You can aggregate that information already with the data at hand. (mmh2o * 3.53) / dBA For real world usage, well, if dBA is equal or lower to the Noctua with higher mmh2o you already have a style that will be better at cooling, given that the cooler would fit into the case. All that aside, the viewers are from all around the world, we appreciate the usage of SI units and not the catering to a minority.
I hope people look at GE jet engines and how they've made the intake way bigger to surround the jet air from ambient atmosphere with more air somehow to break the sound up. It's like air is its own acoustic insulation. I'm pretty sure the bypass air gets heated, and this creates as near a vacuum as possible to surround the fast moving air to reduce the sound waves in most directions.
not surprised the Olimpus sounded like a siren when it was designed after the engine in one of the loudest planes ever made aside from the thunderscreech
I honestly did not think that would work as well is it did, well done! It's amazing that the A12x25, designed by a real company, is being outperformed by so many people!
I've designed, printed, and built a functioning axial compressor/jet engine, but I don't think the Noctua would be able to spin it. I have it driven by an electric motor from an RC car via a 3:1 drive belt setup, and the compressor spins anywhere from 20-30k RPM depending on what fan design I put in it. It's also around 250mm long (or over 400mm with the turbine section attached) so fitment may also be an issue. It's been a long running work in progress, maybe one day I'll finally make a video of it.
I have a design ive been working on for a screw blower type supercharger and i would be afraid to submit it for the same reason, the noctua fan just doesn't have enough power/rpm unless he wants to print and assemble some sort of crazy gearbox 😂
What’s the advantage of using multi-stage air compression for airflow or cooling, besides cutting down the CFM? turbine compression is typically for delivering the right air volume for your combustion situation. Feels like chasing perpetual motion- limits due to energy constraints. 12v density vs petroleum...getting the compressors to spin fast enough for increased airflow would be nearly impossible without a denser energy source.
I feel like the problem with the olympus was the angle of the blades, the first stage looks like the angle of attack would be really high causing severe flow separation and the angle doesn't really look like it changes as it goes through the compressor stages which you would normally expect.
Agreed. The blade pitch of all three stages is way too high. The first stage should be fairly shallow and each of the following stages should be slightly greater. The blade pitch should be set to move about the same volume at each stage. This is a complex design project to do properly.
Alternatively, use gearing to spin them at different speeds, which simplifies the blade disk design and lowers the total drag (if the gearbox drag can be kept low).
yeah I taught that, I doesent Look calculated to me, maybe I could reupload a better version of my multistage fan, it was a catastrofe, but i calculated the speeds and blade angles, season 5 episode 4 btw
This video peaked my interest as I use to be a gas turbine technician in the Navy. Typically, each progressive set of fan blades would get smaller and smaller as each stage would be compressed at a higher pressure. The built ones were essentially the same diameter. But really not a deciding factor on a PC fan. As I assessed this design would be the most inefficient design for a number of factors, higher friction and tolerances were not that great. But the most significant factor…the power to drive three fan blades rather than one, would slow the speed down significantly. Typically an axial flow gas turbine has a high power to weight ratio at the cost of high fuel consumption. As many guessed, the electric motor was not powerful enough to drive such a fan setup.
Involute needs the tips connected to the back of the previous blade with a thin strut. Hybris made me wonder if it didn't just act like it had thick blades. Grain Dryer is just a nice simple high speed fan with no skew and rather small blades. Maybe try a 10 bladed version for more blade area?
Paused at beginning to say: if this is an actual 3-stage compressor with some math in it (:: not just 3 of the same fans, but, sized for actual function), then, yes, absolutely, these ARE the droids I'm looking for!
Too bad the stabilizer didn't stabilize, or that whatever tolerances were out, or both, or whatever :/ wish there was a way to print a good one!! It would be hella sweet!!
Maybe, axial compressors use a ton of energy and are traditionally prefered for lower pressure, higher flow applications. For actually cooling a radiator, I think an axial (with a more ideal design) may do better, but in this test that is only examining static pressure, it is hard to imagine anything beating a centrifugal blower (at least of an a12x25). I think our big issue is that the static pressure test is not a great one for cooling performance. Proposed alternate tests: 1: time to fill an air bladder (maybe a plastic bag) when mounted to a radiator. Or same with a volumetric flow sensor behind the radiator. I think he may have done something similar a few seasons back, but I don't remember.
@@JOEDHIGGINS He used to to do volumetric flow, and then I guess the comments convinced him that static pressure was more important for cooling a season or two ago and he's been trying to figure out how to best measure that ever since.
@@GremlinSciences I think static pressure is important to push air through a radiator, but not if it reduces the volume flow. Measuring only one thing allows cheesing if the fan has a reduced cross section in exchange for higher pressure. Best would be to measure volume at some flow restriction similar to a radiator.
The problem is the fan RPM. Axial compressors are running at 10s of thousands of RPM. The design has to change to be optimised for low speed operation.
I'd like to see him do a season where he uses a pulsed power supply so all fans can be locked to something like 2000 rpm and X watts. And they gain or lose points if they can hit 2000 rpm at a lower wattage. Or, do a season where the fans are run through 2 tests. 2000 rpm (any power) 30W (any speed) He'd have to change up what he's using to drive them probably, but they make industrial cabinet fans in 120 and 140 with sufficient motors.
And actually have size limitations. Maybe a class for "in-frame" (i.e. fits inside the standard frame) and an exo-frame class where fans can be anywhere in an additional area.
All these designs are really pleasing to look at and feel like they should perform quite well with some adjustments. Involute reminds me a lot of how Arctic fans look, and I bet would perform REALLY well with a more rigid material or slightly thicker blades.
Looks like they were the opposite direction to me, the green blades were all rotor blades, the stators were white. The reason it didn't "impress" is probably mostly due to the fact that type of design requires a lot of power to work well. It requires high air flow, which requires a lot of power.
I'm here for the jet engine fan but gotta say for once the add was spot on, they chose the right guy which opinion we might trust with the right audience that might be nerded into buying the thing, ifixit should sponsor you harder, sure I'm not the only math is cooler than math guy that now wants one, who else here loves to go to the cables, adapters and electronics section?
If I can make a suggestion .. Maybe add a ruler or a yardstick to the side of the smoke test set up so we can measure how far the smoke is pushed. It would give a nice indicator of the actual differences in airflow.
The olympus was not carefully modeled, or the creator did not understand the purpose of the stators. That smoke in the smoke test had a lot of rotation indicating a lot of wasted power.
They have enthusiasm but not expertise, not that I'm any better in that regard, I respect the attempt, would be awesome to see them give it another go with some help/research
I got invested in the sponsor segment because I need a new soldering iron. I love it when sponsored segments are "look at this thing, it's a good thing, here why!" Instead of rote script reading
@@geekswithfeet9137 I love it when channels use sponsor parts, because it really helps them and i don't have to see them, because I use sponsorblock like a normal person. It is 2min 31sec btw
Not to steal ifixit's thunder, but take a look at pinecil. I got it as a small backup soldering iron, but it ended up becoming the main iron i use. It's made mostly for precise work, but with wide tip and a good usb-c charger you can put down close to 100w of heat if you need to.
Can designers utilize a common bearing size? Like say a 608 bearings? That would allow much better complex compressor designs, and a standard bearing size would make James's job easier. It would be awesome to see a well engineered axial compressor with bearings and tested tolerances
Maybe for a future season you can measure both CFM and static pressure, or just mass flow? Also, it would be interesting to see what some of the lower-RPM designs that have been on the show can do with Noctua's A14 industrialPPC-3000 frame and motor?
If you have to add a bearing, remove the seals and remove thick grease and change to very little thin oil or thin grease, keep the seals off to reduce drag. Ideally you could do this with no bearing added. Shallower angle on the blades should increase rpm and static pressure.
large greased rubber shielded bearings have a lot of friction btw. (for a fan) If you pull the shields out (just a pick or something will pull rubber shields out) then wash out the grease and replace it with a light oil like sewing machine oil it'll have way way less friction. For the ultimate you can run them dry but it's for show more than practical as they will rust and things. Metal shielded bearings can be no contact and you can with some effort swap the grease for oil in light load applications so it's not really an unfair test overall to degrease and oil.
How much material is too much for that motor? How or when is to much? Maybe a video idea like 5 of the same basic fans with different infill and Thickness and see how much is too much or too little.
I'd love to see how the Involute went if a couple mm high rim joined the tips. Or even better if it joined the tips and formed a slight velocity stack with a snug fit around the frame.
I love how through 6 seasons Wes still get that awesome Mattie MaGuire track during the smoke test. Less a couple episodes late in season 2 or early season 3.
Grain drier: Looks simple, easy to maintain; performs above average. Not surprised, to be honest; that seems exactly like the kind of optimisation you'd see on a farm.
The Axial fan needs Bearings to stop any form of play in the main shaft which will also need to be true, so that the tolerances between the Shroud, Compressor, and Stator blades can be maintained, also the blades will need to be balanced to stop any wobble. The smaller the gap between the shroud compressors and stators the better the performance will be. Be Aware: Deformation and catastrophic failure of Plastic compressor and stator blades is likely to happen at high RPM.. There is a reason these types of things are made of metal.
No big surprises there. Weight is not much of an issue for speed, as long as you're in the range that the bearing can handle without issue. It's an issue for acceleration, of the fan, so saving a few grams won't help the performance much, unless you're measuring how fast the fan spins up. The blades on the involute was also "scooped backwards" near the center. Arctic P series have a similar blade design, but they have those details right.
i always enjoy when i catch one of these videos. I know my low number of videos watched on your channel doesn't lend me a leg to stand on, but nevertheless, i'd like to suggest a season using one of those ridiculously powerful server fans that spin at like 5k RPM lol. I'd look at Delta, they make great ones. maybe you could test some of the previous season winners to see what RPM each of them does best at.
I want to see someone design and test a fan that has the same gear technology that the Pratt and Whitney GTF turbofan engine has on an airbus jet. Just an idea
What I took out of the RC hobby is, that electric turbine style fans are cool and in some cases have a unique wooosh sound but!! In case of watt, ampere usage to thrust these are no good because of the amount of blades which the motor needs to push through the air. From efficiency standpoint what at least o now these are no competition to normal propeller.
Re the Rolls Royce Olympus engine, later models were also used on the Concorde. It differed from a lot of engines at the time in that there were two compressor spools - the high pressure compressors and turbines ran on a separate spool to the low pressure compressors / turbines, allowing them to run at speeds more suited to the low / high pressure conditions.
Interesting note about high performance fans, sound can change how much you're willing to deal with when everything else stays the same except the chassis. Case in point, I've been using 5300RPM fans for years and other than at full song, they were perfectly bearable for peace of mind when doing simulations and big renders. I upgraded to a significantly larger GPU recently, which meant getting back into a full tower case. Yeah, it lasted four days until the noise became unbearable and I ended up swapping all but my exhaust to 3600rpm fans. It runs about 10C warmer but still within the safe zone under load. I never thought I'd be walking backwards on headroom for static pressure, but sanity has a cost when it's sitting next to you for 12+ hours. 😁 As soon as I finish my current project, I'll sit down and finally get a fan piece banged out for this challenge. I've had one started forever but paying jobs and R&D work take precedent unfortunately.
Just excited to be here in the first wave. Jet fans are always interesting. I wanna see you try and cool a water block with gradually more ludicrous stuff, like with a Honeywell T55 engine. I hear it's quite capable of spinning some fans.
Someone needs to design a squirrel cage type of fan. Furnaces use them and have quite a bit more static pressure than that. Like a pull through fan that mounts a blower wheel off the back of the fan motor and attach the housing to the backside of the fan frame surrounding the blower wheel. Basically a one side inlet only squirrel cage/furnace blower.
I worked on 200mw gas turbines which have axial compressors. Here is the challenge. Each blade is an airfoil (like an air plane wing) to work properly it has to have the proper air flow (not stall). They are very sensitive and requiere very tight clearances. A 12" blade only has 50/1000" clearance. Then this repeats on each stage, with each stage having a different shape and blade count. When blades stall it produces alot of turbulence and vibration. Thats probably whats going on with this fans.
idea for s7: get a strong motor that can power nearly any design at a set rpm(1000) for ex, then go for temp on an aircooler. alternatively a slim fan contest would also be cool
If you want to design a fan with extremely thin blades like the Involute you should really include a ring touching the furthest part of the fan tip. This would give it a bit more rigidity allowing it to spin without touching the rim of the fan. An like said in the video the closer you can get the blades to the rim of the fan without touching the better. Noctua spent a lot of time developing the material of their fans to make them rigid enough that they could make them run extremely close to the housing of the fan without risking that the blades would flex enough to touch it. An alternative that some have tried is to have ring connecting the tip of the fan blades that acts like a rim.
Not an engineer, but looking at the sharp angle on the olympus fan it looked optimised for speed not static pressure (makes sense after which it is designed). Higher static pressure = lower airflow, I think a fan that's going to produce pressure should only use the 2cm of outer fan, since that part of the fan spins fastest and thus creates most pressure.
Yes, bearings between each stage. A single bearing would have too high a load as well as balancing issues. Very much like when balancing a tire with lead weights.
in my mind the Involute's fan blades snapped off and scattered around the room during the smoke test, but the footage and music continued like nothing happened. it was an amusing thought
6:00 I have Artic P14 fans in my PC currently and had coolermaster fans before that, both have the same kind of blade shape but the coolermaster ones were significantly louder in my experience.
The RR Olympus engine was also used in the Concorde supersonic passenger plane! The fan design here was facing far too much friction, perhaps from the internal bushings?, and what seemed like far too much air load on the blades, perhaps a lower blade pitch would work better?
I know this season is all about static pressure, but why don't you check airflow as well? Many of these fans (like the Involute), while very low on the pressure scale, seem to have very high CFM - maybe higher than the top rated ones from season one
Oh wow, James, interesting that this is a USB-powered soldering iron! OK, so when it's plugged into is battery pack, those knobs control something about its power. What about when you plug it into a regular USB charger plug or a power strip with USB in it (a computer, for all we know)? I mean what is it set on by default because you don't have those knobs when you plug it into a different USB jack?
One of the main complaints about the Ifixit soldering iron is the lack of controls on the iron itself. You basically have to use it with their own battery bank if you want full functionality. The Pinecil, Fnirsi irons, and other USB irons all offer controls and a screen on the iron itself and for a much lower price.
@@guycxz: Hmm, yeah, I can see the expense being a problem, but as for just using it without controls, that wouldn't bother me much because the only soldering irons I've ever had only had zero or one control: 15/30W; most of mine haven't even had that. Why have I bought several? Because tips don't last that long and replacement tips are hard to find, so just replacing the iron after a while works when whole irons are very cheap.
plug it into a comp, set the temp via serial command or ifixit site [which fires the serial command anyway] then plug it in your other usb-c PD power source, it remembers the temp it was last set to
This might actually be the first youtube ad spot to work on me xd I want the fix hub for working on audio in my cars and i'll probably ask my wife to get it for me for the upcoming holiday lol
Just noticed the Sawblade fan from S6E4 has the static pressure in the dBA column, it should say 46. Surely someone else has already mentioned this, but just in case no one has :)
What will win this challenge is a vacuum impeller design. Nobody's really done it yet, but the first one to take apart their dyson and scale up that impeller is going to take home the W.
I designed "The Wonder From Downunder" which was the first Compressor fan with a really big win. I have another design entered which easily beats that and the current first place. I hope he wants the record broken because it will be something to see, I'm talking almost double the current winner.
The involute needs to be designed such that the air pressure it creates counteracts the centrifugal force so that there's a balancing centripetal force to keep it together.
My RUclips has been messing up a lot lately. In the middle of videos it will randomly change to a completely different video. With your video, it started playing at 2x speed and I was horrified because you were talking FAST 🤣🤣🤣 changed it back to normal and video played just fine lol. Great video even at 2x speed!
5:05 Arctic has fans that look kind like this for example ACFAN00121A. Noiseblocker has a Fan that has 3 mmH²O NB-eloop X B12x-P. Maybe someone uses it as inspiration. I know it isn't as good as the Top. But it is in a 120x120x25 space.
Id like to see an unlimited category of the fan show down where the only rules are it has to fit in certain physical dimensions. I wanna see stuff powered with whatever motor the designer wants and really crazy stuff.
Involute fan needs aredesign. Run it at rpm and see how much it deflects. Compensate for that in the original at rest shape so when the fan is spun up it achieves it's final working shape that is the desired shape. Result is super low weight fan blades and intended airflow.
G'day Major & Cooper, The Involute by Barak looks like the Design that CoolerMaster used years ago for the Nepton AIO Fans plus the AMD Wraith Spire & Stealth stock coolers which I think are also made by Coolermaster & the Arctic P12, none of which Expand at speed to rub the Frame.
That support bearing was killing the RPMs of the fan. Remove those tight grease seals, wash out the grease with IPA, lightly oil, and give it another go. Probably won't be amazing but I would expect you would see a noted improvement.
3:04 The air/gas exhausting the thrust cone of a turbine engine is at or near the same ambient pressure of the air around the engine. The air Pressure is increased internally returns to ambient in the tail cone.
How about doing a best of the best series? Take say top 5 designs from each Swanson and run them against each other through all the tests? That would give a true champion of champion deeign. 😊
7:00 I mean they're not too bad even today, though of course they're waaaayyyyy more limited in scope than they used to be. Rolls-Royce is still one of the top contenders especially in jet engines, but are also pretty high up in aerospace, defense and energy departments too.
Do you think it would be possible to "disqualify" the fans that are clearly designed for bigger numbers (for example have allowable build area, inside the fan and the fan "dangerzone" maybe so something like a cone wouldn't be enough to DNF but WFDU and centrifugal madness is out) and colour their placement in the sheet, it would be the perfect compromise for the competitive peeps who still care about useable pc fan and peeps who just like to see number go up, you could accept the designs that will dnf and still have them on the list with static pressure in descending order. Also links in the description if people mention that design is available somewhere?
If I remember it correctly then compressor style fans have layers that rotate at different speeds and have different pitch angles. So a si gle motor without any gearbox will not create the compression necessary for it to work.
Maybe you're thinking of bypass engines like used in airliners? I wanted to design an axial compressor after the last episode so I dug into it a bit. Most of the designs I read about had one solid shaft that all the blades attached to. Pressure is created by reducing volume and blade pitch between stators like you said. The engineering involved gets complicated enough that I just couldn't dedicate the time to learning.
Correct! I would not say different speeds as all rotate at the same speed. However, several stages of compressor blade are rotateable. That is to prevent compressor stall at different speeds. Not a make or break design for a PC fan.
What ya need to do is find out how to make the air cold and then pull it into the fan and then heat it (by cooling the pc parts) and treat the cold and hot stages accordingly with their volume.
Oki... great video, but now... I need to get myself an ifixit soldering station! I still use an old Weller station, and it's..... well, it's not portable. :D I've been using gas pen solders for that, with various degree of sucsess.
I think that axial fan needs to not use the whole diameter of the fan area, if it was much smaller, the motor might actually be able to spin it fast enough with reduced mass flow of course
I was lucky enough or maybe unlucky? enough to see and hear the Vulcan bomber fly ,,, in 40 some years its still the loudest thing ive ever heard , and i live by Cape Canaveral
I'd like to someday see the top 10 fans from each season go head to head in all tests to crown the over all top 5 best fans. I'm not actually all that interested in the "cheater" style fans as they don't really practically work in a pc, but I am interested in the standard sized ones.
![Seungmin "그렇게, 천천히, 우리(As we are)" | [Stray Kids : SKZ-PLAYER]](http://i.ytimg.com/vi/kAzmhLHePqU/mqdefault.jpg)
Next season should be a "CFM per dB" competition to see who can make the most air for the least noise for real world PC use.
You can aggregate that information already with the data at hand. (mmh2o * 3.53) / dBA
For real world usage, well, if dBA is equal or lower to the Noctua with higher mmh2o you already have a style that will be better at cooling, given that the cooler would fit into the case. All that aside, the viewers are from all around the world, we appreciate the usage of SI units and not the catering to a minority.
ah yes, the power/weight ratio of the fan world
I hope people look at GE jet engines and how they've made the intake way bigger to surround the jet air from ambient atmosphere with more air somehow to break the sound up. It's like air is its own acoustic insulation.
I'm pretty sure the bypass air gets heated, and this creates as near a vacuum as possible to surround the fast moving air to reduce the sound waves in most directions.
No no no. The most air with the MOST noise. I wanna hear them scream.
@@fallingwater
You enjoy the fans torturing the air? What kind of fan club you got there?
not surprised the Olimpus sounded like a siren when it was designed after the engine in one of the loudest planes ever made aside from the thunderscreech
To be fair, it made the noise because it was rubbing.
The design probably had the same amount of stator blades as compressor blades, which will make it sound like a siren
@@Skracken You can clearly hear that the blades are rubbing. It's as simple as that in this case.
@@mytube001 No that's a different sound.
@@Skracken If that's what you want to believe, I won't stop you.
Hey! The Grain Dryer was my fan! I had no idea it would beat the A12x25. That was my goal
Congratulations!
Great job!
I honestly did not think that would work as well is it did, well done! It's amazing that the A12x25, designed by a real company, is being outperformed by so many people!
@AHSEN. Yeah I literally threw it together in about 2-3hrs. I plan on making a V2 that is of a better dryer fan design
simple and effective
I've designed, printed, and built a functioning axial compressor/jet engine, but I don't think the Noctua would be able to spin it. I have it driven by an electric motor from an RC car via a 3:1 drive belt setup, and the compressor spins anywhere from 20-30k RPM depending on what fan design I put in it. It's also around 250mm long (or over 400mm with the turbine section attached) so fitment may also be an issue. It's been a long running work in progress, maybe one day I'll finally make a video of it.
Would be cool if he made a noctua compatible 12v motor with more amps. Throw random blades on it and see what happens.
The hub size constrains what you can do quite a lot. But if 12V is out the window we can go for 48 or 220V and make a very insane fan.
I have a design ive been working on for a screw blower type supercharger and i would be afraid to submit it for the same reason, the noctua fan just doesn't have enough power/rpm unless he wants to print and assemble some sort of crazy gearbox 😂
What’s the advantage of using multi-stage air compression for airflow or cooling, besides cutting down the CFM? turbine compression is typically for delivering the right air volume for your combustion situation. Feels like chasing perpetual motion- limits due to energy constraints. 12v density vs petroleum...getting the compressors to spin fast enough for increased airflow would be nearly impossible without a denser energy source.
I feel like the problem with the olympus was the angle of the blades, the first stage looks like the angle of attack would be really high causing severe flow separation and the angle doesn't really look like it changes as it goes through the compressor stages which you would normally expect.
Agreed. The blade pitch of all three stages is way too high. The first stage should be fairly shallow and each of the following stages should be slightly greater. The blade pitch should be set to move about the same volume at each stage. This is a complex design project to do properly.
Alternatively, use gearing to spin them at different speeds, which simplifies the blade disk design and lowers the total drag (if the gearbox drag can be kept low).
there is a function for that @@wingsounds13
yeah I taught that, I doesent Look calculated to me, maybe I could reupload a better version of my multistage fan, it was a catastrofe, but i calculated the speeds and blade angles, season 5 episode 4 btw
That kind of design just won't be able to outperform the leaders in this type of testing. Not enough motor torque and too low rpm.
This video peaked my interest as I use to be a gas turbine technician in the Navy. Typically, each progressive set of fan blades would get smaller and smaller as each stage would be compressed at a higher pressure. The built ones were essentially the same diameter. But really not a deciding factor on a PC fan. As I assessed this design would be the most inefficient design for a number of factors, higher friction and tolerances were not that great. But the most significant factor…the power to drive three fan blades rather than one, would slow the speed down significantly. Typically an axial flow gas turbine has a high power to weight ratio at the cost of high fuel consumption. As many guessed, the electric motor was not powerful enough to drive such a fan setup.
*piqued
Invelloop looks 98% like the Arctic P-12 fan 👏🏼
looks like a noise blocker a little
Yeah Cooler Master has been using this design for many years too, I have one of their Nepton AIOs that uses them
5:52 Arctic cooling would like a word.
The same goes for Cooler Master and their Master Fan. I still have the one that came off of my Hyper 212 Black as backup fan.
i was always curious how a P12 would perform since its considered a static pressure fan
@rayzian for the price it's amazing.
@@rayzian My thoughts also. Or the BioniX P120.
Involute needs the tips connected to the back of the previous blade with a thin strut. Hybris made me wonder if it didn't just act like it had thick blades. Grain Dryer is just a nice simple high speed fan with no skew and rather small blades. Maybe try a 10 bladed version for more blade area?
Paused at beginning to say: if this is an actual 3-stage compressor with some math in it (:: not just 3 of the same fans, but, sized for actual function), then, yes, absolutely, these ARE the droids I'm looking for!
Too bad the stabilizer didn't stabilize, or that whatever tolerances were out, or both, or whatever :/ wish there was a way to print a good one!! It would be hella sweet!!
it CAN work. the problem is blade deformation and an underpowered motor.
I'd like to see all the blades supported on bearings as a modification
Maybe, axial compressors use a ton of energy and are traditionally prefered for lower pressure, higher flow applications.
For actually cooling a radiator, I think an axial (with a more ideal design) may do better, but in this test that is only examining static pressure, it is hard to imagine anything beating a centrifugal blower (at least of an a12x25).
I think our big issue is that the static pressure test is not a great one for cooling performance.
Proposed alternate tests:
1: time to fill an air bladder (maybe a plastic bag) when mounted to a radiator. Or same with a volumetric flow sensor behind the radiator.
I think he may have done something similar a few seasons back, but I don't remember.
@@JOEDHIGGINS He used to to do volumetric flow, and then I guess the comments convinced him that static pressure was more important for cooling a season or two ago and he's been trying to figure out how to best measure that ever since.
@@GremlinSciences I think static pressure is important to push air through a radiator, but not if it reduces the volume flow. Measuring only one thing allows cheesing if the fan has a reduced cross section in exchange for higher pressure. Best would be to measure volume at some flow restriction similar to a radiator.
The problem is the fan RPM. Axial compressors are running at 10s of thousands of RPM. The design has to change to be optimised for low speed operation.
I'd like to see him do a season where he uses a pulsed power supply so all fans can be locked to something like 2000 rpm and X watts. And they gain or lose points if they can hit 2000 rpm at a lower wattage.
Or, do a season where the fans are run through 2 tests.
2000 rpm (any power)
30W (any speed)
He'd have to change up what he's using to drive them probably, but they make industrial cabinet fans in 120 and 140 with sufficient motors.
And actually have size limitations.
Maybe a class for "in-frame" (i.e. fits inside the standard frame) and an exo-frame class where fans can be anywhere in an additional area.
High AOA blades with leading edge slats would probably get the job done.
All these designs are really pleasing to look at and feel like they should perform quite well with some adjustments. Involute reminds me a lot of how Arctic fans look, and I bet would perform REALLY well with a more rigid material or slightly thicker blades.
Nice up to @speeed reference
They are brothers
With all the craziness in the world, thanks for something both fun and informative
The stator blades in a jet engine are angled the opposite way to the compressor blades, thats why the Olympus isnt that impressive.
Looks like they were the opposite direction to me, the green blades were all rotor blades, the stators were white. The reason it didn't "impress" is probably mostly due to the fact that type of design requires a lot of power to work well. It requires high air flow, which requires a lot of power.
You're not going to get meaningful compression with a fan spinning that slowly, so stators are just extra impedance to airflow.
I'm here for the jet engine fan but gotta say for once the add was spot on, they chose the right guy which opinion we might trust with the right audience that might be nerded into buying the thing, ifixit should sponsor you harder, sure I'm not the only math is cooler than math guy that now wants one, who else here loves to go to the cables, adapters and electronics section?
If I can make a suggestion .. Maybe add a ruler or a yardstick to the side of the smoke test set up so we can measure how far the smoke is pushed. It would give a nice indicator of the actual differences in airflow.
The olympus was not carefully modeled, or the creator did not understand the purpose of the stators. That smoke in the smoke test had a lot of rotation indicating a lot of wasted power.
That is important, if your goal is to create thrust, not so much if you just want static pressure.
They have enthusiasm but not expertise, not that I'm any better in that regard, I respect the attempt, would be awesome to see them give it another go with some help/research
I got invested in the sponsor segment because I need a new soldering iron. I love it when sponsored segments are "look at this thing, it's a good thing, here why!" Instead of rote script reading
3 minutes long at the very start of the video is a bit much tho
@@geekswithfeet9137 You can fast forward videos on YT
@@geekswithfeet9137 I love it when channels use sponsor parts, because it really helps them and i don't have to see them, because I use sponsorblock like a normal person.
It is 2min 31sec btw
Not to steal ifixit's thunder, but take a look at pinecil. I got it as a small backup soldering iron, but it ended up becoming the main iron i use. It's made mostly for precise work, but with wide tip and a good usb-c charger you can put down close to 100w of heat if you need to.
@@DrathVader Both of them are top notch.
Can designers utilize a common bearing size? Like say a 608 bearings? That would allow much better complex compressor designs, and a standard bearing size would make James's job easier.
It would be awesome to see a well engineered axial compressor with bearings and tested tolerances
Maybe for a future season you can measure both CFM and static pressure, or just mass flow? Also, it would be interesting to see what some of the lower-RPM designs that have been on the show can do with Noctua's A14 industrialPPC-3000 frame and motor?
If you have to add a bearing, remove the seals and remove thick grease and change to very little thin oil or thin grease, keep the seals off to reduce drag. Ideally you could do this with no bearing added. Shallower angle on the blades should increase rpm and static pressure.
large greased rubber shielded bearings have a lot of friction btw. (for a fan) If you pull the shields out (just a pick or something will pull rubber shields out) then wash out the grease and replace it with a light oil like sewing machine oil it'll have way way less friction. For the ultimate you can run them dry but it's for show more than practical as they will rust and things. Metal shielded bearings can be no contact and you can with some effort swap the grease for oil in light load applications so it's not really an unfair test overall to degrease and oil.
Thanks for your service on this Veterans Day. Fellow AF vet here.
How much material is too much for that motor? How or when is to much? Maybe a video idea like 5 of the same basic fans with different infill and Thickness and see how much is too much or too little.
I'd love to see how the Involute went if a couple mm high rim joined the tips. Or even better if it joined the tips and formed a slight velocity stack with a snug fit around the frame.
I love how through 6 seasons Wes still get that awesome Mattie MaGuire track during the smoke test.
Less a couple episodes late in season 2 or early season 3.
@1:54 The soldering on that wire was painful to watch.
Oh yes. Terrible.
Grain drier: Looks simple, easy to maintain; performs above average.
Not surprised, to be honest; that seems exactly like the kind of optimisation you'd see on a farm.
The Axial fan needs Bearings to stop any form of play in the main shaft which will also need to be true, so that the tolerances between the Shroud, Compressor, and Stator blades can be maintained, also the blades will need to be balanced to stop any wobble. The smaller the gap between the shroud compressors and stators the better the performance will be. Be Aware: Deformation and catastrophic failure of Plastic compressor and stator blades is likely to happen at high RPM.. There is a reason these types of things are made of metal.
No big surprises there. Weight is not much of an issue for speed, as long as you're in the range that the bearing can handle without issue. It's an issue for acceleration, of the fan, so saving a few grams won't help the performance much, unless you're measuring how fast the fan spins up. The blades on the involute was also "scooped backwards" near the center. Arctic P series have a similar blade design, but they have those details right.
The problem with this season is that people don't understand that it's not about moving air, it's about maintaining pressure inside.
I lost this channel like a year or 2 ago and just now found it again lol so glad your still going
i always enjoy when i catch one of these videos. I know my low number of videos watched on your channel doesn't lend me a leg to stand on, but nevertheless, i'd like to suggest a season using one of those ridiculously powerful server fans that spin at like 5k RPM lol. I'd look at Delta, they make great ones. maybe you could test some of the previous season winners to see what RPM each of them does best at.
I want to see someone design and test a fan that has the same gear technology that the Pratt and Whitney GTF turbofan engine has on an airbus jet. Just an idea
What I took out of the RC hobby is, that electric turbine style fans are cool and in some cases have a unique wooosh sound but!! In case of watt, ampere usage to thrust these are no good because of the amount of blades which the motor needs to push through the air. From efficiency standpoint what at least o now these are no competition to normal propeller.
Re the Rolls Royce Olympus engine, later models were also used on the Concorde. It differed from a lot of engines at the time in that there were two compressor spools - the high pressure compressors and turbines ran on a separate spool to the low pressure compressors / turbines, allowing them to run at speeds more suited to the low / high pressure conditions.
Only the second 2 spool engine ever produced.
Interesting note about high performance fans, sound can change how much you're willing to deal with when everything else stays the same except the chassis. Case in point, I've been using 5300RPM fans for years and other than at full song, they were perfectly bearable for peace of mind when doing simulations and big renders. I upgraded to a significantly larger GPU recently, which meant getting back into a full tower case.
Yeah, it lasted four days until the noise became unbearable and I ended up swapping all but my exhaust to 3600rpm fans. It runs about 10C warmer but still within the safe zone under load. I never thought I'd be walking backwards on headroom for static pressure, but sanity has a cost when it's sitting next to you for 12+ hours. 😁
As soon as I finish my current project, I'll sit down and finally get a fan piece banged out for this challenge. I've had one started forever but paying jobs and R&D work take precedent unfortunately.
Just excited to be here in the first wave.
Jet fans are always interesting. I wanna see you try and cool a water block with gradually more ludicrous stuff, like with a Honeywell T55 engine.
I hear it's quite capable of spinning some fans.
doesn't the olympus need fewer fan blades at the front and more towards the exit to increase pressure?
Someone needs to design a squirrel cage type of fan. Furnaces use them and have quite a bit more static pressure than that. Like a pull through fan that mounts a blower wheel off the back of the fan motor and attach the housing to the backside of the fan frame surrounding the blower wheel. Basically a one side inlet only squirrel cage/furnace blower.
I worked on 200mw gas turbines which have axial compressors. Here is the challenge. Each blade is an airfoil (like an air plane wing) to work properly it has to have the proper air flow (not stall). They are very sensitive and requiere very tight clearances. A 12" blade only has 50/1000" clearance. Then this repeats on each stage, with each stage having a different shape and blade count. When blades stall it produces alot of turbulence and vibration. Thats probably whats going on with this fans.
"Something weve all seen...grain dryer"
Nope, didnt know that existed until this video.
idea for s7: get a strong motor that can power nearly any design at a set rpm(1000) for ex, then go for temp on an aircooler. alternatively a slim fan contest would also be cool
If you want to design a fan with extremely thin blades like the Involute you should really include a ring touching the furthest part of the fan tip. This would give it a bit more rigidity allowing it to spin without touching the rim of the fan. An like said in the video the closer you can get the blades to the rim of the fan without touching the better. Noctua spent a lot of time developing the material of their fans to make them rigid enough that they could make them run extremely close to the housing of the fan without risking that the blades would flex enough to touch it. An alternative that some have tried is to have ring connecting the tip of the fan blades that acts like a rim.
Thank you for your service, and happy Veterans Day to all who served this great nation!
Not an engineer, but looking at the sharp angle on the olympus fan it looked optimised for speed not static pressure (makes sense after which it is designed).
Higher static pressure = lower airflow, I think a fan that's going to produce pressure should only use the 2cm of outer fan, since that part of the fan spins fastest and thus creates most pressure.
I think the 3 stage need support between each stages … without support it’s won’t work.
Yes, bearings between each stage. A single bearing would have too high a load as well as balancing issues. Very much like when balancing a tire with lead weights.
in my mind the Involute's fan blades snapped off and scattered around the room during the smoke test, but the footage and music continued like nothing happened. it was an amusing thought
6:00 I have Artic P14 fans in my PC currently and had coolermaster fans before that, both have the same kind of blade shape but the coolermaster ones were significantly louder in my experience.
The RR Olympus engine was also used in the Concorde supersonic passenger plane! The fan design here was facing far too much friction, perhaps from the internal bushings?, and what seemed like far too much air load on the blades, perhaps a lower blade pitch would work better?
The smoke test showed the Olympus was out of balance as well, which I'm sure didn't help.
I know this season is all about static pressure, but why don't you check airflow as well? Many of these fans (like the Involute), while very low on the pressure scale, seem to have very high CFM - maybe higher than the top rated ones from season one
Oh wow, James, interesting that this is a USB-powered soldering iron! OK, so when it's plugged into is battery pack, those knobs control something about its power. What about when you plug it into a regular USB charger plug or a power strip with USB in it (a computer, for all we know)? I mean what is it set on by default because you don't have those knobs when you plug it into a different USB jack?
One of the main complaints about the Ifixit soldering iron is the lack of controls on the iron itself. You basically have to use it with their own battery bank if you want full functionality.
The Pinecil, Fnirsi irons, and other USB irons all offer controls and a screen on the iron itself and for a much lower price.
@@guycxz: Hmm, yeah, I can see the expense being a problem, but as for just using it without controls, that wouldn't bother me much because the only soldering irons I've ever had only had zero or one control: 15/30W; most of mine haven't even had that.
Why have I bought several? Because tips don't last that long and replacement tips are hard to find, so just replacing the iron after a while works when whole irons are very cheap.
plug it into a comp, set the temp via serial command or ifixit site [which fires the serial command anyway] then plug it in your other usb-c PD power source, it remembers the temp it was last set to
that involute needs a ring at the tips to keep them inline and give it another shot
great stuff as always - keep up the good work !
This might actually be the first youtube ad spot to work on me xd I want the fix hub for working on audio in my cars and i'll probably ask my wife to get it for me for the upcoming holiday lol
Just noticed the Sawblade fan from S6E4 has the static pressure in the dBA column, it should say 46.
Surely someone else has already mentioned this, but just in case no one has :)
What will win this challenge is a vacuum impeller design. Nobody's really done it yet, but the first one to take apart their dyson and scale up that impeller is going to take home the W.
I designed "The Wonder From Downunder" which was the first Compressor fan with a really big win. I have another design entered which easily beats that and the current first place. I hope he wants the record broken because it will be something to see, I'm talking almost double the current winner.
@TheRattleSnake3145 hell yeah brother
Editor went crazy on this one 🤣🤣🤣🤣🤣🤣🤣🔥🔥
The involute needs to be designed such that the air pressure it creates counteracts the centrifugal force so that there's a balancing centripetal force to keep it together.
My RUclips has been messing up a lot lately. In the middle of videos it will randomly change to a completely different video. With your video, it started playing at 2x speed and I was horrified because you were talking FAST 🤣🤣🤣 changed it back to normal and video played just fine lol. Great video even at 2x speed!
The Involute looks like the Arctic P12... just in green. An awesome fan for its low price by the way.
5:05 Arctic has fans that look kind like this for example ACFAN00121A. Noiseblocker has a Fan that has 3 mmH²O NB-eloop X B12x-P. Maybe someone uses it as inspiration. I know it isn't as good as the Top. But it is in a 120x120x25 space.
2:30
What is the reparability score on this thing?
Id like to see an unlimited category of the fan show down where the only rules are it has to fit in certain physical dimensions. I wanna see stuff powered with whatever motor the designer wants and really crazy stuff.
I've been toying with this idea of stacked planetary fans, possibly using a planetary gear....
Involute fan needs aredesign. Run it at rpm and see how much it deflects. Compensate for that in the original at rest shape so when the fan is spun up it achieves it's final working shape that is the desired shape. Result is super low weight fan blades and intended airflow.
I have an idea: sandwich 2 fans toghether (back to back, counter rotation) and try differet blade designs to get optimal air flow
that fix hub is pretty sick looking. not usually intrigued by ads lol
It was bound to happen, glad to see somone made a jet engine
G'day Major & Cooper,
The Involute by Barak looks like the Design that CoolerMaster used years ago for the Nepton AIO Fans plus the AMD Wraith Spire & Stealth stock coolers which I think are also made by Coolermaster & the Arctic P12, none of which Expand at speed to rub the Frame.
Would love to see commercially available fans on the show down, a la project farm
Logical next step is an a12x25 into a fully functional gas turbine!
the sawblade still has that error on the dBA scale, it says 2.6 but it should be 46 as seen in the video that it is from
Involute: Perhaps a stabilising outer ring like the arctic P12 Max would help stabilise it?
Very common fan blade design on the involute, dating back about 10 years. Cooler master, Arctic. Surprised that you thought it was rare.
That support bearing was killing the RPMs of the fan. Remove those tight grease seals, wash out the grease with IPA, lightly oil, and give it another go. Probably won't be amazing but I would expect you would see a noted improvement.
Where's Rubin's turbo fan compressor from episode 7? That thing didn't rub.
3:04 The air/gas exhausting the thrust cone of a turbine engine is at or near the same ambient pressure of the air around the engine. The air Pressure is increased internally returns to ambient in the tail cone.
I love the fan showdown so much 🥰
Need to take the seals and lubricant out of that bearing. Maybe some very light oil, but the seals and grease will cause a bunch of drag
Doesn't the involute feature on the Arctic P12/14 and the CM Silencio 120?
How about doing a best of the best series? Take say top 5 designs from each Swanson and run them against each other through all the tests? That would give a true champion of champion deeign. 😊
7:00 I mean they're not too bad even today, though of course they're waaaayyyyy more limited in scope than they used to be. Rolls-Royce is still one of the top contenders especially in jet engines, but are also pretty high up in aerospace, defense and energy departments too.
11:05 You have a typo on your spreadsheet: Sawblade (16th) has its static pressure duplicated as its dBA. I just checked the S6E4 video and it was 46.
Do you think it would be possible to "disqualify" the fans that are clearly designed for bigger numbers (for example have allowable build area, inside the fan and the fan "dangerzone" maybe so something like a cone wouldn't be enough to DNF but WFDU and centrifugal madness is out) and colour their placement in the sheet, it would be the perfect compromise for the competitive peeps who still care about useable pc fan and peeps who just like to see number go up, you could accept the designs that will dnf and still have them on the list with static pressure in descending order. Also links in the description if people mention that design is available somewhere?
If I remember it correctly then compressor style fans have layers that rotate at different speeds and have different pitch angles. So a si gle motor without any gearbox will not create the compression necessary for it to work.
Maybe you're thinking of bypass engines like used in airliners? I wanted to design an axial compressor after the last episode so I dug into it a bit. Most of the designs I read about had one solid shaft that all the blades attached to. Pressure is created by reducing volume and blade pitch between stators like you said. The engineering involved gets complicated enough that I just couldn't dedicate the time to learning.
Correct! I would not say different speeds as all rotate at the same speed. However, several stages of compressor blade are rotateable. That is to prevent compressor stall at different speeds. Not a make or break design for a PC fan.
What ya need to do is find out how to make the air cold and then pull it into the fan and then heat it (by cooling the pc parts) and treat the cold and hot stages accordingly with their volume.
Oki... great video, but now... I need to get myself an ifixit soldering station! I still use an old Weller station, and it's..... well, it's not portable. :D I've been using gas pen solders for that, with various degree of sucsess.
It is possible to put sand paper on the walls and then make it spin until it stop rubbing?
The Olympus has way to much pitch on the blades.
I think that axial fan needs to not use the whole diameter of the fan area, if it was much smaller, the motor might actually be able to spin it fast enough with reduced mass flow of course
That note that you hear from the Olympus is commonly known as the blue note in the aviation industry.
Isn't involute shape used by Arctic in their P-series fans?
Are the designs for the fans available anywhere?
Hey, I just checked S6E4, your data table has an error. The Sawblade (pos 16) had a dBA reading of 46, not 2.6
I hope there will be a second attempt with a fixed Involute design that connects the blade tips with a thin ring.
I was lucky enough or maybe unlucky? enough to see and hear the Vulcan bomber fly ,,, in 40 some years its still the loudest thing ive ever heard , and i live by Cape Canaveral
I'd like to someday see the top 10 fans from each season go head to head in all tests to crown the over all top 5 best fans. I'm not actually all that interested in the "cheater" style fans as they don't really practically work in a pc, but I am interested in the standard sized ones.
I'm gonna need you to start every video like that