Best model for measuring maximum volumetric flow rate?
Here's an example of what I mean:
I checked thingiverse, but found nothing.
I'll soon be switching hotends and nozzles, so I figure now is a good time to run some benchmarks so as to better compare their effect on maximum volumetric flow rate.
Anyone have any different models to print for that purpose and/or associated methods, or is the method and model shown above as good as it gets?
RE: Best model for measuring maximum volumetric flow rate?
P.S One you know the Max flow rate, then you can use it to drive prusa slicer and the world's your oyster: https://help.prusa3d.com/en/article/max-volumetric-speed_127176
RE: Best model for measuring maximum volumetric flow rate?
I've got notes here on the process I use. I don't use prints as the slicer adds too many variables. Straight free-air extrusion yields good results.
and miscellaneous other tech projects
He is intelligent, but not experienced. His pattern indicates two dimensional thinking. -- Spock in Star Trek: The Wrath of Khan
RE: Best model for measuring maximum volumetric flow rate?
A big flat model, a square . the plane will not be nice.
Then the prusa cannot print this, you need to adjust nozzle temp(higher) or lower speed.
RE: Best model for measuring maximum volumetric flow rate?
A big flat model, a square . the plane will not be nice.
Then the prusa cannot print this, you need to adjust nozzle temp(higher) or lower speed.
Understanding not. Elaborate nice or comprehend cannot.
RE: Best model for measuring maximum volumetric flow rate?
If you involve a slicer -- any slicer -- it's going to introduce features that may throw accurate measurements off. Volumetric rate involves the extruder and hotend. Ideally, calibrate using those and avoid introducing other factors like slicer "smarts". The best model is no model.
and miscellaneous other tech projects
He is intelligent, but not experienced. His pattern indicates two dimensional thinking. -- Spock in Star Trek: The Wrath of Khan
RE: Best model for measuring maximum volumetric flow rate?
If you involve a slicer -- any slicer -- it's going to introduce features that may throw accurate measurements off. Volumetric rate involves the extruder and hotend. Ideally, calibrate using those and avoid introducing other factors like slicer "smarts". The best model is no model.
However, doesn't normal printing (with the nozzle pressed up near the surface it's extruding against) introduce some amount of backpressure? So, air printing would be lacking that. i.e. if the measurements are made while air printing, then you'd get a somewhat loose upper-bound on maximum volume extrusion rate. I haven't tested that, but I'm guessing so. Maybe you have tested it and know differently?
If that guess is more right than wrong, perhaps here's an alternative that might be more accurate: how about printing a pure cylinder vase at constant speed and incrementing the extrusion width (as suggested by the youtuber in the OP) at different height triggers. However, by using a cylinder vase, unlike the youtuber's or devilhunter's models, there's no cornering, so there's nothing for the slicer to be "smart" about (well, at least not that I'm able to imagine). When you reach the layer interval where the extrusion rate starts to fail, you could then look at what the extrusion rate was in the slicer view, which is easy to do given the color coded extrusion rate view that prusa slicer kindly provides. The extrusion rate in the interval before that layer interval would be the maximum volume extrusion rate (to whatever extrusion rate resolution is being tested). [ Edit: Of course, this may only be accurate if the actual print time exactly matches the slicer's estimated print time. ]
Actually, I'm not sure whether vase mode would be needed or not. Since I almost never use that mode in my printing, I'd probably rather not use it for the testing. For a cylinder shell I'm not sure it would make any difference anyway.
RE: Best model for measuring maximum volumetric flow rate?
Printing in air does not mean without a nozzle. You're just extruding through the entire hotend -- including the nozzle -- as fast as possible.
The problem with using a sliced print is that every modern slicer uses settings like acceleration and jerk to tweak speeds based on model geometry. Add in coasting, wiping, linear advance and other settings, not to mention slicer math under the hood, and you're measuring something else.
Not to mention that a print will probably take a LOT longer.
No matter what method you're using, you are only identifying hotend capacity on that configuration. Actual usable printing will likely require much lower rates as you move up with larger nozzles. It's just good to know what the hotend can handle.
and miscellaneous other tech projects
He is intelligent, but not experienced. His pattern indicates two dimensional thinking. -- Spock in Star Trek: The Wrath of Khan
RE: Best model for measuring maximum volumetric flow rate?
@bobstro
It would be easy to get a ballpark number and de-rate it by half to have plenty of cushion for margin for error, but plainly that would run contrary to, say, running as fast as possible without sacrificing quality. So, just to confirm: are we in agreement that the goal is to find the best possible MVS plug-number for Prusa Slicer?
For the benefit of anyone else who might be reading this: In a perfect world there would be PS presets for every possible configuration (nozzle, hotend, extruder, etc) that would maximize speed without sacrificing quality, but Prusa has said that even the PS default presets that we are given for the stock Prusa i3 MK3/S/+ are slower than what Prusa himself uses on the Prusa printfarm. I guess that's because of allowances for how well people build and/or maintain their machines plus maybe manufacturing tolerances in the first place. It would be nice to know what presets the Prusa printfarm uses so that we might know what a stock machine can realistically aspire to, but we're here because we don't know that, as well as because some of us may have upgrades (like matchless nozzles or Phaetus Dragons) which can do even better, and then the question is: exactly how much better can an upgraded machine do? The answer to that would be more useful that what 99.9% of product reviews offer, because most reviews only hand-wave and never quantify the true performance benefits, which leaves you to wonder whether or not any particular upgrade or constellation of upgrades is worth the cost and effort.
RE: Best model for measuring maximum volumetric flow rate?
@bobstro
I received the high-flow phaetus dragon today, and the matchless nozzles prior, so I'll be upgrading to them sometime soon. Before I do though I'll try measuring my as-is stock MK3 using your method and maybe some other method (for comparison) to have a baseline. Then I'll repeat the measurements after upgrading to the high-flow phaetus dragon and the 0.8mm matchless nozzle. I'll report back afterward and we can compare notes and print profiles. I can typically run stock i3 MK3 with a 0.6mm nozzle at the "200% speed" without incident using real-time tune, but 300% sometimes doesn't print as well. At a minimum, I'd expect the upgrades to do better than the "200% speed" factor, not counting the added benefits that accrue from increasing nozzle diameter from 0.6mm to 0.8mm.
I'll see how that goes before deciding on a super volcano. Another option would be running two maxiwatt hotends in series by linking them with a regular volcano nozzle. If done that way then up to 80watts can be channeled into the hotend, which is what some of the amazon review writers say they have done. Because it's a toroid, whether just single or doubled up, it might (theoretically) be the best hotend geometry for uniformity.
Anyone else is also invited to join in. The more the merrier, as maybe then we can compare different hotend and nozzle upgrades and thereby potentially learn more.
RE: Best model for measuring maximum volumetric flow rate?
[...] I received the high-flow phaetus dragon today, and the matchless nozzles prior, so I'll be upgrading to them sometime soon. Before I do though I'll try measuring my as-is stock MK3 using your method and maybe some other method (for comparison) to have a baseline.
That's one thing I wish I'd been more consistent with. It's always to gather baseline data before you start making a bunch of mods. My Mk3 is "mostly stock", but I've upgraded the nozzle, heater block, and heatbreak. I'm really torn on swapping in the TriangleLab Dragon because it's been working so well.
Just a note on max volumetric rates in general. With a stock 0.4mm nozzle mounted and using 120% of the nozzle size for extrusion width, you have to run at 120mm/s at 0.2mm layer heights or 74mm/s at 0.32mm (max recommended) layer heights before you hit the conservative 11.5mm^3/s limit of the stock V6. Many users can print along merrily and never come near these limits. The reason I pushed into the topic was realizing that pushing 200mm/s for infill using Prusa's original 0.2mm profile was overly optimistic, and their 15mm^3/s MVS setting was a bit high for reliability. They've since moderated their speed settings a bit.
The beauty of using MVS in PrusaSlicer is that it frees you from worrying about too much speed. Set up print profiles using whatever speeds you need for strength and finish. Then let PrusaSlicer use the MVS value to throttle speeds when and only if you push too hard. This is PrusaSlicer's big secret weapon compared to other slicers. On those, you have to spend time reverse engineering safe values based on your extrusion width and layer height for every nozzle combination. MVS is "set and forget" once you know the upper limit of what you can push with a particular filament and nozzle.
That said, it's a theoretical limit for the most part. Your printer will extrude the filament at that rate, but most of the time, you will want to stick to much lower print speeds to get good finish and adhesion. Regardless of whatever method you decide to use, absolute precision is not needed and is probably misleading. The main takeaway for me was that PLA tends to have a lower maximum flow rate closer 20mm^3/s regardless of nozzle and hotend, while PETG can be pushed at faster rates with some added heat... provided you're not after good finish and can take some stringing.
and miscellaneous other tech projects
He is intelligent, but not experienced. His pattern indicates two dimensional thinking. -- Spock in Star Trek: The Wrath of Khan
RE: Best model for measuring maximum volumetric flow rate?
@bobstro
I received the high-flow phaetus dragon today, and the matchless nozzles prior, so I'll be upgrading to them sometime soon. Before I do though I'll try measuring my as-is stock MK3 using your method and maybe some other method (for comparison) to have a baseline. Then I'll repeat the measurements after upgrading to the high-flow phaetus dragon and the 0.8mm matchless nozzle. I'll report back afterward and we can compare notes and print profiles. I can typically run stock i3 MK3 with a 0.6mm nozzle at the "200% speed" without incident using real-time tune, but 300% sometimes doesn't print as well. At a minimum, I'd expect the upgrades to do better than the "200% speed" factor, not counting the added benefits that accrue from increasing nozzle diameter from 0.6mm to 0.8mm.
I'll see how that goes before deciding on a super volcano. Another option would be running two maxiwatt hotends in series by linking them with a regular volcano nozzle. If done that way then up to 80watts can be channeled into the hotend, which is what some of the amazon review writers say they have done. Because it's a toroid, whether just single or doubled up, it might (theoretically) be the best hotend geometry for uniformity.
Anyone else is also invited to join in. The more the merrier, as maybe then we can compare different hotend and nozzle upgrades and thereby potentially learn more.
Hi! How are you? Did you calculated the MVS for the Phaetus Dragon High Flow?