Automated method from CNC Kitchen to evalute the maximum volumetric speed.  

RE: Automated method from CNC Kitchen to evalute the maximum volumetric speed.

@dimprov - I'll add those notes to the data sheet. Thanks!

RE: Automated method from CNC Kitchen to evalute the maximum volumetric speed.

Reporting back:  I've moved up from a 0.8mm nozzle to using a 1.0mm nozzle, and the prints are coming out better because I can get the same line width but with better nozzle coverage.  I'd try 1.2mm and maybe even 1.5mm, but I've only found 1.2mm nozzles for the volcano, and I haven't found any 1.5mm nozzles at all.  With a 1.0mm nozzle for sure the dominating factor is MVS, and so I may move back to using a volcano.

RE: Automated method from CNC Kitchen to evalute the maximum volumetric speed.

Posted by: @dimprov

Reporting back:  I've moved up from a 0.8mm nozzle to using a 1.0mm nozzle, and the prints are coming out better because I can get the same line width but with better nozzle coverage.  I'd try 1.2mm and maybe even 1.5mm, but I've only found 1.2mm nozzles for the volcano, and I haven't found any 1.5mm nozzles at all.  With a 1.0mm nozzle for sure the dominating factor is MVS, and so I may move back to using a volcano.

I found some very crappy 1.2 and 1.5mm V6-style nozzles. They did not use the normal V6 M7 format, so I had to use pliers to mount them. Quality was nothing special and using the stock Prusa extruder, the speeds were so low as to not gain me anything. I have found some decent quality Volcano nozzles up to 1.2mm. I've dumped a bunch of info on the nozzles in the ongoing nozzle discussion. 

RE: Automated method from CNC Kitchen to evalute the maximum volumetric speed.

@bobstro

Yeah, without a high flow hot-end I found that although I could actually push more plastic (i.e. more mm^3/sec) without jamming by increasing the layer height and using a 1.0 nozzle, the extruded plastic wasn't getting hot enough for the thicker layers to properly fuse together, so it became counter-productive.  

RE: Automated method from CNC Kitchen to evalute the maximum volumetric speed.

That was with PLA by the way.  If it turns out that if, for some reason, more heat gets transfered to PETG than to PLA when passing through the same hot end, it might yet turn out to be worthwhile.  Or, maybe even if more heat isn't transferred with PETG but PETG layers just fuse together better, it might also turn out to be worthwhile.  I can imagine that such a thing could be true, but I haven't tried that yet.  

RE: Automated method from CNC Kitchen to evalute the maximum volumetric speed.

Anyhow, it does shed some light on the original OP test, because it shows that when approaching the limit, it's not just about underextrusion, it's also about how well the extruded plastic can be fused together and about whether the resulting layers are fused strong enough to be useful or just barely sticking together.

RE: Automated method from CNC Kitchen to evalute the maximum volumetric speed.

Probably with small nozzle diameters you hit the underextrusion limit first, and for larger diameters like 1mm+, you hit the thermal limit first.

RE: Automated method from CNC Kitchen to evalute the maximum volumetric speed.

Reporting back: Using the same apparatus as above, and the same gcode, and the same 1mm nozzle (in fact, everything the same but a higher hotend temperature of 230C instead of 210C), I found that the max volumetric flow rate of PETG was lower than for PLA.  Instead of 18mm^3, maximum flow topped out at more like 15mm^3.  At 18mm^3 I got very frequent stepper skipping.  It was Sainsmart Black PETG, 1.75mm +/- 0.02mm.

I had thought that with a 1mm nozzle I'd get a much higher maximum volumetric flow rate for PETG than what was actually observed.  Perhaps the difference in pigment was partly responsible?  I don't know, as it is but one datapoint.

RE: Automated method from CNC Kitchen to evalute the maximum volumetric speed.

Posted by: @dimprov

@bobstro

Is there any way to have PS vary the MVS values depending on the z-height?  It would be nice if I could create an MVS tower, similar in concept to a temperature tower except that the change in MVS values occurs as a function of z-height during slicing rather than run time.

The only way I see is to do a post-processing script to multiply speed values. But in my mind that would multiply all values, even the ones that are not capped by MVS. The test part may be designed with that in mind.

RE: Automated method from CNC Kitchen to evalute the maximum volumetric speed.

I was going to like several of the comments here, but what is the point really, you have to define what your doing, what your intention is, what are you printing, what materials?, TPU key Chains? or running a print farm? Nerf parts? what are you looking to do faster (Max) doesn't necessarily mean better, people get hung up on this, and we see so many problems because of it.

Stop it, get a grip, it's so specific, work with what you are doing and optimize that.

My comments are not pointed at anyone specifically, but the community as a whole, you are dealing with thermodynamics and kinetic energy, the point is to take a molten material and form it into a shape, the speed & accuracy at which this happens depends on the material and the shape (assuming vast experience) you will manage this, the only thing Stephan's method is offering is a bench mark, and your going to have to do it with every single filament that you use, and that number is going to change with every project.

I'm sure many are not going to understand my comments, but there they are.

Good luck,

Swiss_Cheese

RE: Automated method from CNC Kitchen to evalute the maximum volumetric speed.

Posted by: @swiss_cheese

I was going to like several of the comments here, but what is the point really, you have to define what your doing, what your intention is, what are you printing, what materials?, TPU key Chains? or running a print farm? Nerf parts? what are you looking to do faster (Max) doesn't necessarily mean better, people get hung up on this, and we see so many problems because of it.

This is definitely not a topic everybody needs to worry about. I'm not sure exactly what posts you're referring to, but just to clarify:

• Stefan's approach is focused on ensuring that your maximum volumetric "speed" (rate) (MVS) is set low enough not to affect the amount of filament extruded during the print. Avoiding under extrusion as I understand it. He evaluates this based on comparing print weights. His appears to be a quality-focused approach (IMO).
• My "free air" approach is only intended to identify the upper "red line" maximum that a particular combination of hotend, nozzle, filament, and settings can handle. This is all about avoiding hotend skips and jams, not quality. I find that determining a good MVS speed saves me a lot of tweaking speeds for particular prints. In practice, I stay well below the maximum. 

"Faster doesn't necessarily mean better" was actually my motivation for going down this rabbit hole. Somewhere on these forums is an old 2018 thread with a subject along the line of "how fast can you go?" in which users were earnestly trying to figure out how fast they could print with the Mk3. Over a period of weeks, it was clear that there was more to it than just turning up speeds, but I was frustrated at the lack of information explaining what governs the limits. Someone, fortunately, clued me in to the max volumetric rate calculation and I started looking for more information. Surprisingly, the only good info that I could find was deeply buried in old reprap forum posts. Outside of that, I only found vague mentions in E3D marketing literature and mysterious, unexplained values plugged into the Prusa presets. This was the spark for my "speed is an illusion" post that got my 3D printing notes kicked off.

Stop it, get a grip, it's so specific, work with what you are doing and optimize that.

I do agree, but for me, that's the point. I can't fully optimize without knowing what the upper end of my hotend capacity is. That's going to depend on hotend, nozzle, filament, and print settings. A few things are apparent after having worked with several combinations:

• With an E3D V6 hotend and 0.4mm nozzle, you don't need to know or care. Even using 120% of nozzle size for extrusion width and 80% for layer height, you can use 70mm/s speeds without overwhelming the hotend. This is at the upper end of quality for most current filaments. Where it does become an issue is with overly aggressive presets with settings like 200mm/s for infill. We see a lot of "mysterious hotend clicking" posts, although those have come down since Prusa started backing off on some of the preset values.
• This information is a lot more valuable for anybody using cheaper printers with "mystery brand" hotends. There is very little information provided by the manufacturers, and a lot of mythology and bad information out there for these users. I read a post yesterday by a guy convinced he was printing a small Benchy at 300mm/s.  With a hotend capacity of half the V6, those users hit the "clicking" wall much more quickly. No hotend has unlimited capacity, but that's a tough concept for anybody focused on "speed".
• When using larger nozzles, I find this information absolutely vital for maximizing printer output. This really hit home during the PPE printing push of 2020 when trying to maximize output printing frames. You are absolutely correct that "the speed & accuracy at which this happens depends on the material and the shape". I would add that "this only makes sense if you are trying to produce the same thing with the same setup with the same material repeatedly". In this instance, that is exactly what I was trying to do. Knowing what the "red line" hotend capacity was, I was able to tune print parameters to drop the production time of a frame on the Mk3 from 3 hours to under 1 hour. I was able to print 2 frames at a time on my Mk3, and share this information with others including some with multi-printer farms. While testing for a specific filament and nozzle combination took a few minutes, in a case where you want high repeatability, it's worth it. I have no final counts, but I know the group produced thousands of frames for local hospitals at the time, and cutting production times was important. Now, I use that same information to reduce print times for full-bed prints with big fat nozzles. Usually for exciting things like bird feeders. 

My comments are not pointed at anyone specifically, but the community as a whole, you are dealing with thermodynamics and kinetic energy, the point is to take a molten material and form it into a shape, the speed & accuracy at which this happens depends on the material and the shape (assuming vast experience) you will manage this, the only thing Stephan's method is offering is a bench mark, and your going to have to do it with every single filament that you use, and that number is going to change with every project.

Definitely true. I think, though, that it's important that users understand that there are no "magic" settings. There are reasons you can print at one speed with one configuration and totally different speeds with another. I was surprised to realize that PETG can print much more quickly than PLA, for example, having bought into the "slow and steady" mantra for PETG. For large, functional parts, knowing how fast I can reliably push filament through a large nozzle is worth spending such time. I cringe whenever I see someone requesting a "magic profile" for a printer and wish they'd instead learn why some settings affect a specific print.

I'm sure many are not going to understand my comments, but there they are.

It's always interesting to discuss these topics. If I hadn't wandered into the "how fast can you go?" thread, I might still be operating under a bunch of bad assumptions. I've certainly learned a lot from others here and discussion is always productive.

RE: Automated method from CNC Kitchen to evalute the maximum volumetric speed.

I want to make clear that my comments were in no way intended to criticize or quell this topic or anyone participating, I enjoy this topic. However it seemed to me to be coming across as it progresses as a "how fast can you go?".  Based on several recent assists that I was involved with, I felt compelled to make that post, as an informative gesture for newer or less experienced users reading this information and thinking " Oh I'll just do this and print it as fast as I can, using these numbers" these guys are doing it. (It's an advanced topic)

@bobstro

You have addressed my Comments and articulated my intentions better then I ever could,

Thank you,

Swiss_Cheese

RE: Automated method from CNC Kitchen to evalute the maximum volumetric speed.

Posted by: @swiss_cheese

I want to make clear that my comments were in no way intended to criticize or quell this topic or anyone participating, I enjoy this topic.

No worries, I always assume these exchanges are being made with a conversational intent unless things go sideways. We've got a lot of smart folks here and I've learned from many of them.

However it seemed to me to be coming across as it progresses as a "how fast can you go?".  Based on several recent assists that I was involved with, I felt compelled to make that post, as an informative gesture for newer or less experienced users reading this information and thinking " Oh I'll just do this and print it as fast as I can, using these numbers" these guys are doing it. (It's an advanced topic)

That's something I have tried to convey. I do want to update my notes to emphasize that it's a "red line" and not a "top reliable speed" adjustment. Another thing that is often not discussed is the filament itself. Most filament manufacturers clearly state that speeds should be kept with a relatively low range, and print strength can really suffer. That's why I prefer a larger nozzle moving more slowly.

One of the challenges in discussing these topics is that there are so many inter-related factors that it's easy to overlook something important, or assume it's been addressed.

RE: Automated method from CNC Kitchen to evalute the maximum volumetric speed.

@swiss_cheese

Let's say you have a Dragon HF with 0.6 nozzle and random PC. How reasonable amount of plastic you can push per second. I really have no clue. Then a measurement is really nice to decide which way to go.

RE: Automated method from CNC Kitchen to evalute the maximum volumetric speed.

Posted by: @ron

[...] Let's say you have a Dragon HF with 0.6 nozzle and random PC. How reasonable amount of plastic you can push per second. I really have no clue. Then a measurement is really nice to decide which way to go.

I'd expect performance to be comparable to using a Volcano hotend. Exact MVS will depend on your filament, but 70-90mm^3/s as an upper limit is certainly doable.for PETG. I've not tried PC. Expect to print at half maximum for quality.

It's easy to do a quick test for yourself. I've got a Dragon hotend, but am in no hurry to mount it.

Looking for that testing gcode v3 zip.

So the link for the v3 testing gcode is a bit broken, and I can't get it.
Anyone have a copy of it laying around?
I'd like to run this test on my volcano with 0.25, 0.4, 0.6 and 0.8mm nozzles. and like 5 different materials.
And then use the results to build a better model for calculating optimized 3D printing settings.

Attatched is a screenshot of the model, let me know if you want any numbers run through it for testing.
The data used to build the model is from an E3D blog post and that Dyze flow rate optimization post from earlier.

Thanks for the help in advance.

RE: Automated method from CNC Kitchen to evalute the maximum volumetric speed.

Hello!

When I try to download the .zip I get a 404, would you mind sharing it again?

RE: Automated method from CNC Kitchen to evalute the maximum volumetric speed.

I don't follow activities here too much...
Feel free to change values in blue cells in the joined spreadsheet to generate gcode (in yellow cells)
Since then I done a lot of extrusions but not on Prusa. My latests are for LGX + Dragon HF + Bondtech CHT (0.4 & 0.6).
One new tip I have is to make the first two extrusions the same flow rate now. As all underextrusion measurement are based on that first extrusion, you have to know how accurate it is.

RE: Automated method from CNC Kitchen to evalute the maximum volumetric speed.

PDF version
MaximumVolumetricSpeed-Simple

RE: Automated method from CNC Kitchen to evalute the maximum volumetric speed.

And the zip files

VolumetricSpeed-r04-8spots-VS02-16

Those gcode are for volumetric speed from 2mm3/s to 16mm3/s which is in the range of the E3Dv6. Now I do series 4mm3/s -> 46mm3/s.
With the XLSX file you can generate the serie you want.