You can make a decent dynamic pressure rig. Think ping pong ball in a tube with a pull spring attachment and very tiny vent holes in the tube as the spring and ball stretch from the air pressure. You will also get to see dynamic pressure oscillation if the system is sensitive enough. A. Static pressure test is not a good indication of working flow and it's also a little boring.
I disagree, since they publish their stats we would only be testing the manufacturer's trustworthiness. I think the noctua gives a decent enough baseline that allows us to test submitted designs, which is why I watch this channel.
@@StereoTyp0 lol no, it would just be used to compare with the fans in his tests. We want to see how they compare not only to noctua but the 3d printed fans.
Please! I run a thick 360 rad and right now run super fucking loud fans (Arctic p12 max) because none of the other fans I had pushed enough air and I'd really like to run fans that don't sound like a jet engine at 100% 😂
Basically that. Also, it's not static pressure optimized, it's dynamic pressure optimized - it's meant to maintain a good pressure while blowing air through a radiator, not when blowing air into a sealed duct.
Also- it's a "best for all uses" type of fan. It was not specifically designed for static pressure, but as a "good for most uses" type of fan that's also quiet. Many many years ago you had to choose a static pressure fan design or an airflow fan design.
@@BReal-10EC ok but all of the 3d printed fans shown here are ALSO "best for all uses" since they beat the a12 in airflow too. The only thing the a12 has over them is quietness
@@ulamss5 It's true, though. If you don't care about noise, just get some Delta fans. It would be better if he actually included a noise normalized test.
I hope you do one episode with previous season winners. I'd love to see what kind of pressure the Cheater can manage. I don't see any major issues with the testing setup - though I'd like to point out that the A12x25 is *not* optimized for static pressure - it's specifically optimized for good pressure while maintaining airflow, such as blowing through a radiator. Look up the Noctua marketing for the A12x25 - the P/Q curve diagram in particular. (Also, if P/Q curve doesn't ring a bell, you should probably read up on that in preparation for season 6. It's highly relevant.)
He has done something similar pretty recently. The more accurate the data on a season for the actual benefit from the printed fan, the more likely he will do a run of old winners with that new data gathering.
I used to think I had a really good immune system before I had a kid. Turns out I was just really good at avoiding people who sneeze and cough directly onto my eyeballs.
If you're interested in a Noctua fan designed more for static pressure, you should check out the NF-F12 industrial series. Their PPC-3000 PWM claims 7.63 mm H2O, albeit at 43.5 dB.
I have 2 NF F12 120mm fans on my Bitmain S9 BTC miner. They are half the RPM as the server fans that were on this machine and 2.4A less than them as well and I have them on at 1500RPM keeping my underclocked miner at low temps. They are great fans for the price at $27 each. They are noisy though at 3000RPM but not as bad as other 3000rpm fans.
The other bit of mojo with the Noctua industrial fans is that they're rated for 5 years, 24/7 operation at 100% speed. Completely different ballgame than your standard desktop fan.
I currently run 4 of these in a push-pull on a blackice dual row 240... most the time, I have the pulls turned off... but when I need/want it, it will move some serious air.
true! the T30 in particular could lift itself and hover on a flat surface so he also needs to fasten the fan mounting bracket down with said wing nuts.
@@kitame6991 definitely, I just said make it heavy because that would be faster than having to do each wing nut by hand. But do something, especially for static testing.... Or at least smoke test it 😅
How about putting some wings on the fan plates and the wind tunnel's flange, and then using some small spring-based hand clamps to keep the mating surfaces firmly together? You'd get a nice airtight seal (with an O-ring), but still be able to swap them out quickly without messing with screws and nuts.
Nice to see static pressure being measured correctly. :) For what it's worth, the A12x25 is an extremely weak fan regarding static pressure performance; the only fan I've measured that performs at the same level is the Toughfan 12. Every other fan I own is stronger. As a tip to anyone designing a fan for optimal static pressure performance, don't use slender, forward-skewed blades like the A12x25; keep the blades straight and broad like flower petals.
Love your videos, suggested your channel in another comment, but it looks like youtube ate it. I'm curious though - I feel like thicker fans usually do better for static pressure, would it make sense to make a thicker 'blade section' that protrudes beyond the 25mm body? It feels right to my brain, but I'll defer to your expertise there.
Thing is, designing a fan for static pressure makes no sense outside of extremely rare applications. Noctua has some nice interviews on that. You also don't want to design a fan for airflow only (no restrictions). For pc fans, designing them for the highest airflow when restricted by a high density rad is one end, and a low density air cooler on the other end.
@@thorwaldjohanson2526 i never even considered that, i was just thinking in terms of pressure and noise together (dont know the terminology but in my terms pressure as in better airflow i guess)
4:39 There is a difference between the pressure with and without airflow. In any real life application, you would seek flowing air through something flow restricting, not blowing up a ballon😉 TLDR, leave a hole at the end or cover the end up with woven fabric or felt.
I love this, this is fun, but just to be clear the maximum static pressure a fan is capable of has basically no impact on it's ability to remove heat from a radiator or a tower cooler. But it is nice to know if you need to fill a balloon with air 🙂 Can't wait to seed the designs, my money are on a blower stile fan or maybe a compressor if anyone have the skills to pull that off 🙂
The beauty of the A12x25 is that at a given _sound level_ it's great compared to other fans. It's not particularly pressure-y, but it's quiet and the noise it makes is pleasant. You also don't need _that_ much static pressure performance to push air through a radiator, performance-wise that a12x25 is good even at a lower SP. You may want a phanteks T30 (edit: yep), or a Noctua "industrial" fan (the ones that spin faster) for performance! In general faster spinny = more pressure.
ok because of the vertical orientation multi fans with staters will be easier to pull off and have some crazy static pressure especially if somebody finds a way to make counter-rotating blades nested within staters, thats the strat
So I heard everybody here say they want you to standardise the speed of the fans for loudness dB and then test for air movement. 👌 You should get on that, don't let us down Major!
id say good enough for me using industry standard value and you matched it it doesnt need to be perfect it needs to be repeated and comparable and i think this nails it
You can still do the smoke tests. They were never done on the tester anyways, so it's pretty easy to do the smoke tests and then throw them on the new tester.
Glad to finally see a proper static pressure setup, congratulations on getting more and more scientific about the fans! Maybe that is ne non-american in me speaking, but is there any reason you are using mmH20 instead of pascals as a unit for the pressure? I get that many fan manufacturers give their static pressure rating in this unit, but I would encourage you to use pascals instead of mmH20, as it not only offers more precision (1 mmH2O = 9.80665 pascals (Pa)) leading to a higher resolution in the fan ratings, but is widely accepted abroad the scientific community. Looking forward and keep up the great grind
Agreed. I've been doing static pressure testing using a magnehelic gauge which has a very high level of accuracy and can measure fractions of a pascal, which is sometimes needed when measuring the pressure of PC fans - if there is any airflow at all, the pressure drops a fair bit.
I would say don't change units. I understand that in a general sense pascals are used in the scientific community but this isn't for the scientific community. If you working within an industry you stay within the common units for that industry. As for the higher resolution you'd have to see how the actual device measures. If it displays one decimal while using mmH20 i would expect it to display no decimals for pascals leading to essentially no difference since the ratio is extremely close to 1:10.
Id suggest adding a 10mm hole so you can test static pressure without anywhere for air to go, and static pressure with airflow. A fast squirrel cage with a tiny output could produce very high static pressure, but little airflow, making it useless as a fan.
In that case, better to use an adjustable grate to map the entire P/Q curve. I think that would settle the debate between fans designed for high air flow vs high static pressure. By seeing the entire curve, we can clearly see where each fan lies in terms of tradeoffs, and if any fan design does actually dominate the other across the curve.
@@reverse_engineered I put in a comment about using changeable orifices to test pressure with different levels of flow but it got buried. Some fans will achieve higher pressure but there will be no flow, but others might be slightly less pressure but substantial flow
That’s a great meter I hope you got the pitot tube too. I’m guessing you’ll try to make fan curves on these at some point? Find a good test for static pressure with some flow? I’m sure you know but this test is just the stall pressure. Fans can also surge (flow changes direction around the blades) at stall and lose a lot of pressure. Some decisions make more pressure with a little flow.
It looks like your setup is solid. That said, as long as you use the same setup between fans, the results will be relative to each other, and can be ranked accordingly.
This looks to me like a valid measurement of static pressure matching the season goal. I suggest a procedure to check for leaks when assembling this for a day, although it likely doesn't actually matter for sane fan designs (difference between true static and almost static)
Really want to see rpm and noise readings on the pressure test rig. Most likely a static pressure test will favor blades with more angle of attack as all fans are likely to slow down relative to the air flow tests, but we won't know if we don't measure it
measuring pressure this way with no air flow has little to no meaning, and since the fan has no flow you can be assured that its blades are stalled, generating less "static pressure" than what it could generate with some airflow. You should just measure the fan efficiency given motor electric power and measuring the total pressure rise across it. Or create a manually (electronically) variable orifice at the end of the tube and find the optimal operating point (in terms of airflow) for the fan to reach maximum static pressure, look up a fan characteristic curve.
When you do the charts definitely weight the noise heavily against the static pressure. That way it limits people trying to just brute force the highest static pressure to get to the top of the leaderboard.
Nothing wrong with that. Only for Noctua fanboys It seem. Fans aren't make to be quiet but to cool. Get over it and just buy what you need, without force your personal preferences to others. Bye.
If you're not going to use the wing nuts to secure the gray mounting plate to the tube, I think you should ensure it's *relatively* extremely heavy to ensure a well performing fan doesn't artificially reduce its own score by allowing some air to escape.
My suggestion is to have dual tests. Static pressure and airflow. Can easily have a fan that moves lots of air with no pressure, or one that makes lots of pressure without airflow. Both of which I would classify as a "useful" fan for cooling sake.
Once I looted some Server Fans out of the electronics trash of our University. They had small diameter metal fans, ball bearings, extreme angle of attack and 10.000 rpm. They went to 70dB, but boy did they move air.
Looks like a really good setup, I follow your logic and it seems like the tester lines up exactly with the manufacturer data so it seems like it's good :)
you could also just push water up a clear tube, preferably angled at 45° to increase measurement precision. this set-up would definitely look interesting for the videos
RPM helps static pressure alot, because the gap in the blades effectively becomes smaller, allowing the fan to pump for longer, assuming the blades don't stall.
Let me get this straight, they have more airflow AND more static pressure than the A12x25 ? And the only downside is that they are marginally less silent? Which you know given some of the setups we see thrown around in the comments is negligable due to the casing since standard cooling and noise tend to go out of the window with a lot of those aforementioned setups? Yeah I'd buy these things many of them have a not as harsh sound profile either so I certainly can't and wont complain!
I think the mounting plate should be screwed, or fastened however, the positive pressure inside will try to escape wherever it can, and the fan helps it by generating an upward force on the plate. A gasket would be nice, if not overkill.
That fancy Fluke 922 meter you have there is quite the device. Should you read the literature that came with it in regards to it's functions, you will find that the measurement range is +/- 16mm of H20 with a resolution of .001. (info found with a quick search) You need to properly seal the end of the tube where you shoved it into your stand. Proper sealing will yeild reuslts that are more accurate. Also, to increase the accuracy of the results, you should leave each fan on for a specified time, ie. 30 or 60 seconds. Always ensure the fan is properly fastened before testing. Remember - repeatable results require rigorous testing. That meter can also measure back pressure, so turn the fan around so it is sucking and the measurement will be through the meter connetion only. This is another reason to make sure all your 3D printed press fit fittings are properly sealed, no matter how 'snugly' it feels. Air is small. Really small. Silicone sealent will work just fine for that. Use the silicone to seal a proper fitting for your flexible measuring hose. Properly conducted experimentation will yeild better results and improve your ability to increase your reputation in the community. While I initially started watching because of an obtuse interest in the strange facination that people have with everyday objects, I kept watching because I saw your method change and was intrigued as to where this might go. This obtuse interest often leads to personal revelations of interesting science and in general interesting people from all over. Purchasing that Fluke meter (I have a 34 year old Fluke multimeter, I remember how much it was!) showed that you are either serious or making too much money. I believe it is the former, however true the latter is. Bravo in your endevours and I shall be watching some of your back catalouge out of sincere interest.
For anyone who is watching and doesn’t have the propper context. What he is referring to at the beginning of the video is Rotavirus. Rotavirus is incredibly nasty it's over fairly quickly, but it takes you a while to recover. It affects all people differently, but it was quite literally one of the worst sicknesses I had had since COVID. I spent roughly 7 1/2 hours throwing out not being able to keep fluid or solid down. Pray that you don't get Rotavirus. If the correct pro cautions are not taken when you have it you could be hospitalized for dehydration amongst other things. It has been sweeping the entire upper half of the United States the past few weeks. Also, this is different from the virus that causes you to have a non-stop cough for 35 days because of nasal drainage going down the back of your throat and irritating your throat. That is a whole other virus that has been devastating in the same area.
I'd like to suggest using the other piece of tube as a second test with an automotive style mass airflow (MAF) sensor, pulling a raw ohm-resistance reading .
If it's possible, please test the following retail fans for additional points of comparison: 1. Thermalright TL-B12 (cheapest of the "big hub" fans at around 10 USD, and a rather potent "sleeper brand" offering great performance at low cost) 2. Scythe Grand Tornado (their successor to the Gentle Typhoon, and the second cheapest "big hub" fan at around 20 USD) 3. Nidec Gentle Typhoon (I believe you tested it, or at least mentioned it, once against the A12x25 in an earlier season) 4. be Quiet Silent Wings 4 Pro (a similar "luxury" big hub fan that competes against the A12x25 in performance and noise) 5. Thermaltake ToughFan Pro (designed specifically to try and leave no gaps between the fan frame and radiator frame) 6. SuperFlower MEGACOOL (think T30s on steroids)
Now you gotta do a stat card for each fan! Weight ☆☆ RPM ☆☆☆ Airflow ☆☆☆ Static Pressure ☆☆☆☆ But with the values and maybe 0-100% bars. Just like with engines where RPM, Horsepower, Max Speed, Acceleration and Torque are relevant stats that make them good for different applications. Showcasing the strenghts and weaknesses and which ones are jacks of all trades would make for interesting graphics and analysis, and would also let more designs shine.
I like this, it seems more repeatable than the door. Also, the A12x25 blades are thin and more numerous, making me think it is optimized for higher airflow, rather than pressure. It would probably beat the others you showed in an airflow metric, because that's just a tradeoff with noise and pressure. Anyways, I like this idea for Season 6.
It would be very good to have airflow speed/cfm just for reference and comparison. Seems like that would help sort of categorise the "purely static pressure" with the practicality of a pc fan that needs both
Dude, right there with you on the gift of endless illness from the little one. I've had a shocking run of illness since my little girl arrived 5 months ago. Worth it.
A slack tube manometer is easy, fast, and maybe even free in you have some transparent tubing, a ruler, and some water. I do have to concede that in your case a digital display is better for your application though.
Need gaskets or seals on the capped end and the big square piece the fans are mounted to. Just make sure no leakage, even though I think it would be minimum.
Smaller blades equal less pressure. Any fans I've bought for static pressure (for radiator use) always have wider and fewer blades. Looking forward to see what the season brings!
That Fluke 922 is the meter I suggested in your previous video. It's a good mix of accuracy, granularity, and price. There's even more accurate micromanometers from Alnor and Shortridge, but you're looking at a few thousand dollars. However, if you bought the meter with the kit, you should be using the pitot tube that came with it for measuring static pressure. You'll want to use the side port that connects to the perimeter holes, not the center one.
Max Flow vs Max pressure is a very different task. Pressure is 'maximise unit energy per molecule' => fewer molecules Flow is 'maximise molecules per energy unit' => lower pressure.
That test retest reliability looks really good. Fluke is also my meter of choice and there are no parts to wear during repeated testing so I'm liking this test stand setup. Looking forward to season 6. As for kids, I always dreaded the start of school because we knew they were going to bring home something fun. Good luck.
Thanks for the effort put into this! Unfortunately, with static pressure test and open flow test you're now only get the far ends of the fan curve and missing out on all the real life practical performance data in the middle, which would actually be really cool to see! You'd want to take air and pressure data points at different airflows and chart them to show you where that particular fan shines and where it isn't ideal. It's a lot of work! If you have a logging setup to your PC you could gradually close an orifice and log the data, which would be ideal. You'd probably also want a mass air flow sensor instead of an anemometer, but now we're really getting scope creep :D
Two suggestions: 1. Mount a fitting to attach the sensing tube. 2. Run a normalised test at 1000 rpm. 3. Split the bunch into categories a) Stock: fan disk must not exceed stock fan disk dimensions. b) Mod: fan disk is allowed to exceed stock fan disk dimensions, aerodynamic parts are not allowed. c) Unlimited: as above but aerodynamic parts allowed.
I think that while static pressure testing is cool and should be done I think there should be some airflow allowed out for it to be more "realistic" and so fans that move very little air but have really good static pressure still have to flow air
Static pressure can't be measured while allowing airflow, the point of static pressure is how much a fan can push without allowing air to flow. I'm expecting that designs which seal the gap in the duct would work better
Noctua openly states that they design for maximim airflow AS WELL AS being as quiet as possible, also being balanced and durable. Static pressure is just one metric. Ideally you would test, CFM/Static Pressure/Volume together.
The A12x25 doesn't actually have very high static pressure compared to other commercially available PC fans. The A12x25 offers high airflow at low noise, but in terms of static pressure, I expect a lot of submitted fans to beat it. For static pressure, you could also use the Corsair ML120, Thermaltake ToughFan 12 Pro, Phanteks PH-F120MP or EK Vardar Evo 120 as a benchmark.
I like this setup a lot. With that said, six seasons in and we're still waiting for two classes separating fans that fully fit in the frame, and an "unlimited class" that can be the ones with add on parts that exceed the frame.
i feel your pain man, i think i had the same thing about 2 weeks ago, it was absolutely brutal. I had no idea something like that could come out of my digestive system.
id like to see all the new fans designed for the back fan. it would allow some of the heavy fans to preform that have failed due to lack of rpm in the previous seasons. right now you have to sacrifice some airflow design for rpm to compete.
The one thing I would add is "performance at a specific noise level". You can often just keep adding speed and get more performance, some of us want the best performance without killing our eardrums. Pick a specific noise level thats maybe noticeable but not uncomfortable and adjust the speed of the fans to match that noise for a second data point. Some people just want maximum cooling, other people want good cooling with fan noise being controlled. Why can't we test both?
A suggestion for better comparison: log the rpm of the fan and compare pressure between fans at the same rpm. Rationale: the pressure build up is directly proportional to the speed of the fan. The speed will depend on how much work it does moving air (which it won’t in this test) and the weight and air friction of the fan wheel since the motor only have so much torque. This means that a light fan with few blades will spin faster than a heavier fan with many blades and thus, it will look like the light fan performs better in the test. The T30 has the rally high rpm, and thus easily wins the competition.. Love the tests though! :D
I'd love to keep including the measured dB along with the other metrics. For my environment, I would prioritize a quiet fan over one 10% stronger but 40% noisier.
Some 10 years ago corsair used to do AF and SP fans, AF had more blades but they were skinny where SP had less blades but their surface area was much bigger. These days most fans is somewhat a combination of both types into one blade design.
Nice setup. Only thing I might add would be a gasket/sealant where the tube goes into the base. Seems like a good fit now, but I assume it's probably not air tight.
![[#2024MAMA] G-DRAGON - 무제(Untitled, 2014)+POWER+HOME SWEET HOME+뱅뱅뱅+FANTASTIC BABY | Mnet 241123 방송](http://i.ytimg.com/vi/Ox29z5Nf1Uk/mqdefault.jpg)
Play World of Tanks here: tanks.ly/3uadktR
Noctua fans suck, they are using a design that is at lest 2 decades old.
@@administrator4728 Wow what a load of bs. Any explanation for that wildly inaccurate claim?
@@administrator4728 cool story bro
You can make a decent dynamic pressure rig. Think ping pong ball in a tube with a pull spring attachment and very tiny vent holes in the tube as the spring and ball stretch from the air pressure. You will also get to see dynamic pressure oscillation if the system is sensitive enough. A. Static pressure test is not a good indication of working flow and it's also a little boring.
@@triedproven9908 yep
You gott test more retail fans during Season six submissions too.
I agree with this statement.
I disagree, since they publish their stats we would only be testing the manufacturer's trustworthiness.
I think the noctua gives a decent enough baseline that allows us to test submitted designs, which is why I watch this channel.
@@StereoTyp0 lol no, it would just be used to compare with the fans in his tests. We want to see how they compare not only to noctua but the 3d printed fans.
@@StereoTyp0 The people demand to KNOW what fan performs best in all relavant testing scenarios!
Please! I run a thick 360 rad and right now run super fucking loud fans (Arctic p12 max) because none of the other fans I had pushed enough air and I'd really like to run fans that don't sound like a jet engine at 100% 😂
Pretty easy to explain why the A12x25 came last. It's quieter. It's not hard to make a strong fan. It's hard to make one, that is also quiet.
Basically that. Also, it's not static pressure optimized, it's dynamic pressure optimized - it's meant to maintain a good pressure while blowing air through a radiator, not when blowing air into a sealed duct.
Also- it's a "best for all uses" type of fan. It was not specifically designed for static pressure, but as a "good for most uses" type of fan that's also quiet. Many many years ago you had to choose a static pressure fan design or an airflow fan design.
@@BReal-10EC ok but all of the 3d printed fans shown here are ALSO "best for all uses" since they beat the a12 in airflow too. The only thing the a12 has over them is quietness
Lol the amount of cope by the noctua fanboys. Get over it, they're not standout amazing fans anymore since like 2016.
@@ulamss5 It's true, though. If you don't care about noise, just get some Delta fans. It would be better if he actually included a noise normalized test.
I hope you do one episode with previous season winners. I'd love to see what kind of pressure the Cheater can manage.
I don't see any major issues with the testing setup - though I'd like to point out that the A12x25 is *not* optimized for static pressure - it's specifically optimized for good pressure while maintaining airflow, such as blowing through a radiator. Look up the Noctua marketing for the A12x25 - the P/Q curve diagram in particular. (Also, if P/Q curve doesn't ring a bell, you should probably read up on that in preparation for season 6. It's highly relevant.)
This needs more likes!
@@iansabrewolfe Agreed!
Champion of Champions
He has done something similar pretty recently. The more accurate the data on a season for the actual benefit from the printed fan, the more likely he will do a run of old winners with that new data gathering.
I used to think I had a really good immune system before I had a kid. Turns out I was just really good at avoiding people who sneeze and cough directly onto my eyeballs.
"really good at avoiding people who sneeze and cough directly onto my eyeballs"
You're a snot-dodger!
Matched Noctua spec is all setup validation I need to see. Really noice.
That's my thought too
I would also do a smoke test to make sure you don't have any leaks. Your static pressure measurement will be HEAVILY influenced by air leakage.
Provided they’re all tested on the same setup, all results are comparable irrespective of leakage
If you're interested in a Noctua fan designed more for static pressure, you should check out the NF-F12 industrial series. Their PPC-3000 PWM claims 7.63 mm H2O, albeit at 43.5 dB.
I have 2 NF F12 120mm fans on my Bitmain S9 BTC miner. They are half the RPM as the server fans that were on this machine and 2.4A less than them as well and I have them on at 1500RPM keeping my underclocked miner at low temps. They are great fans for the price at $27 each. They are noisy though at 3000RPM but not as bad as other 3000rpm fans.
Those are LOUD. I switched them pretty quickly for eloops
The other bit of mojo with the Noctua industrial fans is that they're rated for 5 years, 24/7 operation at 100% speed. Completely different ballgame than your standard desktop fan.
I currently run 4 of these in a push-pull on a blackice dual row 240... most the time, I have the pulls turned off... but when I need/want it, it will move some serious air.
Seriously one of my favorite tech channels out right now, the fun he has with it and the community engagement is seriously awesome
same 👍
"No matter how tall you go, its fine." Wait is that a challenge?
I'm waiting for a 12-stage axial compressor entry called "The tower of power".
Also waiting for someone to Rise to the occasion er challenge
Put a soft gasket on the mating surfaces to limit and homogenize air leakage. Maybe make the cap and fan mount heavier too.
true!
the T30 in particular could lift itself and hover on a flat surface so he also needs to fasten the fan mounting bracket down with said wing nuts.
@@kitame6991 definitely, I just said make it heavy because that would be faster than having to do each wing nut by hand. But do something, especially for static testing.... Or at least smoke test it 😅
How about putting some wings on the fan plates and the wind tunnel's flange, and then using some small spring-based hand clamps to keep the mating surfaces firmly together? You'd get a nice airtight seal (with an O-ring), but still be able to swap them out quickly without messing with screws and nuts.
just use 4 small spring clamps, 1 on each side of the base flange that sits on top of the sealed test duct
That and just tighten some wing nuts. Why make it heavier
Nice to see static pressure being measured correctly. :) For what it's worth, the A12x25 is an extremely weak fan regarding static pressure performance; the only fan I've measured that performs at the same level is the Toughfan 12. Every other fan I own is stronger. As a tip to anyone designing a fan for optimal static pressure performance, don't use slender, forward-skewed blades like the A12x25; keep the blades straight and broad like flower petals.
I was wondering if I was gonna see you here LOL. Love your testing!
Love your videos, suggested your channel in another comment, but it looks like youtube ate it. I'm curious though - I feel like thicker fans usually do better for static pressure, would it make sense to make a thicker 'blade section' that protrudes beyond the 25mm body? It feels right to my brain, but I'll defer to your expertise there.
Thing is, designing a fan for static pressure makes no sense outside of extremely rare applications. Noctua has some nice interviews on that. You also don't want to design a fan for airflow only (no restrictions). For pc fans, designing them for the highest airflow when restricted by a high density rad is one end, and a low density air cooler on the other end.
oh 100%, but we're not talking about designing for the real world, but for the horse race that is this channel. @@thorwaldjohanson2526
@@thorwaldjohanson2526 i never even considered that, i was just thinking in terms of pressure and noise together
(dont know the terminology but in my terms pressure as in better airflow i guess)
legit test rig. might have been fun to make a rig that actually pushed water up a tube, but the meter is way easier to read.
4:39 There is a difference between the pressure with and without airflow. In any real life application, you would seek flowing air through something flow restricting, not blowing up a ballon😉
TLDR, leave a hole at the end or cover the end up with woven fabric or felt.
DO THIS FOR BATHROOM VENTS 🎉 THIS IS IMPORTANT INFO MOST CONSUMERS WOULD NEVER SEE
I love this, this is fun, but just to be clear the maximum static pressure a fan is capable of has basically no impact on it's ability to remove heat from a radiator or a tower cooler.
But it is nice to know if you need to fill a balloon with air 🙂
Can't wait to seed the designs, my money are on a blower stile fan or maybe a compressor if anyone have the skills to pull that off 🙂
The beauty of the A12x25 is that at a given _sound level_ it's great compared to other fans. It's not particularly pressure-y, but it's quiet and the noise it makes is pleasant. You also don't need _that_ much static pressure performance to push air through a radiator, performance-wise that a12x25 is good even at a lower SP.
You may want a phanteks T30 (edit: yep), or a Noctua "industrial" fan (the ones that spin faster) for performance! In general faster spinny = more pressure.
You made a great call by going with precision. I’m excited for this season.
Lol at the start that looks exactly like a gigantic aeropress coffee maker!
for the whole video my head was like:
manometer, düp düüp düdü düp,
manometer, düp düdüp düp..
I was wondering if someone already made this reference. Glad I wasn't the only one.
This should be added every time he says it
LOL
ok because of the vertical orientation multi fans with staters will be easier to pull off and have some crazy static pressure especially if somebody finds a way to make counter-rotating blades nested within staters, thats the strat
Feel like we will see alot of turbo inspired fans
So I heard everybody here say they want you to standardise the speed of the fans for loudness dB and then test for air movement. 👌 You should get on that, don't let us down Major!
Static pressure will produce really unpractical designs for any real world use but will be interesting to watch
Not true. You want to pick a fan for its static pressure rather than just airflow for water cooling rads
Your test rig wasn't flush to the ground for every fan test - check out the wire it's balanced on at 8:26
id say good enough for me using industry standard value and you matched it it doesnt need to be perfect it needs to be repeated and comparable and i think this nails it
RIP smoke test b-rolls. The t30 states the static pressure as 7.11mm/h20, so your setup is pretty accurate.
I'd still like to see the smoke b rolls as usual!
Pretty sure they'll still be there, just this one is fans that already have smoke b-roll and it's purpose is to talk about the setup.
You can still do the smoke tests. They were never done on the tester anyways, so it's pretty easy to do the smoke tests and then throw them on the new tester.
Glad to finally see a proper static pressure setup, congratulations on getting more and more scientific about the fans! Maybe that is ne non-american in me speaking, but is there any reason you are using mmH20 instead of pascals as a unit for the pressure? I get that many fan manufacturers give their static pressure rating in this unit, but I would encourage you to use pascals instead of mmH20, as it not only offers more precision (1 mmH2O = 9.80665 pascals (Pa)) leading to a higher resolution in the fan ratings, but is widely accepted abroad the scientific community. Looking forward and keep up the great grind
Agreed. I've been doing static pressure testing using a magnehelic gauge which has a very high level of accuracy and can measure fractions of a pascal, which is sometimes needed when measuring the pressure of PC fans - if there is any airflow at all, the pressure drops a fair bit.
I would say don't change units. I understand that in a general sense pascals are used in the scientific community but this isn't for the scientific community. If you working within an industry you stay within the common units for that industry. As for the higher resolution you'd have to see how the actual device measures. If it displays one decimal while using mmH20 i would expect it to display no decimals for pascals leading to essentially no difference since the ratio is extremely close to 1:10.
I just looked it up the datasheet for the tool he is using accuracy is based on 1 pascal or .1mmH20 so no difference in resolution.
You should test for pressure and vacuum, you know just flip the fan around
Best sanity check would be to do a season 1 style computer temperature test and see how well that correlates with the static pressure these got
Id suggest adding a 10mm hole so you can test static pressure without anywhere for air to go, and static pressure with airflow. A fast squirrel cage with a tiny output could produce very high static pressure, but little airflow, making it useless as a fan.
In that case, better to use an adjustable grate to map the entire P/Q curve. I think that would settle the debate between fans designed for high air flow vs high static pressure. By seeing the entire curve, we can clearly see where each fan lies in terms of tradeoffs, and if any fan design does actually dominate the other across the curve.
@@reverse_engineered I put in a comment about using changeable orifices to test pressure with different levels of flow but it got buried. Some fans will achieve higher pressure but there will be no flow, but others might be slightly less pressure but substantial flow
That’s a great meter I hope you got the pitot tube too. I’m guessing you’ll try to make fan curves on these at some point? Find a good test for static pressure with some flow?
I’m sure you know but this test is just the stall pressure. Fans can also surge (flow changes direction around the blades) at stall and lose a lot of pressure. Some decisions make more pressure with a little flow.
Reproducing the production specs is real good for a testing of a setup. It gives you a know variable as a reference.
It looks like your setup is solid.
That said, as long as you use the same setup between fans, the results will be relative to each other, and can be ranked accordingly.
I like this setup, any thoughts on flipping the fan and blowing air the opposite way as well? (Push vs Pull performance)
This looks to me like a valid measurement of static pressure matching the season goal. I suggest a procedure to check for leaks when assembling this for a day, although it likely doesn't actually matter for sane fan designs (difference between true static and almost static)
Really want to see rpm and noise readings on the pressure test rig.
Most likely a static pressure test will favor blades with more angle of attack as all fans are likely to slow down relative to the air flow tests, but we won't know if we don't measure it
measuring pressure this way with no air flow has little to no meaning, and since the fan has no flow you can be assured that its blades are stalled, generating less "static pressure" than what it could generate with some airflow. You should just measure the fan efficiency given motor electric power and measuring the total pressure rise across it. Or create a manually (electronically) variable orifice at the end of the tube and find the optimal operating point (in terms of airflow) for the fan to reach maximum static pressure, look up a fan characteristic curve.
When you do the charts definitely weight the noise heavily against the static pressure. That way it limits people trying to just brute force the highest static pressure to get to the top of the leaderboard.
Nothing wrong with that. Only for Noctua fanboys It seem. Fans aren't make to be quiet but to cool. Get over it and just buy what you need, without force your personal preferences to others. Bye.
If you're not going to use the wing nuts to secure the gray mounting plate to the tube, I think you should ensure it's *relatively* extremely heavy to ensure a well performing fan doesn't artificially reduce its own score by allowing some air to escape.
My suggestion is to have dual tests.
Static pressure and airflow. Can easily have a fan that moves lots of air with no pressure, or one that makes lots of pressure without airflow. Both of which I would classify as a "useful" fan for cooling sake.
Immediate thought. Probably need to seal the end piece. Blue-tac? Simple rubber gasket? Just to make sure there is no leakage.
Once I looted some Server Fans out of the electronics trash of our University. They had small diameter metal fans, ball bearings, extreme angle of attack and 10.000 rpm. They went to 70dB, but boy did they move air.
Looks like a really good setup, I follow your logic and it seems like the tester lines up exactly with the manufacturer data so it seems like it's good :)
Matching the Noctua spec with the test setup is a great sign. I think this looks like a valid way to test, to me.
0:25 Isn't that when you pass the 30 years mark?
you could also just push water up a clear tube, preferably angled at 45° to increase measurement precision.
this set-up would definitely look interesting for the videos
Crazy to see that with the T30 fan, a 1000rpm boost in speed more than doubles the pressure output seen at 2000rpm. Intriguing stuff, keep it up!
RPM helps static pressure alot, because the gap in the blades effectively becomes smaller, allowing the fan to pump for longer, assuming the blades don't stall.
Let me get this straight, they have more airflow AND more static pressure than the A12x25 ? And the only downside is that they are marginally less silent? Which you know given some of the setups we see thrown around in the comments is negligable due to the casing since standard cooling and noise tend to go out of the window with a lot of those aforementioned setups? Yeah I'd buy these things many of them have a not as harsh sound profile either so I certainly can't and wont complain!
I would like a roundup of the top 2 or top 3 of the other 4 seasons in the new set up!
I think the mounting plate should be screwed, or fastened however, the positive pressure inside will try to escape wherever it can, and the fan helps it by generating an upward force on the plate. A gasket would be nice, if not overkill.
You could create the pressure flow curves by creating a variable restriction at the end and then record pressure and flow !
That fancy Fluke 922 meter you have there is quite the device. Should you read the literature that came with it in regards to it's functions, you will find that the measurement range is +/- 16mm of H20 with a resolution of .001. (info found with a quick search) You need to properly seal the end of the tube where you shoved it into your stand. Proper sealing will yeild reuslts that are more accurate. Also, to increase the accuracy of the results, you should leave each fan on for a specified time, ie. 30 or 60 seconds. Always ensure the fan is properly fastened before testing. Remember - repeatable results require rigorous testing. That meter can also measure back pressure, so turn the fan around so it is sucking and the measurement will be through the meter connetion only. This is another reason to make sure all your 3D printed press fit fittings are properly sealed, no matter how 'snugly' it feels. Air is small. Really small. Silicone sealent will work just fine for that. Use the silicone to seal a proper fitting for your flexible measuring hose. Properly conducted experimentation will yeild better results and improve your ability to increase your reputation in the community. While I initially started watching because of an obtuse interest in the strange facination that people have with everyday objects, I kept watching because I saw your method change and was intrigued as to where this might go. This obtuse interest often leads to personal revelations of interesting science and in general interesting people from all over. Purchasing that Fluke meter (I have a 34 year old Fluke multimeter, I remember how much it was!) showed that you are either serious or making too much money. I believe it is the former, however true the latter is. Bravo in your endevours and I shall be watching some of your back catalouge out of sincere interest.
That's a big one 😲
I still feel like you should measure pressure, velocity, & db.
For anyone who is watching and doesn’t have the propper context. What he is referring to at the beginning of the video is Rotavirus. Rotavirus is incredibly nasty it's over fairly quickly, but it takes you a while to recover. It affects all people differently, but it was quite literally one of the worst sicknesses I had had since COVID. I spent roughly 7 1/2 hours throwing out not being able to keep fluid or solid down. Pray that you don't get Rotavirus. If the correct pro cautions are not taken when you have it you could be hospitalized for dehydration amongst other things. It has been sweeping the entire upper half of the United States the past few weeks. Also, this is different from the virus that causes you to have a non-stop cough for 35 days because of nasal drainage going down the back of your throat and irritating your throat. That is a whole other virus that has been devastating in the same area.
It's so fun to see you getting progressively more scientific in your testing and adding more and more parameters.
I'd like to suggest using the other piece of tube as a second test with an automotive style mass airflow (MAF) sensor, pulling a raw ohm-resistance reading .
Now that would be cool, having both readings on each fan!
If it's possible, please test the following retail fans for additional points of comparison:
1. Thermalright TL-B12 (cheapest of the "big hub" fans at around 10 USD, and a rather potent "sleeper brand" offering great performance at low cost)
2. Scythe Grand Tornado (their successor to the Gentle Typhoon, and the second cheapest "big hub" fan at around 20 USD)
3. Nidec Gentle Typhoon (I believe you tested it, or at least mentioned it, once against the A12x25 in an earlier season)
4. be Quiet Silent Wings 4 Pro (a similar "luxury" big hub fan that competes against the A12x25 in performance and noise)
5. Thermaltake ToughFan Pro (designed specifically to try and leave no gaps between the fan frame and radiator frame)
6. SuperFlower MEGACOOL (think T30s on steroids)
Now you gotta do a stat card for each fan!
Weight ☆☆
RPM ☆☆☆
Airflow ☆☆☆
Static Pressure ☆☆☆☆
But with the values and maybe 0-100% bars.
Just like with engines where RPM, Horsepower, Max Speed, Acceleration and Torque are relevant stats that make them good for different applications.
Showcasing the strenghts and weaknesses and which ones are jacks of all trades would make for interesting graphics and analysis, and would also let more designs shine.
So, now we're designing air compressors? Cool. I'm for it.
I like this, it seems more repeatable than the door. Also, the A12x25 blades are thin and more numerous, making me think it is optimized for higher airflow, rather than pressure. It would probably beat the others you showed in an airflow metric, because that's just a tradeoff with noise and pressure. Anyways, I like this idea for Season 6.
It would be very good to have airflow speed/cfm just for reference and comparison. Seems like that would help sort of categorise the "purely static pressure" with the practicality of a pc fan that needs both
Dude, right there with you on the gift of endless illness from the little one. I've had a shocking run of illness since my little girl arrived 5 months ago.
Worth it.
A slack tube manometer is easy, fast, and maybe even free in you have some transparent tubing, a ruler, and some water. I do have to concede that in your case a digital display is better for your application though.
The pressures are too small to get accurate measurements with water. You can't measure the water level to 0.1mm with a ruler.
@@Lahhacould use a floater/lever mechanism in order to amplify the range of motion.
But honestly, its easier just to use the digital manometer.
So to win this season you will see blade designs with more chord and blade overlap.
The previous season was airflow thru a radiator. Don't have to make as much airflow to produce more static pressure.
Good video topic, I am selecting a fan for my upgraded silica gel drying bed.
A 40x28 does pretty damn well for V1
Need gaskets or seals on the capped end and the big square piece the fans are mounted to.
Just make sure no leakage, even though I think it would be minimum.
Smaller blades equal less pressure. Any fans I've bought for static pressure (for radiator use) always have wider and fewer blades. Looking forward to see what the season brings!
Good setup. Great way of measuring static pressure.
Most fans will have either a great airflow or a great pressure drop. Not both.
Why would I prioritize static pressure over CFM? Air flow, not pressure, is going to give me my cooling, no?
That Fluke 922 is the meter I suggested in your previous video. It's a good mix of accuracy, granularity, and price. There's even more accurate micromanometers from Alnor and Shortridge, but you're looking at a few thousand dollars.
However, if you bought the meter with the kit, you should be using the pitot tube that came with it for measuring static pressure. You'll want to use the side port that connects to the perimeter holes, not the center one.
Max Flow vs Max pressure is a very different task.
Pressure is 'maximise unit energy per molecule' => fewer molecules
Flow is 'maximise molecules per energy unit' => lower pressure.
That Fluke is a pretty good brand, and a good manometer!
Use the wingnuts because you're increasing the margin of error.
This seems good. Not the most resolution so you may get a number of ties, but enough for discerning the really good from the mediocre.
That test retest reliability looks really good. Fluke is also my meter of choice and there are no parts to wear during repeated testing so I'm liking this test stand setup. Looking forward to season 6. As for kids, I always dreaded the start of school because we knew they were going to bring home something fun. Good luck.
Thanks for the effort put into this! Unfortunately, with static pressure test and open flow test you're now only get the far ends of the fan curve and missing out on all the real life practical performance data in the middle, which would actually be really cool to see! You'd want to take air and pressure data points at different airflows and chart them to show you where that particular fan shines and where it isn't ideal. It's a lot of work! If you have a logging setup to your PC you could gradually close an orifice and log the data, which would be ideal. You'd probably also want a mass air flow sensor instead of an anemometer, but now we're really getting scope creep :D
Two suggestions:
1. Mount a fitting to attach the sensing tube.
2. Run a normalised test at 1000 rpm.
3. Split the bunch into categories
a) Stock: fan disk must not exceed stock fan disk dimensions.
b) Mod: fan disk is allowed to exceed stock fan disk dimensions, aerodynamic parts are not allowed.
c) Unlimited: as above but aerodynamic parts allowed.
That test rig looks much better. You even matched the factory spec!
The turbo fans with stators should do well in this test
you can put a separate gasket between the wind tunnel (black) and the fan mount plate (grey) just to be extra careful
I think that while static pressure testing is cool and should be done I think there should be some airflow allowed out for it to be more "realistic" and so fans that move very little air but have really good static pressure still have to flow air
Static pressure can't be measured while allowing airflow, the point of static pressure is how much a fan can push without allowing air to flow. I'm expecting that designs which seal the gap in the duct would work better
Someone is now going to take that challenge of... "no matter how high you go...."
Noctua openly states that they design for maximim airflow AS WELL AS being as quiet as possible, also being balanced and durable. Static pressure is just one metric. Ideally you would test, CFM/Static Pressure/Volume together.
I'm glad you're feeling better, and you've figured it out.
looks good. ship it. maybe use two or four clamps instead of no screws to hold it all tight.
The A12x25 doesn't actually have very high static pressure compared to other commercially available PC fans. The A12x25 offers high airflow at low noise, but in terms of static pressure, I expect a lot of submitted fans to beat it. For static pressure, you could also use the Corsair ML120, Thermaltake ToughFan 12 Pro, Phanteks PH-F120MP or EK Vardar Evo 120 as a benchmark.
I like this setup a lot. With that said, six seasons in and we're still waiting for two classes separating fans that fully fit in the frame, and an "unlimited class" that can be the ones with add on parts that exceed the frame.
Well done on the upgrade on test methodology. Generally legit setup.
i feel your pain man, i think i had the same thing about 2 weeks ago, it was absolutely brutal. I had no idea something like that could come out of my digestive system.
I cant wait to see all the compressors/ fixed displacement pumps
id like to see all the new fans designed for the back fan. it would allow some of the heavy fans to preform that have failed due to lack of rpm in the previous seasons. right now you have to sacrifice some airflow design for rpm to compete.
In the test, one cable was under the "sealer", so be careful next time. because in the "control" was not.
Having a good test bench, is critical.
The one thing I would add is "performance at a specific noise level". You can often just keep adding speed and get more performance, some of us want the best performance without killing our eardrums.
Pick a specific noise level thats maybe noticeable but not uncomfortable and adjust the speed of the fans to match that noise for a second data point. Some people just want maximum cooling, other people want good cooling with fan noise being controlled. Why can't we test both?
A suggestion for better comparison: log the rpm of the fan and compare pressure between fans at the same rpm. Rationale: the pressure build up is directly proportional to the speed of the fan. The speed will depend on how much work it does moving air (which it won’t in this test) and the weight and air friction of the fan wheel since the motor only have so much torque. This means that a light fan with few blades will spin faster than a heavier fan with many blades and thus, it will look like the light fan performs better in the test. The T30 has the rally high rpm, and thus easily wins the competition.. Love the tests though! :D
I'd love to keep including the measured dB along with the other metrics. For my environment, I would prioritize a quiet fan over one 10% stronger but 40% noisier.
Some 10 years ago corsair used to do AF and SP fans, AF had more blades but they were skinny where SP had less blades but their surface area was much bigger. These days most fans is somewhat a combination of both types into one blade design.
Nice setup. Only thing I might add would be a gasket/sealant where the tube goes into the base. Seems like a good fit now, but I assume it's probably not air tight.
Oh boy new numbers to make bigger with crazy designs.