I cut open my very first hotend for this one.
So this is what the iTopie i3 looks like now -- you’ll notice that Luke decided to go with a modern penta-color paint job with the classic white and black main elements, metallic silver and gold on the feet to match brass and steel details on the printer’s mechanics, and they all nicely complement the bright red printed parts. So as you can tell, it’s going to be a very classy printer once he actually gets around to finishing it. Today though, i want to specifically focus on one element, which isn’t even in here yet: The hotend.
Hi everyone, Tom here, and picking a hotend that is right for your idea of the printer you want to build can be somewhat daunting with so many choices out there. But once you start breaking them down, there really are only a few factors that define what type of hotend you’re dealing with and what sort of performance you should expect from it.
Now, the biggest factor and the one you’re going to see right away because it’s going to be shoved in your face is “all metal” or PTFE-based. Let’s start with the very first hotend i used, this thing, not an all-metal design, obviously, it’s got a mount made from PEEK, which is a super-strong plastic that doesn’t melt until, like 300°C, so way beyond what you’d typically use with 3D printing filaments. And the PEEK provides both the insulation between the heater block and the mount as well as the mechanical strength, after all, there is an extruder pushing filament into this thing. You can see that the heater block down here, where the filament melts, has pretty much a direct thermal path up to the PEEK part. But because PEEK isn’t really slippery, the hotend would constantly jam if it didn’t have a PTFE aka Teflon insert in here, and while you often still see PTFE inserts with newer all-metal hotends, this one reaches down all the way into the heater block. All-metal hotends, like the E3D v6, the recently reviewed DyzeEND-X or many other hotends, either have the PTFE liner stop somewhere in their heatsink or don’t use a liner at all. Now, the consequences from using a PTFE liner are diverse, the most obvious one being that you can use all-metal, or, more precisely PTFE-free hotends up to a higher maximum temperature. Reason being, PTFE starts to really soften beyond 250, 260°C and somewhere in that range, also start releasing gaseous neurotoxins -- which is the same reason why you shouldn’t leave Teflon-coated pans on the stove for too long without anything in them. So while all all-metal hotends easily go up to 300°C and some even further, the safe usable temperature range for hotends that use a PTFE-liner or insulator for anything that’s touching the heater block ends at those 260°C max. As far as robustness and reliability goes, you could argue both ways -- all-metal hotends use a heat break that is machined to have a very thin stainless steel tube section as an insulator, and naturally, that section is fairly easy to bend and break and at the same time, is pretty tricky to manufacture with a smooth inside bore to prevent filaments like PLA from sticking to the hot metal sides. On the other hand, the PTFE liner can leak if assembled with too little pressure and, worst case, kink and block the filament path if assembled with too much pressure and used with too high of a temperature, but they tend to provide an easier filament pushing experience for the extruder, which can be particularly helpful with flexible filaments. You can also get hybrid types, like the E3D lite6, Printrbot Ubis 13s or the soon-to-be-reviewed Flexion hotend, those still have a PTFE liner all the way into the melt zone but also use stainless steel as a structural and insulating element, cutting out the cost for a large chunk of PEEK as an insulator or for a precisely machined heat break.
Both PTFE-based and all-metal hotends can work extremely reliably with PLA, PET and many other not-so demanding plastics, but for regular use with ABS or even more extreme plastics, you should definitely go with an all-metal hotend.
Ok, so that’s been a lot of talk on one subject, let’s look into what else i’d look for in a hotend. And let’s start out with the overall geometry -- how big is it, will it fit your particular printer and extruder? Is it too short or too long? All very simple things, but easily overlooked. So the de facto standard for hotend mounting, is this 16mm groovemount. While many hotends might look similar or compatible, there are minor differences between them that keep them from being universally cross compatible -- not even the E3D v5 and v6 use the exact same mounting geometry. And really, for that, there’s not much else to say than “in case of doubt, check with the manufacturer”, there are simply too many different extruder mounts and hotends out there. Next up, length and girth. Now, a long hotend is usually not going to hurt anything, at worst, you might lose a few millimeters of vertical build space, but a hotend that is too short can cause issues in two spots, other than self-confidence -- one, you might end up bottoming out your Z-drive, and two, at the actual carriage that, well, carries, the hotend, there’s typically going to be bearings and sensors and what not sticking out the bottom there, so make sure the hotend is actually the lowest part out of those. Then, the circumference of the entire hotend assembly. This is the size of a classic PEEK-based hotend like the original Ubis or the Jhead or whatever, and this is what a modern all-metal one looks like, so, obviously, if the printer you want to build only has space for a super slim one, then obviously, buying one that is this big isn’t really the smartest thing to do. And keep in mind you might need to fit things like a bed sensor and part cooling fan around the hotend as well. Though, i’ve got to say, newer printer designs typically have the space to fit even the larger hotend types.
Ok, so one more thing about geometry, there’s basically three zones in any hotend, and ones that have these zones clearly separated tend to have the most predictable performance. There’s the cold end, the transition zone and the melt zone, and the mistake some manufacturers have made with all-metal types was to not properly define the transition zone, so depending on what print settings you used, the filament would start to soften at an arbitrary point somewhere above the melt zone, and cause some whacky behaviour. This isn’t as crucial in PTFE-lined hotends, as the liner will insulate the filament and compensate a lot for this sort of wonky behavior, but for all-metal types, having a sharp transition from the cold zone to the transition zone to the hot zone is pretty crucial. So the cold end will typically not be much warmer than ambient if you have a fan blowing over it, as most hotends do these days. It’s not really crucial how exactly the cold end looks as long as it’s doing its job of cooling well enough, so let’s move on to the transition zone, which is the weakest, but also the most important part of any all-metal hotend. Essentially, a longer transition zone leaves more plastic in a semi-molten sticky gooey state, so ideally you want that zone to be as short as possible. Most hotends get this about right, but be careful with ones that seem to have a particularly long transition zone. On hotends with a PTFE liner, there often isn’t a dedicated transition zone per se, which is ok for them. And lastly, the melt zone, this one will be at a consistent temperature throughout, and, basically, the longer it is, the faster you’ll be able to print with a given set temperature. Which is why the E3D Volcano is so ridiculously long. But with a longer melt zone you do lose some precision and increase stringing and blobbing, while a shorter than usual melt zone will need higher temperatures to print since the plastic has less time to heat up in there.
One more very important point is serviceability- sooner or later, you’re going to have a jam or break something on the hotend. Now, if you’ve got one that has like a one-piece nozzle heat break connector thing, there’s no way you’ll clean that in any decent amount of time, and if you have to replace it for example should you bend it, you’ll obviously have to replace the entire thing. And while we’re at nozzles and the likes, the standard nozzle size today is 0.4mm, which works for a lot of different applications, if you want slightly faster build speeds and are ok with slightly thicker layers, for example for a larger printer, 0.5mm or 0.6mm are still very flexible sizes, and if you want to go the opposite way and get maximum precision and detail resolution out of your printer, 0.3mm would still be an acceptable choice. I wouldn’t recommend 0.2mm unless you really want to live on the bleeding edge, since you’ll need to use insanely low layer heights and speeds to even profit from that kind of nozzle size and you’ll be at constant risk of jamming your nozzle with even the tiniest particles. But if you get a hotend that you can disassemble, you can always just swap in and try out a different size if you want to.
And one last note, especially with the hotend, please do yourself a favor and buy genuine parts if you’re building your first printer or don’t have the engineering or machining knowledge or simply not the time to figure out how to make a clone work. I mean, they all look great in the pictures, but there are so many things a Chinese manufacturer will be able to mess up when he’s only machining these by some incomplete drawings and pictures he found online, some of the sellers even claim to have improved the original designs, but it’s more like “well downright cloning didn’t work, so here’s something we huddled up to make it somewhat functional” when clearly, the ones manufacturing these don’t even know how the parts should properly fit together. Look at this, this is completely unusable. Yes, i only paid 4$ for this, but then again, you get what you pay for, and this sort of stuff isn’t what i would want to deal with when getting a printer up and running.
So in general, the big names when it comes to hotends like E3D, the UBIS used in the Printrbots, the Hexagon as used in Lulzbot’s printers, those are good all-round choices. But if you want to go for something completely different you should now have a some guidelines of what to look for.
Anyways, thanks for watching, leave me a thumbs up if you learned something, get subscribed if you want to learn even more in the future or maybe even consider directly supporting this channel on Patreon. And that’s about it for today, see you in the next one!
“Gymnopedie No. 1” Kevin MacLeod (incompetech.com)
Licensed under Creative Commons: By Attribution 3.0 License
Eric Satie -- Gymnopedie No.1 (Baptiste Remix)
Creative Commons Attribution 4.0 International License
You can support me without spending a single penny!