MakerBot Print is our newest print-prepration software, which supports native CAD files and STL assemblies,
allows you to interact with all your printers via the Cloud, and many other exciting new features.

Download Now


CR-10/10S/Ender3 FANG for 5015 Fan - Smooth Design

by amass3d Sep 10, 2018
Download All Files

Thing Apps Enabled

Please Login to Comment

Hello! Would it be possible to get the cad file? I would like to modify a little bit? Thx!

Hello! i saw a lot of remix of you fang duct, and i do not understand how they could modify, did you share the cad file with them? if yes would you share with me also? Thx!

I don't have a CAD file. I modeled this using traditional polymodeling with subdivision.
The models are very high density and my guess is that they used Meshmixer to splice things together.


Anyway, if you mind to send or share the original file would be great!


Is there / has anyone, modified the BL Touch mount for this? The regular one doesn’t fit because the body of the cooler is fatter than the original.

any change of tweaking it for dual 5015 fans one per duct and bl touch

Hi Everyone! Id love to make a 40 + 50m fan version of this but ..and this baffles me...im having a hard time converting the mesh properly and get it down to a lower number or polygons........still....baffled. Anyways...i dont want to ask the original creator for his files as that would not be appropriate to ask him to share his original creation. But if any of you remixers have been able to optimize this model somewhat please send me a PM...thanks fellas ;)

Hi there! I really like this fang and I wondered if I could trouble you for the CAD file for it please? Or isn't that the done thing? I won't be offended if you say no. Did you design it in Fusion360?

i sent you a PM amass3d

Hey, I didn't design it in a CAD software, it was done through taditional polygon modeling :)

Excuse my ignorance, but what is ‘polygon modeling’? Sounds like I’m missing a trick here

It's just a general 3D modeling method that can be done in software like Blender, Cinema4D, Autodesk 3Ds Max etc... You can find good explanations for that if you search the web. :)

Comments deleted.

very good part. I'm looking also for the back white extruder fan holder but I could'nt find it. Can you help me?
I found a CR-2 holder, it works but it does'nt fit good.

I'm sorry, I don't understand what you mean. The rest of the assembly for the CR-10 should be all in ozfunghi's files. I listed in the description which files you need from his pack. What do you mean with CR-2?

It s a great work !
Is it possible to make a version in order to use 5020 blower ?

Surely but I don't see what for. A 5015 fan is already more than strong enough for most prints with this design. You would only introduce more unneeded weight to the print head resulting in ghosting. If more people request it I could put in the time to make it.

i usualy use 5020 because they are stronger, quieter and for 24v version the pwm control is really better, more smooth.
the difference in weight is very low, a few grams.
and they are so much more powerful ! i think it s a must hav :)

I see but you really don't need that much power if the fang offers efficient air flow. I have my 5015 running at 60-70% in most cases, because it's sufficient enough and even quiter.

It's been over 200 hours that I use a 5020 instead of a 5015 and I will not return back, it's a real gain. You should try instead to try to convince me to stop using it :)
And I think that combined with your design would be even more effective in addition to being an interesting technical achievement.

Noone is trying to convince you to stop doing anything. If anything it's you who is trying to convince me to do work I don't actually need to do. The design as it is already works perfectly for me, much so that I've been reducing the fan speed. It's one thing to assume, another to actually test. I've done countless tests of stringing, bridging and overhangs at various fan speeds with my setup and I've reached a point where there is just no improvement beyond 70% speed anymore. Increasing the volume of air being pushed through the fang can actually even worsen the performance for bridging because too high air pressure can distort long bridge extrusions.

If you are happy with your solution, then it's awesome and you don't need my fang design in the first place :)

Hello, thank you for taking the time to answer in details.
The design I use today has other small problems that I am trying to correct, and your design in addition to being smart seems perfect.
Currently I can make bridges 180 - 200mm, it's really interesting, but I will try your design with a good quality 5015 fan (I just ordered it), the 5015 that I had tested before were not really great.
I will post the result as well as my adaptation work for the Tevo Tornado & a BL TOUCH.
Best regards

Hi there! Did you make a modified mount for the BL Touch please?

Hey, i added a bl touch onto the fang mount now, i have vibration problems any soulutions?

I'm not using a bl touch so I can't really help. My guess is that maybe the whole extruder head is now too heavy and depending on your jerk settings it can't handle faster movements.

can u make this to fit the cr-10s PRO by chance?

Sorry I don't own one and I only made the fang, not the mount plate. If anything I suggest asking ozfunghi for a new mounting plate for the pro model.
Or you simply model one yourself using the mount model for the normal CR-10S as reference.

This is what I came up with. Any thoughts would be nice.

Ender 3: Smooth Fang Mount with 40mm duct and EZABL

Hey man, great! Looks awesome. Unfortunately I don't own an Ender 3 and can't test it but it looks proper.

The only thing I see as critical is this area because it has an unsupported top. You could easiy fix it by removing just the top little part though (see painted line). Other than that I don't see any problems. I love that you simpled down the assemby itself.


Ah and also, the most simple mount plate doesn't have chamfers/fillets on the fang holder compared to the other versions. I don't think it's that important I think but just wanted to point out that it's different to the outher in that ragard.

I will add your design to my updates in the desciption so those interested in adding it to Ender 3 can check it out!

Keep up the great work :)

I would advise against recommending that mount for the ender 3 by 3DDorsett. If you look closely at the STL model where the two extruding screw holes are, there is a gap. I printed it and those gaps made those important points very fragile.

Thank you for the note. I don't own an Ender3 printer and I think at that time only his mount was made to work with an Ender3 and my fang design. Please feel free to send me a link to a better one!

making one right now.

Is there any way to use this with an EZABL?

This is just the fang that's mounted on the carriage plate by ozfunghi. If anything I suggest you ask him to implement an EZABL. The problem with adding mounting solutions for these things is that you have to own and use them yourself. There's no way to itterate on such a design if you can't test it yourself step by step... atleast not if you want it to work and especially work well for others.

For the exact same reason there are no addons or more variations of my fang design (yet). I'm happy with how it prints and I don't even need/use a bed leveling sensor.

I'm going to make my own lol

that's the spirit! :D make sure to post it here once you are done.

I just finished my first prototype! Printing my first benchy with it right now...

i just instaled this and and yor bracket to mount it but now my extruder fan is to strong and wont let the printer get up above 205c any thorts

First time I'm hearing this, please check that the heat cartridge is secured properly in the hot end block. The fang itself shouldn't be pointing at the heat block itself anyway but just at the very tip of the hozzle. You can always dial down the fan speed too btw... since the cooling is so efficient with the fang I tend to dial down the fan speed in my slicer, because it can already cool well at lower speeds thus making it quieter.

na the other fan the 40x40 one i think mine is 2 strong

  • Posted with 3D Geeks Thingiverse Browser App

Dear Smooth Fang users!

Can you advice me what is the best print position for Smooth Fang?

  1. Fan bracket on the bed, so the Fang stays upwards vertical?
  2. The fan bracket is perpendicular to the bead, so the FANG lays on the bed with two windpipe?


Definitely 1, that's how I optimized it to be printed.

Just make sure to add one layer of brim with 0 distance to the part and give the brim 5-6 outlines. This way it will give the print more surface to stick to the bed while it's very easy to remove once it's printed.

Also I recommend printing the fang by itself due the brim instead of grouping all parts for printing as shown in your screenshots. And just in case something goes wrong with one of the parts during printing you don't have to abort and waste other parts.

Thx for reply, i take your advice.

Do you need a 5015 fan or it will work with the stock fan.

Where can I find the brace to attach the duct and fan in the middle to my ender 3 hot end assembly. in your picture i see it but i am not able to find it. the design look great by the way!!

Please check the description, I have linked to the assembly pieces to both CR-10 and Ender 3.

This design and the latest Petsfang seem like the most promising fang ducts for the Ender and CR10. Can you tell me if there is much of a difference between this design and the Petsfang, or does it just come down to whichever one suits your tastes better? It looks like the Petsfang has the BLTouch mounts closer to the hotend than the d3dorsett mount plate?

I can't give you any comment on that since I never tried the fang that you mentioned. Personally I didn't like the design of the petsfang. I found it looked too blocky with its sharp edges, which I assume will make the fans have to apply more pressure to force the air past those sharp corners.

I took what seemed most appealing to me look and functionality wise and improved it further for my needs. I mainly tried to create a shape for smooth airflow while triying to make the fang as light as possible and easily printable. For example I don't understand why people design or print fangs for two mounted 5015 fans... two are not needed if the design is right, it will only make the whole thing much louder and heavier. There is so much air coming out of my fang with just one 5015 fan (with equal amount on both sides) that I started dialing down the fan speed, simply for a quieter operation.

I'm not using any type of bed leveling sensors, otherwise I would've modeled a fitting holder myself :) I haven't tried the d3dorsett mount plate, I just linked it for those who needed one... I suggest you try out both fang designs if you find the time. Let me know how it went for you and which one you like more.

Will this work with the ezabl mount?

No idea which exact mount you are refering to so I can't tell. If it's a mount using ozfunghi's mount plate as its base then yes.

Yeah, he includes different back plates that come with mounts for the th3d ezabl sensor. Unfortunately, this fang is a little too big. :(

I see, sorry mate.

Comments deleted.

I've since had the chance to install this and it works very well. The 40x20 fan fits perfectly and seems to be doing a much better job pushing air through the heatsink than the 40x10 Noctua.

The fan frame and screw holes feel a bit fragile but they are strong enough when handled with care so I'm happy with it.

One curious observation: I've printed this fang three times now (twice with the stock 4010 fan, once with a mid-power 5015 fan using this fang). Each time there was a small exterior defect due to the extreme overhang at the end of the prongs, though this was significantly reduced with the 5015 fan. The strange thing is, it's always on the right prong (when looking at an installed fang from the front). This makes me wonder if the model is asymmetrical or if my part cooling is not distributed evenly (or maybe it's just a difference in print moves). If I end up printing a fourth one I'll turn it around 180 degrees and see what that does.

EDIT: You can see the defect quite well in the pictures.

Comments deleted.

What could also cause it might be the starting and end points of each layer. If the layer starts always at the same spot (the faulty overhang) try setting it to random, or shift the position to the side which has less overhang.

Thanks for the info. I can assure you though that the model is perfectly symmetrical... in fact I modeled it using symmetry in the first place. I also haven't encountered that problem myself so not sure what's causing it in your case. If it happens due to curling I suggest printing slower (you can set a lower % of speed for layers that take less time in S3D, bring that value even further down so that the tips of the prongs print even slower). It is weird that it happens only on one side though.

Also when it comes to thin walls with overhangs a key aspect to look out for are the settings for perimeters. Try to avoid using 1 outline perimeter since depending on the orientation of the infill per layer that singular outline could curl up easier than with a second perimeter supporting it. Also, depending on the settings of your print width it might by a good try to print with top and bottom perimeters set to 9999. It sounds weird but that forces the slicer to print the model with 100% infill. If you don't set it like that you can clearly see that even with a 100% infill setting, the slicer divides the layers in different sections instead of a consistent singular infill... But all that is really not needed with this design tbh (I printed mine with just 20% infill, 2 outline perimeters at 0.48mm width) ... it's more of an experiment if you might need it.

Thanks for the advice, I'll definitely take it into account on my next attempt. I think you may be onto something with those layer start positions. I don't randomize them and that inner edge of the right prong is indeed the first line drawn on each layer around that height.

I think my base settings should be fine for this. 0.4mm nozzle, 0.2mm layer height, 2700mm/min (45mm/s) with speed reductions, 25% infill, 3 perimeter shells, 5 solid top/bottom layers (though not with that 9999 perimeter setting).

Seeing your base settings I suggest to set it to 0.15mm or even 0.1mm. It is not particularly needed but this will definitely improve the overhangs (less overhang per layer and faster cooldown).

p.s.: the 9999 perimeter setting I mentioned is only for when you plan to print with 100% infill...

Some more user experience follow-up. Been getting great results with it, the best I've ever had. Unfortunately, uhh...


A mysterious tragedy has befallen my hot-end assembly.

Certainly nothing that involved a bench vice and power tools.

After a moment of silence for the stock hot-end I resolved to look upon this as a positive thing (i.e. denial) as it gives me the opportunity to make good on my E3D V6 upgrade plans. For that I will probably have to switch to an entirely different mounting system so... Yeah. My one regret about upgrading is that I won't get to use this beautiful fang anymore.

Sad to hear about the "mysterious" tragedy. Wish you all the best on the E3D V6 endevor. If I every decide to upgrade to that as well I'll make sure to design a new cooling solution :)

Nice design! Any idea if this would fit with a 40x20 fan? (the original V2 fang doesn't)

It should work with a 40x20 fan without problems, since the air ducts in my design are spread apart more than the V2 design. I have only a 40x10 fan here but there's quite a bit of space around it, which tells me a 40x20 should work, too. I will take a photo showing the comparisson between the V2 and my fang once I'm back at home and post it here.

Btw. in case you plan on printing right away, maybe you will want to wait a tiny bit. I'm going to model an additional variant tomorrow which will feature a cable holder. The bowden tube&cables slightly press against the 5015 fan and to prevent them from sliding sideways during the print this holder will come in handy.

Great, thank you. If you look at the top corners of the fan from directly ahead you should be able to see if the fang obstructs the view. I've attached a picture from my V2 fang with a 40x10 fan, looking straight at the top left corner. As you can see the fang passes in front of the corner just a tiny bit, but enough to be in the way. It's the same for both corners of course.

I'll happily wait for the new variant. I haven't bought the 5015 fan yet anyway, still trying to figure out which one to buy. My #1 criterion is for it to be as quiet as possible but it's hard to find solid comparable information on this - especially when they are not running at 100%. For instance the stock 4010 blower I got is fairly quiet at 100% but is way noisier at lower speeds. Any advice on that is welcome, because at this point my best plan is to blow $100 on a whole bunch of 5015 fans and do my own comparison.

I have just added the above mentioned cable holder variant. Happy printing :)

I've reassembled the V2 fang quickly to show you the difference in extruder fan clearance with my design. Check the attached image. I took photos of mine from two different angles so you can see how much space is between the parts. Plus the air ducts are slanted upwards so there should definitely be no issue mounting a 40x20 fan.

I can't tell you much about different 5015 fans but I bought mine based on amazon reviews. And I trusted the reviews since they were written by other 3d print nerds. The fan is a bit more expensive (I paid 20€) than most other cheap offers, too... I hoped that would be noticeable in quality. And so far I really can't complain. A german Amazon link probably won't help you much but try searching for this model: Sunon fan 50x50x15mm MF50151VX-A99 DC 12V 6000 U/min 42dBA

Can't compare it to other 5015 fans but it is a bit louder than the stock fan. It is to be expected though since it is so much more powerful. In general blower style fans are louder than normal ones... I learned that from graphic card cooling.

Thank you for the effort, I really appreciate it. Looks very promising indeed! I'll be sure to print it once I've resolved the partial clog I've been having. That's pretty much the only downside I find with this whole fang design, it becomes quite tricky to remove the bowden tube from the heatsink to do cold pulls.

Ordered the Sunon fan, or well, all 4 speed versions of that fan (6000/5000/4000/3000 RPM). If any visitors from the future are interested in my findings comparing them just give me a poke.

Oh and Amazon DE would've been just fine actually, that's where I buy must of my miscellaneous supplies anyway. Us Belgians are too insignificant to get our own Amazon, the Brits have done their brexit and I'll be damned if I'm going to Amazon FR so that leaves our German neighbours, helpful and reliable as always. :) Thanks a lot!

Can you provide more info for the Sunon fans?
Wich is best (6000/5000/4000/3000 RPM) .
I want to buy the Sunon and i was thinking 6000/5000 rpm.
Will it work? is it to much?
Any info will help.

Certainly. They are all the same fan in essence so there is not a lot of difference between them aside from their top speed (and corresponding sound level).

Single fan setup performance in a nutshell:

  • 3000 RPM = not enough for meaningful cooling, even at 100%
  • 4000 RPM = reasonable cooling but only at 100%
  • 5000 RPM = good
  • 6000 RPM = great

There is little reason to buy the 3000/4000 RPM ones. The 4000 RPM variant could have a niche use if you wanted to go ultra-silent to the point of eliminating the PWM whining noise when running at less than 100%, because the 4000 RPM can be used at 100% to run silent and without PWM noise. However I doubt anyone would find this useful in practice because the Sunon fans do not produce much PWM noise to begin with and your steppers will probably be louder. Mine are, even with dampeners and squash feet.

There's no real difference in noise between them, both from the fan and from PWM. A 6000 RPM fan running at 40% sounds just about the same as 4000 RPM fan running at 60%. Both are 2400 RPM so the fan noise is the same, and the PWM noise is also pretty much the same. This is because the PWM noise comes from the PWM carrier frequency which stays at ~488Hz no matter what percentage you're on. This is of course well within the human hearing range which is why we hear the steady beep tone when running below 100%. The PWM noise of the Sunon fans is comparable to that of the stock 4010 fan, perhaps a bit more silent.

The 6000 RPM is the safest choice since it can do anything the others can and more, without any drawbacks. On a single fan setup I would consider this the ideal choice, although the 5000 RPM will yield very good results as well.

On a dual fan setup 2x6000 RPM is overkill. There's no problem having it but you can't use them at 100% for very long or you will trigger thermal runaway protection. This happens when the marlin firmware notices it's been pumping energy into the hot-end but the temperature isn't going up as it should. I currently use this setup but I only run them at 100% during short bridging. Two seconds of full power can cause the hot-end temperature to drop several degrees, even with a silicone sock. For continuous use I would never put the 2x6000's over 70%, though in practice I most often use them in the 30-40% range.

Since the 6000's are a bit too powerful in a dual setup an argument could be made for using 5000's or even 4000's there. But again, there's really no downside to using the 6000, which is why these are the ones I ended up using. It's nice to have the option of extreme cooling even if you don't use it in most prints.

There's one additional advantage to the fast ones: you can use them to cool down the hot-end WAY faster. I absolutely love them for this. When doing cold pulls and alternating between 210-230 and 90-110, this enables me to work much more quickly. No more waiting 3-4 minutes for the hot-end to cool down, just turn on the fans for 30 seconds and the temperature crashes like a tidal wave.

There's also a tiny disadvantage to using powerful fans: you may find that the heated bed is not able to maintain temperature during the first layers due to the sheer amount of air being blown over it. This is only an issue when you're pushing your bed to its max. Mine maxes out at 81C and if I were to do a print at 80C with a lot of cooling, my bed would probably crash to ~70C before recovering after like 50 layers. That's with a dual fan setup though, I've never had the issue with a single fan.

tl;dr: Get the 6000, or the 5000 as a close second choice.

Please let me know if you have any more questions. For some reason I enjoy this kind of nitpick-comparison. ;)

I hope Rapsey can provide you more information. I wouldn't go for too low RPM though. You can always dile down the RPM in your slicer if you wish for it to be weaker and thus quieter.

It's true that pulling the bowden tube is a bit cumbersome with most fang designs.

I just realized that had I uploaded the wrong version of the cable holder variant yesterday. I accidentaly added the cable holder to an earlier version of my fang design that has attachment issues. I just updated the files, though so in case you have already downloaded the old files, please download again for the proper one.

I don't know what kind of clogging issues you have but I also started having some and at some point I saw filament coming out between the heatblock and the heat break. After unscrewing the heat break I noticed that pieces of the threaded end broke off... see attached photo. It is a budget printer afterall so it's not surprising that some parts will be of bad quality. I had to buy a replacement heat break to resolve the issue.

That's very interesting, I've never seen that happen before.

I don't actually know yet what the source of my clog is. If I remove the nozzle I can feed filament all the way through the (cold) hot-end without applying force, so there are clearly no significant obstructions along the way. But then when I put on a brand new unused nozzle and heat it up I find that I have to push way too hard to get filament through. It does print, but it leads to occasional extruder slippage which in turn leads to underextrusion.

Currently taking the whole thing apart to see if I have damage like yours.

Though unlikely it could also be an issue with temperature... for example the thermister reading higher temperature than it should and thus the heat block underheating. I'd try printing a heat tower ranging from 230-200c.

I suspect that's exactly what it is. During disassembly I noticed that the screw holding the temperature sensor in place was completely loose. I'll know soon enough when it's all back together again, but first I gotta finish dremel-sanding the heat block. Turns out the insulation has taken such a liking to my heat block it decided to mate with it for life...

Loose temperature sensor screw and partially blackened insulation fused to the heat block. If anything this seems to suggest that I was printing at higher temperatures than intended. Not quite sure how this would cause any sort of clogging though.

EDIT: Best theory: the weak cooling potential of the Noctua 40x10 fan combined with printing at higher-than-normal temperatures caused too much heat to creep up into the heatsink. Which is why I was looking to use a 40x20 Noctua fan in the first place.

I wish you luck on that front!

Looks like that fixed it. :) Thanks for the help, printing your latest fang design now.

Nice, tell me how it went. Didn't get even one feedback although over 70 people downloaded it. Kinda eager to hear if it works for others as well as it does for me.

I can't give you any user experience yet as I'm still waiting for the fans to arrive but they both printed just fine. Admittedly the semi-transparent natural PLA is not a great look for this but the one in white looks gorgeous. They both have some minor exterior defects at the ends of the fang tips due to the angles but this is not surprising with the stock part cooling fan. As far as I can see the internal structure of the funnels is as it should be so it shouldn't affect performance. Both of these were printed with DAS FILAMENT. (We love you Germany, please never change)

When the fans get here I was planning to do a comparison using this fang. I'm hoping to find a way to quantify (if not measure) the airflow coming out of the fang so that I can at least know if one fan is better than another. If I can't find a way that is doable I'll just print some bridging tests and do a subjective comparison based on those prints.

The original plan was to compare different fans and fan speed settings on the same fang, but I might expand that and test a few other fangs as well. I reckon the actual print test would be a better choice for that since the cooling effectiveness is also affected by the shape and direction of the funnel exits, not just how much air is coming out of them. I will keep you posted!

EDIT: Oh yeah I did print them with manual supports on the air divider and surrounding frame. I haven't had much luck with bridging in the past so I didn't want to take the chance. Not with the stock part cooling fan anyway.

Thanks for the feedback so far. And thanks for the photos, the prints seem fine.

I use the same filament brand :D great price/perfomance ratio but then again I only tested them and nothing else, and I only print in PLA so far...

The supports aren't really needed, especialy on the surrounding frame but yeah, as always it depends on individual setups. Whatever works best.

It's a good idea to test the fans and other fangs with bridging tests and especially overhang test! Good cooling does wonder where bad cooling leads to curling even on simple overhangs.

For what it's worth, their PETG seems to be equally great. I find it just as easy to print as PLA (if not easier) and get better results with it. Definitely something to try out down the road. ;)

I printed with a PETG sample (translucent red) and it worked ok, but the stringing was severe... even after tramping up the retraction to 8mm.

I also had a lot of stringing when I started printing PETG at 240C. Took it down to 220 after the first layer and it prints just as well with barely any stringing. Increasing retraction beyond a few millimeters never did much to reduce stringing for me, it only causes problems. I use 6.5mm retraction for PETG.

I don't know what slicer you use but in case it's S3D, here's the profile I started from (not my making obviously, credits in the filename). Other than reducing the temperature I don't think I changed anything worth mentioning.

I do find that the stringing produced by PETG is a bit more annoying than PLA. The strings tend to be extremely thin but strong enough not to break. I often throw bits of oozing in the trash only to find that I've drawn an invisible string 2 meters long all the way from the nozzle to the trash can.

Thanks, I will check it out... I do use S3D.

I printed my PETG sample at 230C, haven't tried one at lower temp.

I assume this would fit the ender3 as well?

I know your question is already a few months old, but I can confirm that it fits the carriage of my Ender 3 (which I bought in September 2018) well (I have posted a make).

Thanks for still taking the time to answer

If the hotend carriage of the Ender 3 is the same as the CR-10/10S, then yes it will fit. :)