A lot of people want to print faster. Maybe they don’t like to wait, or they need to print a lot of things. Maybe it is just human nature to want to push things to go faster. The problem is, if you move filament too fast it may not have time to melt inside the hot end. To combat that, some people install a “volcano” — a larger heat block that takes a special longer nozzle. The melt zone is longer so there is more time for the filament to liquefy before shooting out of the nozzle. This is also a problem if you are using a very large nozzle size. But what if you don’t have one of these special hot ends? According to [Stefan], you can use a normal hotend with a volcano-style nozzle just by adding some common nuts. You can see the explanation in the video, below.
The idea came from a few commercial offerings that allow converting between different-sized melt zones. Some of these use the same idea. But, if you are familiar with [Stefan’s] videos, you know he tested the results thoroughly. The tests reveal that a standard V6-style hotend can handle rates of just over 9.8mm/second. with a 0.4mm nozzle at 210C and is usable beyond that. A true volcano hotend. starts deviating from the ideal at about 15mm/second and, also, is usable at even higher rates. But what about just using a long nozzle in a regular block with or without the nuts?
You should watch the video, but spoiler alert: with or without the nuts, it worked quite well. In fact, it worked slightly better than the real volcano. There is a lot of discussion about what makes your extrusion system work — or not work. We always enjoy his analysis and methodical testing.
Towards the end, there is some data about how part cooling does impact the results and what you might do about it. Increasing temperature can also improve melt performance. The difference in performance isn’t enough that we’d suggest you take off a volcano hot end for this method. However, if you were wanting to add a volcano-style hotend, this is a great deal. You only need the nozzle and some nuts. Even better, you can easily switch back and forth between conventional and high-flow setups. After all, you don’t always want the long melt zone since it can contribute to undesirable things like oozing.
If you’ve avoided larger nozzles because you fear losing detail, newer slicers might make you rethink that decision. When you watch the video, watch for an appearance of [Stefan’s] special high-flow nozzles.
For the editor: mm/s != mm^2/s (I’m glad we’re finally measuring performance in mm^2/sec)
Wouldn’t actual performance be volume? I’m sure the video uses cubic millimeter per second at 10:35.
Yes, volume good. But the article from HaD mentions speed instead.
While your original claim “mm/s != mm^2/s” is correct, did you really mean “mm/s != mm^3/s”?
(At least I believe that’s what Frankel is asking.)
Since filament diameter is 1.75mm standard and slicing is done in terms of linear filament speed not volume, filament mm/s is a bit more useful IMO.
Slicers and printer firmwares can work in either linear or volumetric mode. Volumetric is better because you can calibrate the filament diameter in the printer settings (“1.75mm” is only the nominal diameter) and not need to reslice the model.
Everyone talks about nozzles and hotends, but I hardly ever see anyone talk about the accuracy of the thermistor, how it is measured, or the heating element itself. I found much better results with a small amount of boron nitride for the thermal interface to allow more effective heat transfer. Even using a mosquito magnum or some other monster hotend still requires such attention to detail in order to be as effective as possible. Though I rarely push it to its upper limits, there is still a noticable difference in print quality at higher speeds just from this sort of cheap and simple application of thermal interface material.
Deeze nuts!
Somebody had to say it.
Suggested responsorial: Nobody had to say it.
You can get a basic Ender for $190. What is your time worth? Just get a second printer and save on bad prints from pushing the extruders envelope.
This is Hack-a-day, Buy-a-day.
Nobody gets an FDM 3D printer to save time. It is either because it is cheap or because the hobby is interesting.
Edit: This is Hack-a-day, not Buy-a-day.
You do you.
Hacks are time consuming and results iffy. Nobody has time to hack everything.
How many of you still overclock? Why or why not?
At some point you will value stability and predictability more than bragging rights.
50% of hackaday posts are just promos for some lame youtube channel. Then there was the tire wrapped in twine table.
I overclocked my old 2600k i7 because it was easy and directly improved single thread performance. I overclock my DDR4 RAM because it’s easy and makes me feel good, whether or not it makes any workloads practically faster.
“Nobody gets an FDM 3D printer to save time.”
Not true. Plenty of people are interested making things rather than working on cheap or home made 3D printers. My first 3D printer was bought because it was cheap, my second 3D printer was bought because it was reliable (and I didn’t have to spend more time on tuning/repairing it than making things).
Out of curiosity, which one did you get?
Prusa i3 MK3 I got it in early 2018. Since then I spent maybe about 1 hour total on repairs/maintenance, replacing a fan and upgrading the firmware. I could upgrade it to the MK3S, but it works fine as is. I’ve got other 3D printers if I want stuff to futz with, but most of the time I just want to design and make things without going down a 3D printer rabbit hole.
My first printer was the Monoprice Maker Select V2. A fun intro, but got fed up of fixing it, upgrading it to try and get better prints, and leveling the bed all the time.
No regrets.
At work, we have at least half a dozen Markforged Onyx printers cranking out prototype parts. The filament is obscenely expensive for a hobbiest, but it’s a good deal for the business vs. paying for machining.
Exactly.
It’s a tool. It works most reliably when not pushing the envelope.
We do get the benefit of all the ‘hacked up’ knowledge that is coded into the slicer. Not against hacking your printer, but not what I’m doing. As you say prototyping (perhaps a hack).
Most slicing isn’t done in terms of linear filament speed either. It’s done in relation to the speed of the traveling nozzle itself. Slicers generally have both limits for mm^3 and mm/s. But volume is the true measure of a hot ends capabilities, so mm^3 makes more sense; as if anyone compares 2.85mm filament in the future, volume is the only true metric at which it can be compared.