Easy path to high quality > 150mm/s print speed?

[NeverDie]

Assuming you’ve got the flow rate to support it, the speeds in the process profile are going to cap the print speed to no more than 150mm/s:

It doesn’t matter what layer height you pick, and it doesn’t matter even what filament you pick. It’ll be the same.

Now, can I change those numbers myself. Yes. Obviously. But to what? I recently found out that some manufacturers, such as eSun, have published profiles for their filaments. Great. That’s all good. But, if they don’t also publish profiles for the process (and unless I’ve missed it, they don’t), then what good is buying their “high speed” filaments if we’re left to fend for ourselves to tune a new process every time we try a new filament to get the highest speed out of it?

I suppose there is “sport mode” and “ludicrous mode” as options, but that’s pretty high granularity versus a tuned profile, isn’t it?

I think I’m missing something basic here, but I don’t know what. How are you all dealing with this? Any advice or suggestions? Or do you just accept 150mm/s and call it a day?

Edit: If there are good tools for calibrating these speed settings, what are they? I’m not talking about flow rate. That’s already covered in orca slicer.

[NeverDie]

Well, I take it back. In the case of eSun, they do provide more than just the filament parameters, on their website:

even if it apparently didn’t load as an importable upload.

Not a big deal setting this by hand and seeing whether it works. Unfortunately, it is only specified for a 0.4mm nozzle, so I have no idea how/if it would be different for a 0.6mm nozzle.

Therefore, it somewhat answers my original question: 300mm/s is the best we can hope for, according to this particular filament maker’s PLA specs.

Is there a filament out there that would allow even higher speeds? Or is this pretty much the practical limit? I’m not trying to judge whether the actual limit is too high or too low. I’d just like to know where the practical limit truly is–without sacrificing quality–so that I can aim for that.

Edit: Unfortunately, the very same company, eSun, does not appear to have similar guidance on the parameters for their high speed PETG. I’ve only found information on PETG generally, which I guess covers all of their PETG as though they were lumped into one bucket. So, I won’t be ordering that, at least not from eSun. Maybe they’ll do better in the future.

Edit2: Oh jeez. What bozo’s. The parameters they listed above for their high speed PLA was lifted from their document for just regular PLA+. In their additional documentation, they don’t even have parameters for Bambu on their PLA High Speed.

eSUN-Fast-filaments-print-parameters.pdf

Honestly, how can anyone be so incompetent about their own product?

Edit3: One minor side-note of interest, is that comparing their recommendations regarding PLA+, it’s clear that the X1 completely blows away my old Prusa MK3S+ printer, and it even shames the voron. So, those of you who want to build vorons, know that the results might be subpar compared to what you’re already getting:

Edit4: Since eSun has no exact guidance for the X1 in regards to their high speed PLA, I thought I could look at their advice on the K1 as a proxy, to see whether it showed any improvement in print speed. Well, it doesn’t:

So, either their own advice is wrong, or there really is no speed advantage to eSun’s “high speed” PLA as compared to their regular PLA+. So, maybe they aren’t bozos after all–there simply is no difference, and they know it.

[RMB]

eSun’s downloadable profiles have the filament and print profiles to import.

The X1C is advertised as being “capable” of 500mm/s print speeds, 20k mm/s travel speeds. It’s fast but like you have discovered, printing super fast is more often than not pointless because the quality suffers greatly (model dependant), or the hotend or filament simply can’t keep up with the flow needed.
IMO, just be happy that you can get decent prints at a quicker speed than your Prusa did.

[NeverDie]

@RMB

be happy that you can get decent prints at a quicker speed than your Prusa did

Oh, believe me, I am very happy with my X1C, and even more so when I compare it to my Prusa MK3S+. Even though Prusa gives its MK3S+ users the option (at a fairly high cost) to upgrade to an MK4, he’s only claiming that the MK4 is twice as fast (without sacrificing print quality) as the MK3. Well, that’s totally a loosing proposition, as even that is well below what the X1C can do.

Anyhow, my purpose in this thread is just to get the most out of the X1C without sacrificing print quality. If it’s already as good as it gets by following the defaults, then great, we’re all done here. On the other hand, if we can do better with the machine still in its stock configuration, then it’s an easy win. I’m just wondering whether that’s possible, and if so, what exactly is possible, and how?

I’m running out of steam on pursuing this alone. I hope others jump in to the thread and share their suggestions and advice. Otherwise, there’s not much point in even posting.

[NeverDie]

LOL. As it turns out, with a 0.6mm nozzle, it’s practically a wash in regards to speed limits, because even at a flow of 22mm3/s (to use the current example of eSuns filament), it’s the flow rate that becomes the limiting factor, not the speeds in the default process profiles.

Of course, with at 0.4mm nozzle, there would likely be a significant difference between the defaults and what’s truly possible, without sacrificing quality.

QED

[RMB]

Yes, I believe you will need a higher flowing hotend to print 0.6 much faster than 0.4, let alone with a 0.8. Sometimes the time is actually longer with 0.6

[NeverDie]

And lo and behold, the speeds do change automatically to the higher figures if I switch to a 0.4mm nozzle:

So, it’s basically all good. Apparently no need to sweat the details.

[NeverDie]

From watching slant3d on youtube, I learned that all PLA ultimately comes from just a single PLA plastic manufacturer, who has a lock on it through whatever means (patents, trade secrets, whatever). So, I guess others can tweak it by whatever means, but the baseline material is going to be the same regardless.

In that sense, PETG should be much cheaper than it presently is, because the basic raw material starting point is pure commodity that lots of manufacturers make for all sorts of purposes. So, over time, I expect PETG may be priced quite a bit cheaper than PLA, for these reasons. In fact, it already does seem to be trending that way.

[Mossy]

I know I’m a bit late but I have pushed my X1C to it’s limit I believe. using a conversion kit for the hot end to third party nozzles and using a CHT nozzle with .6mm I can sling at a flow rate of ~32. I stay at 31 to be safe. I have my speeds at 300mms with acceleration of 10000mms on normal printing and 5000mms on the outer walls. I also to 500 on first layer and 2000 on top as default. There’s no loss of quality per my testing and the cooling seems to be able to keep up. I plan on tweaking this some more and seeing what I can push without quality loss but the flow rate is going to be my limiting factor.

[NeverDie]

Which filament?

I actually went back to the 0.4mm nozzle because I wasn’t seeing much benefit at 0.6mm, and it just seems all around easier to get quality with a 0.4mm nozzle with very little effort. If you posted your entire print profile, it would undoubtedly help anyone who wants to push the limits, as you are.

I’ll probably take another run at it in the future, but even just running a 0.6mm nozzle makes you an outlier compared to everyone else and the default profiles definitely don’t seem to work as well at 0.6mm as they do at 0.4mm. Hopefully some day that will change. I guess with a CHT nozzle, you are more or less on your own unless you get lucky and find a complete profile that someone has posted somewhere.

[Mossy]

Heres an imgur dump of my profile.
imgurDOTcom/a/6mmoSOh

Damn restrictions on links and images.

Heres my start Gcode:
; filament start gcode
{if (bed_temperature[current_extruder] >55)||(bed_temperature_initial_layer[current_extruder] >55)}M106 P3 S200
{elsif(bed_temperature[current_extruder] >50)||(bed_temperature_initial_layer[current_extruder] >50)}M106 P3 S150
{elsif(bed_temperature[current_extruder] >45)||(bed_temperature_initial_layer[current_extruder] >45)}M106 P3 S50
{endif}

{if activate_air_filtration[current_extruder] && support_air_filtration}
M106 P3 S{during_print_exhaust_fan_speed_num[current_extruder]}
{endif}

And my End Gcode:
; filament end gcode
M106 P3 S0

Everything is built off a modified .6mm default profile so anything not listed can be assumed default for the profile. I should note that I also use a WhamBam for my plate as opposed to the magnetic sheet that comes with. Never have had a bed adhesion issue! My next modification is going to be moving to X1Plus!

Feel free to reach out with questions or respond to this. I will keep an eye on my email for notifications! Hope this helps some folks!

[Mossy]

I forgot to mention I use Matter Hackers Build PLA almost exclusively. I order cases of it for work.

[NeverDie]

@Mossy

I use Matter Hackers Build PLA almost exclusively.

Checking their datasheet, they recommend a print speed of 40-100+mm/sec.

It’s controversial, but some filament manufacturers are claiming up to 500mm/sec print speeds for their particular flavor of PLA. It’s controversial mainly because you don’t know whether it’s pure marketing BS with nothing behind it, or whether there really is something that makes it different. To break that inherent uncertainty, I’ve decided to take their claims at face value so that if it doesn’t perform well at the advertised speeds, back it goes to the vendor it came from. I don’t have enough data yet to make any firm recommendations, but I am noticing what seems like tangible differences that tend to lean in favor the filaments which have the higher advertised speeds. It would be more objective to do the assessments double blind, to rule out bias, but I’m not staffed for that.

That said, the filaments rated for lower speeds are generally cheaper, and you can print almost any filament well if you’re willing to go slow enough. In that sense, there is a real “time is money” valuation that seems to be reflected in the pricing.

[Mossy]

I have found that recommendation and practice tend to very greatly. For instance I have only had one failure at this speed and with this brand. That being between my last comment and now but I was running some silk PLA as opposed to standard and there was some buildup. Cleaned it out and running my standard PLA again. I also would love to test this. I have the resources and time so might so that in a couple weeks when I’m back from vacation.

[NeverDie]

The people who run print farms probably already have done a thorough analysis of this topic, but unfortunately they don’t talk much. It maybe requires knowing an insider to get the real scoop, unless, as you say, you have the resources to confidently figure it out on your own.

It’s a bit like google and hard drives. They had large troves of reliability data as to which hard drives performed well over time and which did not. They decided to publish the info, and then suddenly you could immediately see who the quality manufacturers were and who were only just posing as such.

[Ken-N-Texas]

@Mossy

There’s no loss of quality per my testing and the cooling seems to be able to keep up. I plan on tweaking this some more and seeing what I can push without quality loss but the flow rate is going to be my limiting factor.

Where can I see the pictures? I learned very early in my 3D experience that the term “quality” is used to describe everything from a rat’s nest to a diamond in my opinion. One’s fine art is another’s rim-shot to the basket. “No loss of quality” really doesn’t mean anything if the quality was questionable to begin with.

[GenericUser]

@Mossy

M106 P3

I’m at a loss to the value of the cooling provided by the Chamber fan? P1 and P2 (parts/aux), are the ones that provide the cooling and think are controlled better in the other tab.

[Mossy]

Give me a few I am piecing together my Enterprise 1701-D atm and ill snap some pics for you! Printed with the profile listed!

[Mossy]

Frankly this is default but with the increased speed some extra cooling on each layer will help as hot end fans don’t get much time on each layer to actually cool it. You can see the difference it can make if you print a benchy at high speed while the bow is both facing toward and away from the fan.

[Mossy]

Thankfully I happen to be in a cost analysis stage of setting up a print farm as we potentially expand our offerings so this sort of thing is honestly something I really need to do. I have compared ESUN PLA and HatchBox PLA as well as Bambu PLA to MatterHackers Build PLA. I have some rolls left of each ill throw in the dryer and get some numbers on those at least.