UPDATED GRAPHS with the 5020 blower fan: Google Drive - bit.ly/5020_CHARTS Imgur - imgur.com/a/6oUTb05 For people who are concerned about the different brand performances of the fans, I can say that these tested fans were "generic" types like Gdstime, MX, Usongshine, etc. Their rotor fin designs are basically identical. The main difference is the RPM at which the fans run. That is why in the graphs on each fan performance line you can see the dots, they indicate increments of RPM of 1k (except the 7530fan). So as the designs are the same the most important factor of performance is RPM and the generic-fan-type brand doesn't really matter for this aspect. If you want better-optimized blower fans search for Delta, Sunon, Sanyo Denki, Mechatronics Fan Group, and others. They will perform better when compared at the same RPM but don't expect miracles like 50% better performance and 50% lower noise. Hopefully, this answers the question if you had one.
Thank you this video, it truly helped me figure out what to purchase! I just did my first printer and what a difference, both in flow and noise. Without your video i wouldn't have started this upgrade 👍
One issue I've noticed with bigger fans is during bridging moves. Some of these larger fans don't even spool up until after the bridge has been completed. Kind of like turbo lag on a car.
i used dual 5015 for a while but switched to a single 12032 that is mounted on the side of the printer and connected trough a hose to the printhead. - all the airflow i could ever need - no weight on the printhead - the fan should live longer because its not being shaken by the hotend moving ;) - runs perfectly fine from the fan header of my creality 4.2.7 board :D
@@sandrolauer350 its a lower pitch then the dual 5015 i used before but its by no means silent. but i dont really care about noise since the printer is in the basement ;)
@@2mD I kept reading about all these 3d printed fan mounts, shrouds etc. and I kept thinking that a bigger side mounted fan with a light, flexible duct would be both simpler and more efficient :) Could you please tell what kind of hose you use (e.g. material, diameter, length) and do you simply terminate it close to the nozzle or do you use something fancier to focus or spread the air around the nozzle? Thanks.
I designed a dual 5015 mod for the voron 2.4, and was still able to run 7k accelerations -- with a canbus toolhead board, so weight savings of removing the cable chains, I was able to keep the weight down, and cooling performance is excellent.
You can easily get 20k acceleration with >350mm/s using stock Stealthburner on a 2.4 Voron with very good quality. I'm running that for several month now. Reducing weight with a custom toolhead should yield even more.
I'm running a 4020 axial fan for the hotend cooling and a 5015 blower for part cooling. I tend to run the blower at around 60% for most prints (for a little less noise and to try and eek out lifespan) but for bridging I tend to ramp it up to 100% (using Cura's "experimental" bridge settings). Most of the time this works pretty well for me - so far.
YES! 4020 axial for heatsinks works pretty well and if you have a branded high-quality 5015 fan you definitely can run them at lower speed while still getting a lot of cooling.
@@PrintingPerspective I did purchase some Winsinn 5015 blowers but they won't run at anything other than 100% speed. I guess something in their electronics doesn't like PWM. I might try a Resister/diode/capacitor filter circuit to see if that helps but since buying them I've read that Winsinn aren't that great anyway - I'll have to build the filter to find out for myself I guess.
@@steveellis2829 Curious about your opinion on the WinSinn fans after running them for a while. I've been running WinSinn ball bearing fans exclusively for over a year (primarily because of price). I've found their performance is excellent, BUT noise levels are "hit or miss"...especially with the 5015 blowers. I have several that are VERY quiet, while others are far worse than the crap Creality fans. I have no idea why, it's just what I have noticed. I'd love to find some simiiiarly priced fans that were pretty quiet.
@@tommyc8333 Certainly WinSinn fans aren't the quietest of fans but they have so far proved to be pretty reliable however I do tend to peel back the label and put just a drop of machine oil onto the bearings before I put them into service - this might be helping. I've not noticed any being quieter or noisier than others but then I've not needed to replace any yet so that might change further down the line.
Just as an update - I built the filter for the Winsinn fans and can happily run the fans at lower speeds without issue or strange noises. They've been running flawlessly for a few months now. All good.
Fantastic job! Succinct video that sums up excellent suggestions. Bottom line: 5015 or two that has been used for years now is, in general, the way to go. You have a knack for this type of topics. Wonder if one day you'll disprove a common practice (like CNC kitchen has done a few times). Well done!
Thanks a lot, I am glad you like the videos. Yeah, 5015 fans are great. The only common practice that I know and could have real potential is using blower fans instead of axial to cool the heatsink of the hotend. Using a smaller heatsink with thinner fins and a 4010 blower fan at reduced speed could make the hotend smaller, and fans more silent without losing required cooling. Sprite Extruder has something similar but it uses the almost useless 3010 blower fan which must run at insane speed being very loud to compensate for its size.
Im running a Neptune 3 pro and one of the fans has a pinched wire and was causing thermal runaway when the fans turned on. So I printed the Wind Warden upgrade and snagged some 5015 fans and Im glad to see I made the right choice! I print a lot with a .6 mm nozzle and I have been eyeing a 50-15 fan upgrade for a while. The bridging on a nozzle that large is terrible but hopefully this upgrade can mitigate that. Also the wind Warden upgrade makes the tip a lot easier to access
You can get 5020 large exit, same speed / weight as 5015, less noise and more airflow. Even better stick a couple fixed behind the gantry so they don't weight on the toolhead.
I run a HeroMe gen 7 set up with a Noctua 40x20 on the hot end and two 5015 fans for part cooling. I have a sprite I am thinking about modding as the stock blowers are well...
this is already a GOAT video/comparison, but I'm dying to know how 4x 4010 would compare? Takes up around as much space as 2x 5015, and the 2x 4010 was surprisingly strong.
Yeah, server fans can do that as they operate at 10K+ RPM. Probably not the best airspeed-to-noise ratio but I doubt that is relevant as you are rocking RAT RIG V-CORE 3 at some sick speeds :)
You just listed my setup, I’m looking at making a mod to run 2 4028. This would allow quieter ops running at like 50% while having higher headroom if I want to print a 5 min benchy
I'd love to see several more high-power designs added to the comparison: WS7040 CPAP (used on VZbot) Roborock CPAP (used on CPAP mods for many printers) 4028 axial (used on Ratrig V-Core) 3628 axial (used on Takeoff toolhead)
I have two 5015 fans on my e3d v6 /titan clone setpu, i have a week trying to find why my printer halt, then i figure it out that i need the TEFLON socket because if the fans go up 60% , the air refelected in the model download my hotend temperature so low , thats triger the Marlin error , and stop printig, so PUT A SILICON SOCKET IN THE BLOCK, PROBLEM SOLVE (i musts do that first but i change the termistor, nozzle, try pid calibration , all of that first...good leson for me)
My printer has stock 30mm for E3D Heatsink and 2x 4010 fans. Interestingly the 30mm fan was the big problem on mine not the 4010 fans. So I have to replace the crappy 30mm fan since I got heatcreep issues and it was loud as well.
You know what, i think 5015 centrifugal would do nicely for cooling the e3d coldend heatsink :D With a little voltage regulator to go along with it and drop its noise level.
A 247printing / bambulab aux fan not fitted to the print head seem to be the best upgrade. Not seen it done on a bed slinger yet as you'd have to mount it on the x gantry, but it's totally doable.
Thank you for your comparison testing. I'm only just starting out into the wide world of 3D printing and as such this may sound a bit weird. But has anyone tried to use a high volume air pump? There'd be no noise from a fan per say, and the air flow can be fine adjusted to the precise point at which it is needed and can be located away from the hot end and have no vibration issues
Are high volume air pumps are quiet? I really doubt they are, plus you will still hear the whoosing noise out of the duct hitting the print. Vibration issues? Unless you hit the speed at which fan amplifies vibrations then sure, but even with a fan that has high vibrations it is not issue ruclips.net/video/TXw37ydld5g/видео.html
Something else about 4020 fans, is they blow air out at a 90 degree angle towards one side. This will cause uneven air flow, especially if you have split or more ducting. I found smaller ducting gives it more backpressure and far more even distribution. And going on a deep dive also found blower motors are made to operate in high static pressure environments, i.e. pushing air through long ducts in homes that go up multiple stories while pulling though dirty air filters. Either way, when i tested out a 5015 fan and i found it blows out almost straight out and moves slightly more outside of a duct. BUT i also tested a 4020 blower fan with 0.3A draw, vs a typical 0.08A draw.. literally blew my hair back. Only issue with a 24v 0.3A DC fan is low speeds, might ruin any print that isn't PLA and used mechanically. I'm currently designing a new part cooling duct for the 5015, see how that goes
I go through a 4010 heatbreak fan about every 8 months. I really wish there was a more durable option out there. My 5015 part cooling fans have no issues with the acceleration (25k @ 750mm/s)
The most really important points are completely missed here.... We need to talk about airflow and air pressure. When you open up a spec sheet, these are the values you find. This is your first clue: No manufacturer talks about airspeed.... Also, if you read some specsheets, you will find massive differences between blower fans of the same size. So comparing random models is questionable. Example1: An axial fan delivers a very high airflow, but a very low air pressure. If you mount an axial fan to a long/complex cooling duct, it will be ineffective. Because it cannot press the air trough the duct. That is why we use blower fans for part cooling - they deliver a much higher air pressure. Example2: The BerdAir system uses an actual airpump and delivers a very little amount of air under extrem pressure to the printed part. This outperforms your standard 5015 fan and is commonly used in the highend area. So we further increased the pressure and decreased the airflow. HOWEVER: If you your ducting/delivery sytem is quite open/direct, a higher airflow at lower pressure does the trick as well. It is however less directed to immediately cool the plastic when it is layed down. This might affect the printers performance while printing overhangs. You want to cool these directly out of the nozzel. Conclusion regarding airspeed: If you deliver twice the air at a slightly lower airspeed, your cooling will still be better overall. If you widen your duct for a "lower pressure" airspeed fan, you will immediately increase its overall performance in direct comparsion to a "higher pressure" fan. The "higher pressure" fan will profit less of this change. Meaning your test setup does provide only partially valid data. In general: The RIGHT amount of cooling will deliver the best printing result. Too much cooling will even result in issues - depending on your filament. The required cooling depends on the material flow. If you build a highend-speed-printer, your 5015 fan will not be sufficient. Of if you print with a 0.8mm or 1.0mm nozzle. For your standard printers, like Prusa i3, Ender etc. the print speed is limited. There is no need to make the printhead heavier by mounting an overweight part cooling fan to it. A proper fan duct is most important here - you want equal cooling performance from all sides.
@sourand jaded Well, the air heats up when you compress it and cools down when it expands. So the total energy and therefore temperature remains equal. We could get into tiny details now, where the compressed and therefore heated air radiates some heat away on its way to the air duct. But also an airpump will create more heat than a simple fan and transfer some of it into the air it compresses. So I would assume these effects are neglectable. Regarding radial fans: This was for sure a massive improvement over axial fans for part cooling. It is sometimes strange... with today's knowledge and experience, it all feels so obvious. But it is for sure not, when we hear about these things for the first time. I have been working on a design using two large radial fans connected with a high-diameter hose to the printhead. I am targeting a high airflow approach over maximum pressure. The reason for this design was simple: The airpumps for a berd-air are too loud. By now, the community came up with a similar design, but they use a lower hose diameter and ultra high rpm radial fans. These ones sound like a trubine, which makes it a no-go for me again. But the cooling significantly outperforms the berdair system.
Michael Schaub, I perfectly understand your points, but I think you are taking this video as in your eyes it should be some sort of definitive comparison with the most accurate scientific data. You compress the air into the duct with one of these blower fans, but when the air leaves the duct that air expands resulting in an increase in airspeed. So I don't think my results are misleading when using airspeed as a performance indicator. The whole video is based on giving a general comparison of what you could expect from these fans. And I am using the test duct opening size that is similar to the real ducts for regular 3D printers like Enders, Prusas and etc. Not to mention that there can be slightly better-optimized fans of each type. That is why I think a general comparison is more than enough.
@@PrintingPerspective I do understand your point. My conclusion was, that the data is partially valid. And yes, if you conclude that a bigger fan gets you a better performance to noise ratio, I do agree with that statement in most cases. I just think that the perspective is too limited. When I listen to the 5015 fan on my I3, I do not hear the fan, but I do hear the air passing trough the airduct and I hear the air pushing on the printed part. A different air duct will make a greater change in noise level than your fan ever could. If you now use an approach with higher volume and lower pressure, your noise level will drop - independed from the fan noise. Simply because lower pressure in the fan duct results in lower speed in the fan duct with results in lower noise. Again, huge dependency of the kind of airduct design. And regarding differences between same sized fans: I spent hours searching through spec sheets when I designed my external part cooling system which features a hose to the printhead. We are not talking about a minor difference. The pressure between two fans of the same size can be double - and the fans will look quite similar. Regarding your last comment: You got it the wrong way around: If you compress the air, the air speed increases. The speed is highest, at the most narrow spot. With expansion (wider spots / leaving the air duct) the air speed decreases. However, this is just a sidenote and does not impact your results, since you measured all ducts the same way. In summary: If you want to provide a general overview in the thought of bigger fan = better noise/performance ratio: This statement remains correct as long as you work with short / simple airducts which are not significantly more narrow than the exit of the fan itself. Meaning, for your I3, Ender, etc. the results are generally correct.
This kinda comparison video is so hard to make as there are so many variables. My goal with it was to provide the most relevant data for the most common 3D printers, which is not perfect but at least it is not anecdotal data, you know. I appreciate the in-depth comment, there is always something to learn from those.
I run 2x5015 for print cooling and one 4020 for hot end cooling. I’m impressed with this setting. I have a hard modded Ender 3s1 pro. Build a housing, use clipper, have linear rails on all axis, the z axis is with Pom anti backlash and brass sliders for z wobbling modded. I have a modded breakout board. Installed a bi metal heatbrake, z axis belt tensioner, and a full custom heat bed by a 750 watt 220v silicone heater isolated of course with an 6mm cast and cnc machined aluminum plate on silicone spacers. And a have squash balls feet. My whole electronic is upgraded with safer parts (meanwell power supply, extra fuse for the bed, ssr relay for the bed, some additional fan, cutouts in the botten cover).
That is a lot of mods. So, basically only what has left from Ender is the frame? You should make a name for it because it is nothing like Ender anymore, haha! :)
@@PrintingPerspective I should think about it, I forgot the copper heatblock 😉 Bit i Tell you, the energy saving and consistance of printing is amazing.
I have replaced my original (? RPM.. no datasheet, not even by searching fan part no) 4010 fans ony my SuperRacer to GDSTime 12K RPM ball bearing ones. They blast air, but also loud a.f. :)
Word of note: i tried a Sunon MagLev fan, and it was a catastrophe, the magnetic bearing just cannot deal with high vibration environments. It made horrible noises. I'm still using it but i filled it with so much grease that it's mostly a sleeve bearing fan now :D My previous sleeve bearing fan died quickly, but the converted Sunon seems to be holding on. I think double ball bearing fans are the only fans suitable for a high vibration use. So next a Delta fan i suppose :D Sunon also has another problem, Autorestart feature. It does not play well with PWM, the fan goes crazy. You can either up the PWM frequency, that sort of works but lead to weird noises. I ended up building a Pi filter, but i sort of went a little wild on capacitance, the control delay is hilarious but it works i guess.
it would be interesting then what kind of fan would be ideal for side cooling for z-axis print bed (as opposed to bolted down printbed/bedslinger) considering mass isn't a big deal, what would be the sweet spot of airflow vs noise
I bought 2 4020 by mistake so I use those in pair, I agree with everything you said, they are weak, can rattle at high accel, in pair they are ok but not much or at all better than 2 4010
I tried a 4020 fan but it amplifies the noise from the pwm speed control. I found that 2 4010 fans have good cooling and less noise than the 4020. I don't have space for 5015 in my printer.
Now how does airspeed affect print quality? From what I've read the higher the cooling, the better the pieces look, but the lower the layer adhesion. My question would be where is a good point? I usually run a single 5015 fan at full speed with the Petsfang cooler. Is this maybe already too much?
Cooling does not harm layer adhesion at all unless your hotend temperature is too low and you're relying on the previous layer still being hot and soft in order for the next layer's heat to be sufficient for bonding.
@@daliasprints9798 that depends a lot on what material you print. too much cooling can cause serious layer adhesion problems with filaments such as petg or asa. but for pla its usually fine to blast it with as much cooling as you can.
@@2mD For ASA and ABS, cooling to room temp is problematic but if you're using a chamber, full strength cooling to the chamber temperature should be ok. I don't have experience doing this but the folks I know with chambers & Annex printers seem to do fine. For PETG, I print with my full cooling (over 100 cfm total air flow) and no layer adhesion problems, at 270-290°C.
@@daliasprints9798 my chamber isn't heated ( yet :D ) so it only reaches ~40°C, maybe that's the problem. and 290°C for petg seems quite high. i usually print somewhere between 235 ( extrudr XPETG ) and 250 ( most other PETGs ). and with that i get layer adhesion problems with more then ~40% partcooling ( 12032 blower fan ). but if the higher temp works for you im gonna give it a try :)
I only ever met 40x10 fans that come stock from Creality and they are noisy AF. I've switched Steve's ones to two axial Gdstme 40x20 with Briss Fang, and it was way quieter all this time. Now I've switched to Creality Sprite and it comes with another set of noisy Creality fans. I'm considering a mod for two 40x20 again because I have them, but I think the mod I'm looking at requires one axial and one.. the other one, so I'd still have to buy some more. One 50x20 is considered two... I'm not ready for two of them ;)
The most mind blowing thing is that no company yet made a 3D printer oriented to be silent. Because I think that is what most people want as they have a printer in a living room and the noise gets annoying pretty quick.
@@PrintingPerspective Yes, I'd quite like it to be quiet. With my previous fan setup, I could close one sliding door (the printer lives in a wardrobe) and struggle to remember if it's on, with two I couldn't hear it. Now I still can hear it's on with all three doors. No sleeping with it working unless one really likes white noise with a bit of whining 🙈
Yup. That is why I worded the title - "3D Printer Part Cooling FANS" and tested all the fans in a way that the data would be relevant for this application.
How effective is a 5015 as a heatsink fan, if I can't mount it directly? I currently have a 3010 (or 3015, don't remember exactly) fan mounted directly to the heatsink (v6), and I'm wondering if placing the 5015 fan above the heatsink and ducting it would result in quieter operation?
In general blower fans are more efficient than axial at forcing air through small ducts or through the fins of the heatsink. 3010 achieves good cooling results by running it fast and annoyingly loud. I think 5015 as a heatsink fan would be a little bit overkill, but if the silence is your number one goal then yes it could provide better noise levels with better cooling. Just keep in mind that you have to reduce its speed first and the easiest way would be with a resistor. And if we are talking about ducting 5015 fan that is on the printhead then I don't think you need to worry about the performance loss.
@@PrintingPerspective Thanks for the reply! I'm running almost stock cooling setup still (might be a 4010 fan, but no duct)l, I'm planning on running a 5015 for part cooling, not worried about duct losses there, just the heatsink - I've got extremely limited space - it's a corexy, but there's little space from the build platform to the edges -, and a 5015 would fit better than the 4020, for axial fans I think only the 3010 is possible, don't have the distance from heatsink for a bigger fan with duct.
Oh yes, 5015 fans are noisy as hell :)) I am in the middle of redesigning the hotend cooling, thanks for the data, it's useful and i will change my design
I question your anemometer setup. Ideally an anemometer presents no loading to the airflow, it just spins freely. But using only one third of the anemometer opening is making it behave like a turbine with an attached compressor. The open side is going to push/pull ambient air presenting the fans with a much higher loading (pressure drop) than they would be with just a duct alone. This will bias the measurement, and fans with better pressure performance vs airflow will outshine. Minimally you should block the open inlets of the anemometer so they have less loading.
More known brands like delta, sunon have datasheets so it is very easy to compare fans yourself. Plus sourcing all those would be a nightmare as you can't find them in on place or at least I don't know where. :/
Now when I done the whole video it would. Too bad, I didn't have any on hand and the ordered ones didn't arrive. But when I was looking at the datasheets 5020 fans usually had way lower static pressure compared to 5015's and that is definitely not good if your cooling duct openings are narrow. But then what is better? Wider or narrower duct openings? :D It's like a never ending story, haha!
@@PrintingPerspective Interesting. The ones I've seen are mounted without any complex or or shrinking duct path, for example the Annex usermods Chubby, so indeed they're in a setting where static pressure matters a lot less. But I thought they were comparable. I'll look at the specs on some. Been thinking about changing out my 5015s for 5020s.
I love 50x20 blowers (the ones with intakes on both sides). They're so much quieter for the amount of airflow compared to anything else I've tested, I started incorporating them into my own designs. If you want to design around one, I uploaded a 5020 blower model to thingiverse (same username) some time back. I've been surprised how popular that model has been. Apparently a lot of people like them. :)
as far as i understood partcooling its not about airspeed but airflow. airspeed: the speed of the air in m/s as mentionend in the video. airflow: how much airvolume per second. i want to know about the airflow and not about the airspeed. cause the speed could be very high but the airflow at the same time very low. so this data seems quite useless in my opinion.
You understand that those two things are very closely related, right? Airspeed gives more concentrated cooling. Also higher airspeed gives a higher cooling rate that can be equal to a higher airflow. And the opening size of the ducts on most 3d printers are more orientated to make direct concentrated airflow at higher airspeeds. That provides a more consistent cooling when moving in any printing direction and especialy when printing thin details that needs high cooling rate in a small area. So for you saying that this data is useless clearly shows that you have no understanding of this complex subject.
@@PrintingPerspective OP is right, airspeed is only viable as a proxy for airflow at very low static pressures and that’s exactly what your test is missing. Without static pressure numbers the tests are pretty much meaningless.
@@PrintingPerspective search printables for Noctua A4x20 see what I am talking about. People replace server fans with these to cut the noise at home, 5000 rpm, optimized for static pressure, I believe they are rated at 14.9 db, 5.2 CFM. Using it may not be simple, 12v pwm, would be interesting to replicate. I would love to see how effective different styles of cooling ducts are, not sure how you would judge that though.
NF-A4x20 has 2,26mm H2O static pressure, while 5015 fans at same RPM have ~12mm H2O. If you didn't watched I suggest checking this video where I compared axial vs blower performance - ruclips.net/video/mxb71Ndg6tY/видео.html
It's difficult to find a 5015 blower that doesn't whistle at speed lower than 100%. Often materials like PETG needs 20-40% blower speed, and the whole printing will be with an obnoxious hissing.
Sleeve bearings are only silent when new. Hydraulic bearings are better sleeve bearings with a bigger oil reservoir. Ball bearings stay the same for years.
I mean, I agree with you more than not. But for example, I had a 4512 ball-bearing blower fan that ran at ~7500 RPM and lasted not even a year with a relatively small amount of printing.
Obviously. Because they run at different RPM speeds. That is why I showed graphs with RPM:Performance ratio. Plus unbranded fans from China that almost every printer uses (or you can buy) come with almost identical blade fin designs. There always be variables, but my goal is to minimize them as much as possible and provide the most relevant use-case scenario for a good reference point.
I mean, did you run them at full speed that noise is the issue? Maybe it is the duct that increases the noise because they affect the noise levels too.
@@PrintingPerspective I was using the hero me duct on an ender 3 and theres really a bunch of problems with the duct itself that made ma not a fan of the design... but ya, if you have to run the fans at 50% just to be as quiet as the non-silent steppers there really isnt a point (for me) to use the bigger, higher power draw, louder, heavier fans.
Hey thanks for the video I was searching for something like this for a long time, ty for going so in-depth. I had a question what if I use no nozzle on the front and run a 7530 vs 2x 5015 (price and weight don't matter) will the dual 5015 still pull ahead with unobstructed flow? edit: both noise and heat removal are important as they will be used in a room next to the bedroom
I always feel like the 4020 moves more air than the 5015. Honestly, I hate 5015's. They're too large and I always felt their output sucked but clearly I Was wrong. I really like the 4020's as noise isn't a problem. Not sure why I feel like this. lol
4020 blower could feel that they provide more airflow but another very important aspect when using the fan in a duct is the static pressure. And there is no way that 4020 provides more of it than the 5015 at the same RPM. But I feel you, I thought 4020 would perform better.
I suppose it's completely subjective, but measuring noise level alone doesn't reflect the TONE of the noise. I am much more comfortable with the tone of a 4020 or 5015 fan than the screech of a 4010 fan.
Brand/manufacture is the biggest factor to all aspects here unfortunately. Those appear to be wathai 4020 fans. There is a new version out more that completely annilate the old ones
UPDATED GRAPHS with the 5020 blower fan:
Google Drive - bit.ly/5020_CHARTS
Imgur - imgur.com/a/6oUTb05
For people who are concerned about the different brand performances of the fans, I can say that these tested fans were "generic" types like Gdstime, MX, Usongshine, etc. Their rotor fin designs are basically identical. The main difference is the RPM at which the fans run. That is why in the graphs on each fan performance line you can see the dots, they indicate increments of RPM of 1k (except the 7530fan). So as the designs are the same the most important factor of performance is RPM and the generic-fan-type brand doesn't really matter for this aspect. If you want better-optimized blower fans search for Delta, Sunon, Sanyo Denki, Mechatronics Fan Group, and others. They will perform better when compared at the same RPM but don't expect miracles like 50% better performance and 50% lower noise. Hopefully, this answers the question if you had one.
How do you think the gdstime 4015 would rank?
Thank you this video, it truly helped me figure out what to purchase! I just did my first printer and what a difference, both in flow and noise. Without your video i wouldn't have started this upgrade 👍
The interesting thing about videos about fans is that when the video is really good it can turn yourself into a fan.
Get out.
One issue I've noticed with bigger fans is during bridging moves. Some of these larger fans don't even spool up until after the bridge has been completed. Kind of like turbo lag on a car.
Fan should have a program advance in the slicer.
@@SianaGearz nah I think we need a misfiring system
@@Iisakki3000 Elaborate please?
@@SianaGearz i think he meant keeping the fan spooled up, redirecting the airflow when necessary.
Using compressed air is sort of that like solution.
Finally, I have been looking for this comparison for a long time.
i used dual 5015 for a while but switched to a single 12032 that is mounted on the side of the printer and connected trough a hose to the printhead.
- all the airflow i could ever need
- no weight on the printhead
- the fan should live longer because its not being shaken by the hotend moving ;)
- runs perfectly fine from the fan header of my creality 4.2.7 board :D
That kinda setup is so interesting. One day I will probably try it too :)
Wow now that is an interesting setup! Got any more info to share on it?
What about the noise level is this setup?
@@sandrolauer350 its a lower pitch then the dual 5015 i used before but its by no means silent.
but i dont really care about noise since the printer is in the basement ;)
@@2mD I kept reading about all these 3d printed fan mounts, shrouds etc. and I kept thinking that a bigger side mounted fan with a light, flexible duct would be both simpler and more efficient :)
Could you please tell what kind of hose you use (e.g. material, diameter, length) and do you simply terminate it close to the nozzle or do you use something fancier to focus or spread the air around the nozzle? Thanks.
I designed a dual 5015 mod for the voron 2.4, and was still able to run 7k accelerations -- with a canbus toolhead board, so weight savings of removing the cable chains, I was able to keep the weight down, and cooling performance is excellent.
You can easily get 20k acceleration with >350mm/s using stock Stealthburner on a 2.4 Voron with very good quality. I'm running that for several month now. Reducing weight with a custom toolhead should yield even more.
Which mod is it? Manta or the Owl?
Do you have a link to the setup so I can try and print it?
I'm running a 4020 axial fan for the hotend cooling and a 5015 blower for part cooling. I tend to run the blower at around 60% for most prints (for a little less noise and to try and eek out lifespan) but for bridging I tend to ramp it up to 100% (using Cura's "experimental" bridge settings). Most of the time this works pretty well for me - so far.
YES! 4020 axial for heatsinks works pretty well and if you have a branded high-quality 5015 fan you definitely can run them at lower speed while still getting a lot of cooling.
@@PrintingPerspective I did purchase some Winsinn 5015 blowers but they won't run at anything other than 100% speed. I guess something in their electronics doesn't like PWM. I might try a Resister/diode/capacitor filter circuit to see if that helps but since buying them I've read that Winsinn aren't that great anyway - I'll have to build the filter to find out for myself I guess.
@@steveellis2829 Curious about your opinion on the WinSinn fans after running them for a while. I've been running WinSinn ball bearing fans exclusively for over a year (primarily because of price). I've found their performance is excellent, BUT noise levels are "hit or miss"...especially with the 5015 blowers. I have several that are VERY quiet, while others are far worse than the crap Creality fans. I have no idea why, it's just what I have noticed. I'd love to find some simiiiarly priced fans that were pretty quiet.
@@tommyc8333 Certainly WinSinn fans aren't the quietest of fans but they have so far proved to be pretty reliable however I do tend to peel back the label and put just a drop of machine oil onto the bearings before I put them into service - this might be helping. I've not noticed any being quieter or noisier than others but then I've not needed to replace any yet so that might change further down the line.
Just as an update - I built the filter for the Winsinn fans and can happily run the fans at lower speeds without issue or strange noises. They've been running flawlessly for a few months now. All good.
Fantastic job! Succinct video that sums up excellent suggestions. Bottom line: 5015 or two that has been used for years now is, in general, the way to go. You have a knack for this type of topics. Wonder if one day you'll disprove a common practice (like CNC kitchen has done a few times). Well done!
Thanks a lot, I am glad you like the videos. Yeah, 5015 fans are great. The only common practice that I know and could have real potential is using blower fans instead of axial to cool the heatsink of the hotend. Using a smaller heatsink with thinner fins and a 4010 blower fan at reduced speed could make the hotend smaller, and fans more silent without losing required cooling. Sprite Extruder has something similar but it uses the almost useless 3010 blower fan which must run at insane speed being very loud to compensate for its size.
Great video, I've always thought 4020 was a great middle ground for airflow vs bulk, but apparently not
Im running a Neptune 3 pro and one of the fans has a pinched wire and was causing thermal runaway when the fans turned on. So I printed the Wind Warden upgrade and snagged some 5015 fans and Im glad to see I made the right choice! I print a lot with a .6 mm nozzle and I have been eyeing a 50-15 fan upgrade for a while. The bridging on a nozzle that large is terrible but hopefully this upgrade can mitigate that. Also the wind Warden upgrade makes the tip a lot easier to access
You can get 5020 large exit, same speed / weight as 5015, less noise and more airflow.
Even better stick a couple fixed behind the gantry so they don't weight on the toolhead.
I run a HeroMe gen 7 set up with a Noctua 40x20 on the hot end and two 5015 fans for part cooling. I have a sprite I am thinking about modding as the stock blowers are well...
this is already a GOAT video/comparison, but I'm dying to know how 4x 4010 would compare? Takes up around as much space as 2x 5015, and the 2x 4010 was surprisingly strong.
Is there any way to have a second high velocity fan that only kicks on for overhangs?
Not sure if there are slicers that can add gcode for that?
To my knowledge, there aren't, we definitely need more advanced slicers to utilize cool features like that.
I personally use a 4028 server fan, pushes so much air, and it'g great for fast parts.
Yeah, server fans can do that as they operate at 10K+ RPM. Probably not the best airspeed-to-noise ratio but I doubt that is relevant as you are rocking RAT RIG V-CORE 3 at some sick speeds :)
You just listed my setup, I’m looking at making a mod to run 2 4028. This would allow quieter ops running at like 50% while having higher headroom if I want to print a 5 min benchy
I'd love to see several more high-power designs added to the comparison:
WS7040 CPAP (used on VZbot)
Roborock CPAP (used on CPAP mods for many printers)
4028 axial (used on Ratrig V-Core)
3628 axial (used on Takeoff toolhead)
You should also consider current ratings, bearing type and brand, for example i got a SK branded 5015 woth dual bearings and its so god damnit loud
I have two 5015 fans on my e3d v6 /titan clone setpu, i have a week trying to find why my printer halt, then i figure it out that i need the TEFLON socket because if the fans go up 60% , the air refelected in the model download my hotend temperature so low , thats triger the Marlin error , and stop printig, so PUT A SILICON SOCKET IN THE BLOCK, PROBLEM SOLVE (i musts do that first but i change the termistor, nozzle, try pid calibration , all of that first...good leson for me)
My printer has stock 30mm for E3D Heatsink and 2x 4010 fans. Interestingly the 30mm fan was the big problem on mine not the 4010 fans.
So I have to replace the crappy 30mm fan since I got heatcreep issues and it was loud as well.
You know what, i think 5015 centrifugal would do nicely for cooling the e3d coldend heatsink :D With a little voltage regulator to go along with it and drop its noise level.
@@SianaGearz This may be, but it´s also heavy. I just used a 40mm instead.
A 247printing / bambulab aux fan not fitted to the print head seem to be the best upgrade.
Not seen it done on a bed slinger yet as you'd have to mount it on the x gantry, but it's totally doable.
5015 for parts cooling and 5015 for extruder cooling on my modded Ender3S1.. works great.
Excellent video!
Thank you! Cheers! :)
Thank you for your comparison testing. I'm only just starting out into the wide world of 3D printing and as such this may sound a bit weird. But has anyone tried to use a high volume air pump? There'd be no noise from a fan per say, and the air flow can be fine adjusted to the precise point at which it is needed and can be located away from the hot end and have no vibration issues
Are high volume air pumps are quiet? I really doubt they are, plus you will still hear the whoosing noise out of the duct hitting the print. Vibration issues? Unless you hit the speed at which fan amplifies vibrations then sure, but even with a fan that has high vibrations it is not issue ruclips.net/video/TXw37ydld5g/видео.html
I love your videos so much ❤❤. They are really interesting and thank you for the work you put into into benchmarking!
Glad you like them! :)
What about 75mm fans?
I run dual 5015 fans on my anycubic vyper, and a single 5015 on my svo6 plus.
Great video! Thank you ✨
Something else about 4020 fans, is they blow air out at a 90 degree angle towards one side. This will cause uneven air flow, especially if you have split or more ducting. I found smaller ducting gives it more backpressure and far more even distribution. And going on a deep dive also found blower motors are made to operate in high static pressure environments, i.e. pushing air through long ducts in homes that go up multiple stories while pulling though dirty air filters.
Either way, when i tested out a 5015 fan and i found it blows out almost straight out and moves slightly more outside of a duct. BUT i also tested a 4020 blower fan with 0.3A draw, vs a typical 0.08A draw.. literally blew my hair back. Only issue with a 24v 0.3A DC fan is low speeds, might ruin any print that isn't PLA and used mechanically.
I'm currently designing a new part cooling duct for the 5015, see how that goes
I go through a 4010 heatbreak fan about every 8 months. I really wish there was a more durable option out there. My 5015 part cooling fans have no issues with the acceleration (25k @ 750mm/s)
2 x winnsinn 5015 hydraulic sleeve bearing on sprite extruder,sonic pad speed..the fans are spot on
The most really important points are completely missed here....
We need to talk about airflow and air pressure. When you open up a spec sheet, these are the values you find.
This is your first clue: No manufacturer talks about airspeed....
Also, if you read some specsheets, you will find massive differences between blower fans of the same size. So comparing random models is questionable.
Example1:
An axial fan delivers a very high airflow, but a very low air pressure.
If you mount an axial fan to a long/complex cooling duct, it will be ineffective. Because it cannot press the air trough the duct.
That is why we use blower fans for part cooling - they deliver a much higher air pressure.
Example2:
The BerdAir system uses an actual airpump and delivers a very little amount of air under extrem pressure to the printed part.
This outperforms your standard 5015 fan and is commonly used in the highend area.
So we further increased the pressure and decreased the airflow.
HOWEVER:
If you your ducting/delivery sytem is quite open/direct, a higher airflow at lower pressure does the trick as well.
It is however less directed to immediately cool the plastic when it is layed down. This might affect the printers performance while printing overhangs. You want to cool these directly out of the nozzel.
Conclusion regarding airspeed:
If you deliver twice the air at a slightly lower airspeed, your cooling will still be better overall.
If you widen your duct for a "lower pressure" airspeed fan, you will immediately increase its overall performance in direct comparsion to a "higher pressure" fan.
The "higher pressure" fan will profit less of this change.
Meaning your test setup does provide only partially valid data.
In general:
The RIGHT amount of cooling will deliver the best printing result. Too much cooling will even result in issues - depending on your filament.
The required cooling depends on the material flow.
If you build a highend-speed-printer, your 5015 fan will not be sufficient. Of if you print with a 0.8mm or 1.0mm nozzle.
For your standard printers, like Prusa i3, Ender etc. the print speed is limited.
There is no need to make the printhead heavier by mounting an overweight part cooling fan to it.
A proper fan duct is most important here - you want equal cooling performance from all sides.
@sourand jaded Well, the air heats up when you compress it and cools down when it expands. So the total energy and therefore temperature remains equal.
We could get into tiny details now, where the compressed and therefore heated air radiates some heat away on its way to the air duct.
But also an airpump will create more heat than a simple fan and transfer some of it into the air it compresses.
So I would assume these effects are neglectable.
Regarding radial fans:
This was for sure a massive improvement over axial fans for part cooling.
It is sometimes strange... with today's knowledge and experience, it all feels so obvious.
But it is for sure not, when we hear about these things for the first time.
I have been working on a design using two large radial fans connected with a high-diameter hose to the printhead.
I am targeting a high airflow approach over maximum pressure.
The reason for this design was simple: The airpumps for a berd-air are too loud.
By now, the community came up with a similar design, but they use a lower hose diameter and ultra high rpm radial fans.
These ones sound like a trubine, which makes it a no-go for me again.
But the cooling significantly outperforms the berdair system.
Michael Schaub, I perfectly understand your points, but I think you are taking this video as in your eyes it should be some sort of definitive comparison with the most accurate scientific data.
You compress the air into the duct with one of these blower fans, but when the air leaves the duct that air expands resulting in an increase in airspeed. So I don't think my results are misleading when using airspeed as a performance indicator. The whole video is based on giving a general comparison of what you could expect from these fans. And I am using the test duct opening size that is similar to the real ducts for regular 3D printers like Enders, Prusas and etc. Not to mention that there can be slightly better-optimized fans of each type. That is why I think a general comparison is more than enough.
@@PrintingPerspective
I do understand your point. My conclusion was, that the data is partially valid.
And yes, if you conclude that a bigger fan gets you a better performance to noise ratio, I do agree with that statement in most cases.
I just think that the perspective is too limited. When I listen to the 5015 fan on my I3, I do not hear the fan, but I do hear the air passing trough the airduct and I hear the air pushing on the printed part. A different air duct will make a greater change in noise level than your fan ever could.
If you now use an approach with higher volume and lower pressure, your noise level will drop - independed from the fan noise. Simply because lower pressure in the fan duct results in lower speed in the fan duct with results in lower noise. Again, huge dependency of the kind of airduct design.
And regarding differences between same sized fans: I spent hours searching through spec sheets when I designed my external part cooling system which features a hose to the printhead.
We are not talking about a minor difference. The pressure between two fans of the same size can be double - and the fans will look quite similar.
Regarding your last comment:
You got it the wrong way around: If you compress the air, the air speed increases. The speed is highest, at the most narrow spot.
With expansion (wider spots / leaving the air duct) the air speed decreases.
However, this is just a sidenote and does not impact your results, since you measured all ducts the same way.
In summary:
If you want to provide a general overview in the thought of bigger fan = better noise/performance ratio:
This statement remains correct as long as you work with short / simple airducts which are not significantly more narrow than the exit of the fan itself.
Meaning, for your I3, Ender, etc. the results are generally correct.
This kinda comparison video is so hard to make as there are so many variables. My goal with it was to provide the most relevant data for the most common 3D printers, which is not perfect but at least it is not anecdotal data, you know. I appreciate the in-depth comment, there is always something to learn from those.
Mine use huge blower fans attached to the frame. The air supplied to the hotend through silicone casw, and yes i use liauid cooling.😅
would like to see a cpap fan tested
I run 2x5015 for print cooling and one 4020 for hot end cooling. I’m impressed with this setting. I have a hard modded Ender 3s1 pro.
Build a housing, use clipper, have linear rails on all axis, the z axis is with Pom anti backlash and brass sliders for z wobbling modded. I have a modded breakout board. Installed a bi metal heatbrake, z axis belt tensioner, and a full custom heat bed by a 750 watt 220v silicone heater isolated of course with an 6mm cast and cnc machined aluminum plate on silicone spacers. And a have squash balls feet. My whole electronic is upgraded with safer parts (meanwell power supply, extra fuse for the bed, ssr relay for the bed, some additional fan, cutouts in the botten cover).
That is a lot of mods. So, basically only what has left from Ender is the frame? You should make a name for it because it is nothing like Ender anymore, haha! :)
@@PrintingPerspective I should think about it, I forgot the copper heatblock 😉 Bit i Tell you, the energy saving and consistance of printing is amazing.
I have replaced my original (? RPM.. no datasheet, not even by searching fan part no) 4010 fans ony my SuperRacer to GDSTime 12K RPM ball bearing ones. They blast air, but also loud a.f. :)
Word of note: i tried a Sunon MagLev fan, and it was a catastrophe, the magnetic bearing just cannot deal with high vibration environments. It made horrible noises.
I'm still using it but i filled it with so much grease that it's mostly a sleeve bearing fan now :D
My previous sleeve bearing fan died quickly, but the converted Sunon seems to be holding on. I think double ball bearing fans are the only fans suitable for a high vibration use. So next a Delta fan i suppose :D
Sunon also has another problem, Autorestart feature. It does not play well with PWM, the fan goes crazy. You can either up the PWM frequency, that sort of works but lead to weird noises. I ended up building a Pi filter, but i sort of went a little wild on capacitance, the control delay is hilarious but it works i guess.
What do this Numbers mean? 5015 50mm wide and 15mm outlet or something?
it would be interesting then what kind of fan would be ideal for side cooling for z-axis print bed (as opposed to bolted down printbed/bedslinger)
considering mass isn't a big deal, what would be the sweet spot of airflow vs noise
fabreeko and west3d are what I'm stick with from now on. Both are incredibly good quality and price to performance is great compared to delta.
I bought 2 4020 by mistake so I use those in pair, I agree with everything you said, they are weak, can rattle at high accel, in pair they are ok but not much or at all better than 2 4010
I tried a 4020 fan but it amplifies the noise from the pwm speed control. I found that 2 4010 fans have good cooling and less noise than the 4020. I don't have space for 5015 in my printer.
Now how does airspeed affect print quality? From what I've read the higher the cooling, the better the pieces look, but the lower the layer adhesion. My question would be where is a good point? I usually run a single 5015 fan at full speed with the Petsfang cooler. Is this maybe already too much?
Cooling does not harm layer adhesion at all unless your hotend temperature is too low and you're relying on the previous layer still being hot and soft in order for the next layer's heat to be sufficient for bonding.
@@daliasprints9798 that depends a lot on what material you print. too much cooling can cause serious layer adhesion problems with filaments such as petg or asa. but for pla its usually fine to blast it with as much cooling as you can.
@@2mD For ASA and ABS, cooling to room temp is problematic but if you're using a chamber, full strength cooling to the chamber temperature should be ok. I don't have experience doing this but the folks I know with chambers & Annex printers seem to do fine.
For PETG, I print with my full cooling (over 100 cfm total air flow) and no layer adhesion problems, at 270-290°C.
@@daliasprints9798 my chamber isn't heated ( yet :D ) so it only reaches ~40°C, maybe that's the problem.
and 290°C for petg seems quite high. i usually print somewhere between 235 ( extrudr XPETG ) and 250 ( most other PETGs ). and with that i get layer adhesion problems with more then ~40% partcooling ( 12032 blower fan ). but if the higher temp works for you im gonna give it a try :)
I only ever met 40x10 fans that come stock from Creality and they are noisy AF. I've switched Steve's ones to two axial Gdstme 40x20 with Briss Fang, and it was way quieter all this time. Now I've switched to Creality Sprite and it comes with another set of noisy Creality fans.
I'm considering a mod for two 40x20 again because I have them, but I think the mod I'm looking at requires one axial and one.. the other one, so I'd still have to buy some more. One 50x20 is considered two... I'm not ready for two of them ;)
The most mind blowing thing is that no company yet made a 3D printer oriented to be silent. Because I think that is what most people want as they have a printer in a living room and the noise gets annoying pretty quick.
@@PrintingPerspective Yes, I'd quite like it to be quiet. With my previous fan setup, I could close one sliding door (the printer lives in a wardrobe) and struggle to remember if it's on, with two I couldn't hear it. Now I still can hear it's on with all three doors. No sleeping with it working unless one really likes white noise with a bit of whining 🙈
There are various applications, so the idea will also spread.
Yup. That is why I worded the title - "3D Printer Part Cooling FANS" and tested all the fans in a way that the data would be relevant for this application.
That's a very interesting printhead for KP3s, can you share the link?
Here you go www.printables.com/model/382202-kingroon-kp3s-30-mod-for-e3d-v6-hotend-4010-fan
Have any of you guys tried a 97 33 fan?
How effective is a 5015 as a heatsink fan, if I can't mount it directly? I currently have a 3010 (or 3015, don't remember exactly) fan mounted directly to the heatsink (v6), and I'm wondering if placing the 5015 fan above the heatsink and ducting it would result in quieter operation?
In general blower fans are more efficient than axial at forcing air through small ducts or through the fins of the heatsink. 3010 achieves good cooling results by running it fast and annoyingly loud. I think 5015 as a heatsink fan would be a little bit overkill, but if the silence is your number one goal then yes it could provide better noise levels with better cooling. Just keep in mind that you have to reduce its speed first and the easiest way would be with a resistor. And if we are talking about ducting 5015 fan that is on the printhead then I don't think you need to worry about the performance loss.
@@PrintingPerspective Thanks for the reply! I'm running almost stock cooling setup still (might be a 4010 fan, but no duct)l, I'm planning on running a 5015 for part cooling, not worried about duct losses there, just the heatsink - I've got extremely limited space - it's a corexy, but there's little space from the build platform to the edges -, and a 5015 would fit better than the 4020, for axial fans I think only the 3010 is possible, don't have the distance from heatsink for a bigger fan with duct.
Twin 5015's running 40% seems quieter than the board fan on my little kingroon.
fantastic job, thanks, Number 1
Oh yes, 5015 fans are noisy as hell :)) I am in the middle of redesigning the hotend cooling, thanks for the data, it's useful and i will change my design
I question your anemometer setup. Ideally an anemometer presents no loading to the airflow, it just spins freely. But using only one third of the anemometer opening is making it behave like a turbine with an attached compressor. The open side is going to push/pull ambient air presenting the fans with a much higher loading (pressure drop) than they would be with just a duct alone. This will bias the measurement, and fans with better pressure performance vs airflow will outshine.
Minimally you should block the open inlets of the anemometer so they have less loading.
For relubrication avoid motor oil or any mineral oil in general. Use ATF hydraulic oil or silicone based oils instead. WD40 will ruin a bearing too.
Will the 24v 4010 fan performance be greater than its 12v counterpart and if yes then how much increase?
Dual 5015s FTW
Man do some testing between all different 5015 brands! There seems to be so many (Delta, sunon, th3d, winsinn, creality, MX/topfan, etc)
More known brands like delta, sunon have datasheets so it is very easy to compare fans yourself. Plus sourcing all those would be a nightmare as you can't find them in on place or at least I don't know where. :/
well im a crazy person and i run an omega cooler for my ender 3 v3 se which uses a 80 mm pc fan but i have an adapter and i am usicng a 6015 axial fan
Top notch work!
Oh, to answer your question. I'm running 2x 5015 blowers and 2x 8025 axial fans. 😁 You can see them on my channel.
Nice :) That is definitely a lot of cooling :D
hello can u share stl file for 2 in 1 fan that u use for the test, thx b4
Would be nice to see 5020s. Reportedly they're quieter *and* have better performance than 5015s.
Now when I done the whole video it would. Too bad, I didn't have any on hand and the ordered ones didn't arrive. But when I was looking at the datasheets 5020 fans usually had way lower static pressure compared to 5015's and that is definitely not good if your cooling duct openings are narrow. But then what is better? Wider or narrower duct openings? :D It's like a never ending story, haha!
@@PrintingPerspective Interesting. The ones I've seen are mounted without any complex or or shrinking duct path, for example the Annex usermods Chubby, so indeed they're in a setting where static pressure matters a lot less. But I thought they were comparable. I'll look at the specs on some. Been thinking about changing out my 5015s for 5020s.
I love 50x20 blowers (the ones with intakes on both sides). They're so much quieter for the amount of airflow compared to anything else I've tested, I started incorporating them into my own designs. If you want to design around one, I uploaded a 5020 blower model to thingiverse (same username) some time back. I've been surprised how popular that model has been. Apparently a lot of people like them. :)
When I get the 5020 fan I will test it and post the results on this channel community tab and on Twitter.
as far as i understood partcooling its not about airspeed but airflow. airspeed: the speed of the air in m/s as mentionend in the video. airflow: how much airvolume per second. i want to know about the airflow and not about the airspeed. cause the speed could be very high but the airflow at the same time very low. so this data seems quite useless in my opinion.
You understand that those two things are very closely related, right? Airspeed gives more concentrated cooling. Also higher airspeed gives a higher cooling rate that can be equal to a higher airflow. And the opening size of the ducts on most 3d printers are more orientated to make direct concentrated airflow at higher airspeeds. That provides a more consistent cooling when moving in any printing direction and especialy when printing thin details that needs high cooling rate in a small area. So for you saying that this data is useless clearly shows that you have no understanding of this complex subject.
@@PrintingPerspective OP is right, airspeed is only viable as a proxy for airflow at very low static pressures and that’s exactly what your test is missing.
Without static pressure numbers the tests are pretty much meaningless.
5015s vs 5015 ??
I have seen noctua 4020 server fans put into a duct... I wonder how that would compare.
Server fans usually run at 10K+ RPM, so they probably could match or maybe exceed the performance but noise levels would be insane.
@@PrintingPerspective search printables for Noctua A4x20 see what I am talking about.
People replace server fans with these to cut the noise at home, 5000 rpm, optimized for static pressure, I believe they are rated at 14.9 db, 5.2 CFM. Using it may not be simple, 12v pwm, would be interesting to replicate.
I would love to see how effective different styles of cooling ducts are, not sure how you would judge that though.
NF-A4x20 has 2,26mm H2O static pressure, while 5015 fans at same RPM have ~12mm H2O. If you didn't watched I suggest checking this video where I compared axial vs blower performance - ruclips.net/video/mxb71Ndg6tY/видео.html
@@PrintingPerspectiveyes, that video completely answered my questions... Great job.
It's difficult to find a 5015 blower that doesn't whistle at speed lower than 100%. Often materials like PETG needs 20-40% blower speed, and the whole printing will be with an obnoxious hissing.
You sure it's not your PWM control? You can change the PWM frequency or add a Pi filter.
Great job.
Is a 6028 worth it for having more than 2x the airflow of a 5015? (Also, 2x 5015's weight)
I didn't test the 6028 fan, so unfortunately I can't answer the question.
single delta 5015 on a EVA 2.4.2 toolhead
Sleeve bearings are only silent when new. Hydraulic bearings are better sleeve bearings with a bigger oil reservoir. Ball bearings stay the same for years.
I mean, I agree with you more than not. But for example, I had a 4512 ball-bearing blower fan that ran at ~7500 RPM and lasted not even a year with a relatively small amount of printing.
@@PrintingPerspective sunon makes radial fans that are ok in terms of price and quality.
Like said before, the brand of the fan matters quiet a lot. And most importantly, the design of the fan ducts
Obviously. Because they run at different RPM speeds. That is why I showed graphs with RPM:Performance ratio. Plus unbranded fans from China that almost every printer uses (or you can buy) come with almost identical blade fin designs. There always be variables, but my goal is to minimize them as much as possible and provide the most relevant use-case scenario for a good reference point.
@@PrintingPerspective yes but you should’ve used all same brand fans to have a more accurate comparison between them
They all were MX brand fans that I ordered from Kingroon store on AliExpress...
after running dual 5015s for a few years Im going back to dual 4010s, the noise just isnt worth it.
I mean, did you run them at full speed that noise is the issue? Maybe it is the duct that increases the noise because they affect the noise levels too.
@@PrintingPerspective I was using the hero me duct on an ender 3 and theres really a bunch of problems with the duct itself that made ma not a fan of the design...
but ya, if you have to run the fans at 50% just to be as quiet as the non-silent steppers there really isnt a point (for me) to use the bigger, higher power draw, louder, heavier fans.
Hey thanks for the video I was searching for something like this for a long time, ty for going so in-depth.
I had a question what if I use no nozzle on the front and run a 7530 vs 2x 5015 (price and weight don't matter)
will the dual 5015 still pull ahead with unobstructed flow?
edit: both noise and heat removal are important as they will be used in a room next to the bedroom
interesting cause the voron printers use 4020 fans
Ótimo resultados !
I always feel like the 4020 moves more air than the 5015. Honestly, I hate 5015's. They're too large and I always felt their output sucked but clearly I Was wrong. I really like the 4020's as noise isn't a problem. Not sure why I feel like this. lol
4020 blower could feel that they provide more airflow but another very important aspect when using the fan in a duct is the static pressure. And there is no way that 4020 provides more of it than the 5015 at the same RPM. But I feel you, I thought 4020 would perform better.
I suppose it's completely subjective, but measuring noise level alone doesn't reflect the TONE of the noise. I am much more comfortable with the tone of a 4020 or 5015 fan than the screech of a 4010 fan.
I just installed new 4010 set , BB fans...yes there is a diffrence in RPM between them and Sleeve.There noticeably faster.
Weight is irrelevant. Put the fan on the printer frame and bring the air by a hose to the printhead. Solved.
This accent sounds familiar :)
I'm more of a 6061 or 7075 kind of guy
9GAX0412P3S001 eva
drop mike
Brand/manufacture is the biggest factor to all aspects here unfortunately.
Those appear to be wathai 4020 fans.
There is a new version out more that completely annilate the old ones
You did get a CNC machine, noise should not be that much of a factor.
great vid