
Part of the confusion is that print calculations don’t always follow the same sequence as the actual print, and even when they do, they’re not precise. A simple example is wall line width: you’d expect increasing line width (thicker walls) to speed up prints because fewer passes are required. Yet that’s not what happens.
Here’s an example:
Supposedly, this change increased internal solid fill time from 2m44s to 3m6s. The math doesn’t add up.
This is just one example. There are countless others, but ultimately, fixing this hasn’t been a priority for the developers.
In addition to @Olias explanation on the slicer print time estimates, there are further factors preventing an exploitation of shorter print times. They do depend on the model and individual key settings though.
If, for example, the model contains many overhanging regions with “slow down for overhangs” enabled, this will overrule other settings.
Similarly, printing at low layer heights can prevent the printer from actually reaching max flow.
Another item to consider is the minimum layer time. If the print is slowed in a given region to allow more cooling time, the high flow will not play to its strengths.
And of course, line widths make a bit of a difference.
& 
And yet another restraining factor: high speed needs long straight lines to reach the target speed. Very small models or models with jagged walls are constantly accelerating and decelerating and the printer never reaches the set speed.
The choice of infill also has a big impact for the same reason. E.g. I really like gyroid, but it doesn’t allow for high speed no matter how big the model is. crosshatch is much better due to the long straight lines in most layers.
thanks all, really appreciate the extra information. So have I wasted my money on this nozzle then? Bambu claim it is 50% faster! So is that just marketing that I have fallen for?
Josh
No, i think that is still true, but not under all circumstances. You have the speed benefit with larger models, that is exactly where it matters the most. But it won’t speed up a benchy by much if at all.
For me it works even up to 40mmÂł/s for many materials. I have limited it to 35mmÂł/s to be on the save side. so the +50% are very conservative.
And at regular speeds, consistency and layer adhesion far outperform the stock nozzle.
I have a proprietary test setup for measuring layer adhesion.
For example, regular sunlu white PETG, with the stock nozzle achieved on average 180N at 70mm/s and 152N at 300mm/s.
With Obxidian nozzle, it achieved ~190N no matter the speed.
That alone has convinced me to stick to it.
ok. thanks. That gives me some hope. Also found a lot of unhappy chat at the link below.
Josh
https://www.reddit.com/r/BambuLab/comments/1lmarhc/trouble_with_obxidian_high_flow_nozzle/?utm_source=share&utm_medium=web3x&utm_name=web3xcss&utm_term=1&utm_content=share_button&sort=new
Also be aware, that there are 3 relevant methods (I’m aware of) to make use of the increased max flow rate:
1.) Print speeds: this is the most obvious parameter to increase flow and thereby reduce print times. Downside is potential reduction of quality. And you can increase it only so much. Also at the same acceleration, a motion takes even longer to reach the full speed.
2.) layer height: also quite obvious → when the model is split into less layers and each layer. Downside are optically coarser resolution and reduced overhang performance.
3.) line width: this is less obvious. I often use a 0,8mm line width in single wall vase mode with a 0,4mm nozzle. Works great. For bigger models that have to be sturdy, you would usually use some more walls. By doubling line width for inner walls, you can vastly reduce the number of walls with the same final shell thickness. Keeping outer wall line width at the default value allows you to maintain resolution of fine details.
Just to open up more options for you.
thanks. Much appreciated. Will start altering some settings and see where I get.
Josh
When I installed mine, I did a manual flow calibration based on the flow rate chart provided by E3D. It made a lot of deference between just upping the flow ratio in the hope of reducing time but having poor quality.
As said in the above comments, each model slices differently but I did find there was time saved and the most rime I had saved was 4 minutes.
I’ve since removed the E3D nozzle and changed with a Diamond Back which has given me better quality at the same E3D flow rates.
In conclusion I’d recommend manual calibration if you haven’t done so already of all your filament of choice at said Volumetric flow rates.
The values seem really conservative to me. I have the 0,4mm Obxidian and I did the Orca Max Flow test up to 45mm³/s with two different PETGs and even at 230 instead of 240°C. Both showed no defects at all up to the top. No idea how much faster I could have gone. PLA matte showed first defects around 43mm³/s. I just settled on 35³/s for all filaments.
thanks both. Did you keep slow down for overhangs ticked? I asked as when I untick that the bow of the benchy starts to show gaps in the layer lines.
Josh
I have it enabled, but I didn’t test much if it would work as well without.
thanks, much appreciated,
Josh
Also need help with this nozzle please…
I have 10 kg Anycubic PETG white and I’m not able to print this filament with that nozzle.
I switched back to the standard 0.4mm nozzle, choose Generic PETG and it works fine.
Btw. I dried the PETG 10h with 60°C.
When I witch back to the HF Nozzle using the PETG HF Profile the filament does not stick to the plate, it rips appart and sticks somewhat to the nozzel. Tried some tweeks here and there but can’t get it to work. Does anyone have an eye opener form me what to do?!
Thanks guys
(can’t attach screenshots ?!)
As a start, have you tried printing with the generic profile that worked with the stock nozzle? From there you could try adjusting parameters one by one, e.g. first increase max flow to the point where you see problems.
If you tested the stock nozzle with the generic profile and the Obxidian nozzle with the Bambu HF PETG profile, it is quite possible that the problem is not in the nozzle but in the profile.
When I tested different HF PETG brands, I saw very big differences in quality at higher speeds. I haven’t tried Creality, but it is well possible that it needs quite some fine tuning to print acceptably.
Hi Alex,
thx for your input.
I’m already adjusting since a few days… and i think finally I got the solution. Just minutes ago.
The Anycubic is supposed to print an 230 to 250C…
I just tried it at 220. And what should I say… No issues, perfect first layer. Speed from the nozzle is superb. Just perfect.
It’s just the “Bambu PETG HF” Profile but reduced the temperature for first and other layers to 220°C … Yehaa
I have the same speed problem on the E3D HF ObXidian™ 0.4mm with my P1S …
5h5 minutes vs 5h10m for the same object printed… and the new settings are only for the 0,2 standard layer… you have to modify all the standar settings if you want to have also the other slicing settings…
i used much more the 0.16 and the 0.08 (for bookmarks) and now i have to use the old settings and mod them for the E3D hozzle… annoying… anyone already solved this kind of problem ? (I’m pretty new to 3D printers and also to this forum so I ask sorry if i did something wrong asking here 
)
