A bi-metal heat break, consisting of copper
threads and surgical stainless steel tubing is a pretty interesting concept that can potentially
add quite some performance to your 3D printer. Let me show you how I was able to more than
double the extrusion capability of my Ender-3 what happened when I first cheaped out and
bought a knock-off part and how I was pleasantly surprised when I spent the money and got the
genuine one from SliceEngineering! Let’s find out more! Guten Tag everybody, I’m
Stefan and welcome to CNC Kitchen. This video was supported by Squarespace! Easily
create your own professional website and support the channel at the same time by starting a
free trial today at squarespace.com/CNCKITCHEN. The bi-metal heat break is the next evolution
of the regular heat breaks that we know from our 3D printers. Let me first tell you what
makes this multi-material assembly unique; let’s benchmark the ones I bought for my
Ender-3 with extrusion tests and test prints and finally compare the standard Bowden solution
to the $10 knock-off and $30 genuine part. Let me know down in the comments if you’ve
ever used one of these heat breaks and if you’d rather do this upgrade instead of
changing the whole hotend! So the heat break fits in-between the hotend
and the cold end and is used to have a very sharp temperature gradient between the molten
material and the filament that’s fed to the nozzle. This part is important because
if you don’t have this sharp gradient, the filament tends to get soft way before the
heating zone and eventually blocks the filament path. That’s what we call heatcreep. In
order to accomplish that, all-metal hotend usually have a heat break with a very thin-walled
throat, where only a minimum amount of heat can travel through. Stainless steel is often
used for that application, or even more recent heat breaks use titanium, which features an
even small heat transfer coefficient but comes with other downsides. Bowden hotends have
a slightly different approach and use the even worse thermal conductivity of the PTFE
to isolate the hot- from the cold end, though this just on the side. The bi-metal heat break
improves on the all-metal heat break in a way that it is a multi-part and multi-material
construction. It features a thin metal tube that guides the filament. This has a couple
of advantages. Stainless steel tubes are available with very thin walls, thinner than you would
be comfortable when machining them. The thinner the wall, the less heat can be conducted and
the sharper the temperature transition. Also, due to their manufacturing process, they can
be stronger than regular stainless steel and the surface finish is also very smooth, which
reduces friction. Then, there is the upper part of the heat break, which is made from
copper. So all the heat that made it through the thin tube is transmitted very quickly
into the heatsink where it dissipates and therefore also helps avoiding heatcreep. Let’s
finally get to the lower threads, that are also made from copper for a particular reason.
These sit in the heaterblock and they help to increase the length of your melting zone
and will start heating your filament efficiently all the way from the start. Yes, with the
standard stainless steel, you’ll also have some material melted in that section, but
since it doesn’t conduct heat that well, it won’t be able to melt filament that efficiently,
especially at faster print speeds. The 3 parts are then probably heat shrunk together which
requires all of the parts to be of high precision and the right alloy. If they aren’t, there
is the risk that the press-fit decreases at higher temperatures and the copper sleeves
come loose, which I heard happening with some non-genuine parts.
Speaking of non-genuine. I know bi-metal heat breaks from SliceEngineering hotends, especially
their latest Copperhead. When I first heard about their claims about higher melting rates
in a size that’s equal to a V6 I was intrigued and wanted to try it out. SliceEngineering
doesn’t only offer these heat breaks for their Copperhead hotend but they also sell
them for other machines, like the hugely popular Ender series. One machine, I had serious problems
in that regard was my Ender-3 Pro, I used a lot last year for printing face shields.
The thing was, that my Prusa was more than double as fast in printing those parts. In
terms of number, I had to limit the melting rate of the Ender-3 to 5mmÂł/s, whereas the
Prusa was able to easily print at 12mmÂł/s before the filament started skipping or grinding.
That is, of course, partly due to the double sided Bondtech gears and the short and straight
filament path. So even though the material might not be properly melted, the Prusa was
able to just force the filament through the nozzle. Since the feedsystem of the Ender
is weak, as soon as the filament didn’t melt properly anymore, the backpressure increased
and the filament started skipping. To get an idea what those numbers mean, I
sliced a 3DBenchy for the Ender-3 at a layer height of 0.16mm. What you can see is, that
depending on the feature our volumetric flow-rates are around 2 to 4.5mmÂł/s. What I want to
say with that is, that if you’re only printing slow, at fine layer heights and with the standard
0.4mm nozzle, you’re usually not that concerned with melting rates. Though as soon as want
to go faster, thicker or wider, this becomes important.
To benchmark the change in extrusion capability, I ran an extrusion test in which I always
extruded the same amount of filament, just at different extrusion speeds and weighted
the resulting material spiral. If the filament skips or strips, the real amount of extruded
filament will be smaller, which means we under-extruded. I tested the whole range of extrusion rates
from a slow 1mmÂł/s up to 20mmÂł/s using a simple G-Code. The standard Bowden setup results
show that the faster we extrude, the less material is really pushed through the nozzle.
At 5mm³/s we’re already extruding only 93% and starting at 10mm³/s we really dip
downwards and can see a lot of grinding at the feeder gear. I performed this test with
a freshly re-seated Bowdentube and no gap between it and the nozzle. But like you many
know, if you don’t do anything against it, the Bowden tube will slowly walk upwards over
time, leaving you with a gap, that’s very bad for our printing performance. To simulate
that I intentionally moved the tube upwards by a mm and ran the extrusion test again,
which shows that the extrusion performance is even worse and the dip in the curve due
to grinding happens earlier at only 7mmÂł/s. Just on a quick side-note: If you have ever
wondered why you’re models have bulges in certain regions. Compare them with the flow
rate in your slicer. These are often at locations where the printer pints significantly slower
or faster than elsewhere. Therefore you are more or less under-extruding and that section
becomes thicker or thinner. One reason to keep flow as constant as possible or to make
sure that you’re only printing in a region where your extrusion graph is very shallow.
If you like these videos and investigations, make sure to subscribe and select the bell
to not miss any upcoming ones in the future! So I wanted to find out, if I can increase
the melting rate and therefore decrease the backpressure of the Ender with one of these
bimetal heat breaks. Since I was cheap and I had a hard time finding a part in Germany,
I purchased a knock-off bi-metal heat break for around 15 bucks from a German eBay seller.
Assembly was pretty straight forward. I removed the fan shoud, heated the nozzle, removed
the installed filament and unscrewed the nozzle. Then I removed the Bowden tube, removed the
two screws that additionally secure the heaterblock to the heatsink and loosened the small grub-screw
in the front of the heatsink to slide the heat break out. Since the lower threads were
quite gunked up, I had to use brute force to unscrew it from the heaterblock. I then
used an M6 tap to recut the threads and tho remove the carbonized plastic.
Installing the new bi-metal heat break was pretty straight forward. I screwed the nozzle
into the heaterblock, making sure that there still was a gap left for tightening and screwed
in the heat break from the other side. While installing it with plenty of thermal paste,
I already noticed the first strange thing because it was too short to bottom-out in
the heatsink, so I just pushed it in that it was flush with the heatsink and fixed it
in place with the grub screw. Installing the Bowden tube was then also quite a pain. Since
the distance between the hotend and heatsink got smaller I wasn’t able to add the retaining
screws and therefore had to really pay attention while hot-tightening the nozzle.
Then I reran the extrusion test and oh boy, especially at the higher extrusion rates,
this part helped to push so much more material out o the nozzle than the standard setup.
Only at around 15mmÂł/s the feeder started really badly grinding. That looked promising
with just a $15 part. Well, yes and no. Of course, I had to test its performance in a
real printing test, with retractions and everything and this is where the problems started. On
my 3DBenchys I got serious extrusion problems, and tuning retractions only helped slightly.
So there was obviously something wrong. This was the point where I contacted Dan from SliceEngineering
and openly told him that I bought a knock-off heat break and if they see the same problems
with their genuine parts. He assured me that they engineered their part especially for
less heatcreep, lower retractions and more flow. So I went out and figured out why in
my quick search, I didn’t find genuine BiMetallic heat breaks. It’s Copperhead heat break
and then I found the one for my Ender 3 as an option. Yei, local shipping from Germany!
Once it arrived, it was interesting to see, that the knock-off wasn’t a straight copy,
even though design files and drawings for the Copperhead are Open Source. The copper
sections are longer and SliceEngineering uses a greyish Tungsten Disulfide coating on their
part, probably for less friction on the inside and as an anti-seize on the threads.
Installation was similar to the knock-off. The only thing I noticed was that here, the
copper part on the cold-side seemed to be a bit too long and protruded out of the heatsink,
probably also not optimal. Anyways, I first performed the extrusion benchmark, which was
interestingly slightly worse than the knock-off but still a lot better than the standard setup.
The real exciting part was the print tests. I first calibrated the retraction length using
3DOptimizers retraction test and only needed around 2mm of retraction on a Bowden system,
which was quite impressive. The prints I was now able to produce also didn’t show any
signs of heat creep or jamming anymore and came out beautifully, just as promised! Using
this simple, new heat break seems to add quite a bit to the Ender. For once, it now allows
me to print way quicker. Then, it seems to help print quality by reducing retractions
to only 2mm and fixing the annoying problem with the Bowden tube wandering off. And then
there is temperature capability. The all-metal heat break now allows me to push temperatures
far above what I was comfortable doing with the Bowden setup in order to print ABS, Nylons
or Polycarbonate. I’m not 100% sure why the knock-off didn’t work that well, but
that’s probably a combination of dimensions, parts and the coating. As a small summary:
an all metal heat break can be better than a PTFE lined hotend because it requires less
maintenance and improves the temperature capability; but if not done well, can even make your printing
results worse due to more friction in the feeding section and potentially more heatcreep
if not done right. The Copperhead bi-metallic heat break improves on the all metal concept
with a higher melting rate due to the copper ends, perfect surface finish and an even sharper
temperature transition due to the thin, surgical tubing including the coating they use; But
it is also more fragile and costs a significant amount of money.
I’m being totally honest here: If you have a printer and you’re only processing PLA
at slow speeds, you won’t benefit a lot from this upgrade and might even break things
or generate problems on a printer that works. I can guarantee that 1 of 10 that are trying
to change the heat break will destroy either your heater wires or thermistor wires. But,
if you’re into bigger nozzles, faster printing or more challenging materials, the genuine
BiMetallic CopperHead heat break seems to be a really capable and simple upgrade, especially
if you currently run a Bowden setup! But let me know what you think!
If you by the way also want to test the extrusion capability of your 3D printer, then check
out the website article to this video where I added the G-Code file. This also brings
me today's video sponsor Squarespace that helps me creating this additional content
for you in such an easy way. Squarespace is one of the easiest to use, yet beautiful,
website development tools out there. People often ask me why I don’t just use Wordpress.
I did that in the past, but I always struggled to create a design I liked. Then I permanently
had to take care that my plugins were up to date, and when my blog got hacked, the hoster
didn’t have any recent backup so everything was gone. That will not be the case with Squarespaces
all-in-one platform where you just don’t have to deal with plugins, patches or updates,
ever! It’s just working and so easy to setup without any webdesign knowledge. If you can
use Word, you can create your own website on Squarespace! Give it a try and start a
free trial at squarespace.com/cnckitchen and use code CNCKITCHEN to get 10% off your first
purchase. If you still get stuck at any point Squarespace offers you great 24/7 support!
Try it out totally for free and let them know who sent you by browsing Squarespace.com/cnckichen.
Thank you Squarespace for sponsoring this video and make all of this possible.
Thanks for watching everyone I hope you’re all doing well! If you found this video helpful
than leave a like, share it with the community and make sure that you're subscribed for more.
If you want to support my work, head over to Patreon, become a YouTube member and take
a look at my merch! Stay healthy, auf wiedersehen and I hope to see you in the next one!
Neat
What hot end are people using for ender 3?
I got an ender 3 v2 and want to upgrade my hotend but not sure which route to go down. I see the microswiss might have compatibility issues with the v2. Then I see the microswiss with the direct drive and figure why not all at once. However, people then are saying I would need to get a second z axis screw for the other side to handle the weight of the direct drive.
So I'm kinda like, waiting. In some ways I wish I bought the regular ender 3 pro because there is just more info on the machine. The v2 in so many cases is just like the 3/pro but then it seems in very specific situations it's completely different.
Almost friend my board because creality sent me the wrong wire for my bltouch and now I wont even mess with installing it.
I got one of those trianglelab all-metal hotends which apparently can be paired with a stronger nozzle to enable nylon, but doesn't nylon also need an enclosure?