FAN TESTING, 3 Years Later - Airflow Case Mesh vs. Noctua NF-A12
video description
Date: 2025-02-15
Related videos
Comments and reviews: 20
rediband
In 1981 the IBM PC arrived. It was a metal box containing a small amount of heat-producing electronics, but the tiny amount of heat was not a problem. It was obvious that the sole most important thing was to isolate the electronics from people. Which was absolutely true at the time. Then we had the IBM PC XT and its clones, also no problem. Then the PC AT, same. Millions sold, it was now the defacto business pc, people knew exactly what a pc looked like. All metal boxes enclosing electronics, but no problem because cpu's didn't put out much heat, and graphics cards were in their infancy. Next, in 1995, 30 years ago, they made the ATX standard for motherboards. It was inherent in the design that they must be inside boxes, because there was no standard for anything that was on the motherboard other than the board dimensions and back panel. Then, heat-design-wise, after ATX came... Nothing. That' was it. For decades, nothing was done in the computer industry to design a standard for computers that could put the heat from on electronics on the outside while making the inside inherently cool. Because that's impossible. Getting back to the story, around ten years later, the hardware got prohibitively hot. So a bunch of small companies made billions of dollars making heat pumps to move the heat from the inside the of the metal boxes to the outside. As time went on, these heat pumps became increasingly complicated, a veritable 'Wild West of heat pumps with no standards whatsoever. So to bring absolute clarity to the industry, reviewers that worked for free made careers out of testing which heat pumps were best to overcome the 45-year-old perfect basis for pc computer design. Which made perfect sense because it's impossible for engineers to get around this, pc's and heat are a force of nature, like meteorites burning up in the atmosphere. Of course, engineers at Lockheed Martin did design a heat shield for the Mercury program in 1958. Well, designing pc's is several levels more complicated than rocket science. The latest chapter is that the pc reviewers are now proudly acquiring very expensive test equipment to ensure the pc and heat pump makers are kept honest, because that's what the problem is.
reply
In 1981 the IBM PC arrived. It was a metal box containing a small amount of heat-producing electronics, but the tiny amount of heat was not a problem. It was obvious that the sole most important thing was to isolate the electronics from people. Which was absolutely true at the time. Then we had the IBM PC XT and its clones, also no problem. Then the PC AT, same. Millions sold, it was now the defacto business pc, people knew exactly what a pc looked like. All metal boxes enclosing electronics, but no problem because cpu's didn't put out much heat, and graphics cards were in their infancy. Next, in 1995, 30 years ago, they made the ATX standard for motherboards. It was inherent in the design that they must be inside boxes, because there was no standard for anything that was on the motherboard other than the board dimensions and back panel. Then, heat-design-wise, after ATX came... Nothing. That' was it. For decades, nothing was done in the computer industry to design a standard for computers that could put the heat from on electronics on the outside while making the inside inherently cool. Because that's impossible. Getting back to the story, around ten years later, the hardware got prohibitively hot. So a bunch of small companies made billions of dollars making heat pumps to move the heat from the inside the of the metal boxes to the outside. As time went on, these heat pumps became increasingly complicated, a veritable 'Wild West of heat pumps with no standards whatsoever. So to bring absolute clarity to the industry, reviewers that worked for free made careers out of testing which heat pumps were best to overcome the 45-year-old perfect basis for pc computer design. Which made perfect sense because it's impossible for engineers to get around this, pc's and heat are a force of nature, like meteorites burning up in the atmosphere. Of course, engineers at Lockheed Martin did design a heat shield for the Mercury program in 1958. Well, designing pc's is several levels more complicated than rocket science. The latest chapter is that the pc reviewers are now proudly acquiring very expensive test equipment to ensure the pc and heat pump makers are kept honest, because that's what the problem is.
reply
carlr2837
As an engineer (who does NOT work with fans), let me say you are doing an excellent job, but I have a basic comment. Physics is the underlying science that defines nature, while mathematics is the tool we use to describe it. Engineering is the use of mathematics to describe the way physics applies to real world situations and machines. ;)
Now for some ideas you can test as you explore the fascinating rabbit hole you have discovered. I personally have worked with pump curves, which, while similar, are also very different, because most liquids are essentially non-compressible. One similarity between pumps and fans is that the designer must choose between maximum flow and maximum pressure. That is something you can test, and is a place that will clarify for you the difference in the language chosen by the engineer.
Situation: One fan might produce 60 CFM with low resistance, but drop off to 10CFM with high resistance, while another fan might produce only 40 CFM with low resistance, but only drop to 25 CFM in the high resistance situation.
Hypothesis (i.e.Theory to test): The higher pressure fan would probably be a great choice for a CPU tower cooler, where the fins create a lot of back pressure, while the high flow fan would be a good choice for a case fan, where there is less back pressure.
Situation: Some builds have a lot of wires that will create turbulence and resistance
Hypothesis: A higher pressure fan would work better in situations with lots of wires
Situation: A tower cooler could have a sandwich, with a fan in front of the cooler and one behind, while another might have a push fan, and yet another might have a pull fan
Hypotheses: A push fan is more effective than a pull fan. Adding a pull fan to the back does not increase the flow so much as it reduces the back pressure seen by the push fan, allowing you opt for a higher flow fan as the push fan, rather than a higher pressure fan. Note that this might also apply to case fans in general. In a case with a single fan, a push fan may be more effective than an exhaust fan, while if you use one of each, you might prefer higher flow fans to higher pressure fans.
reply
As an engineer (who does NOT work with fans), let me say you are doing an excellent job, but I have a basic comment. Physics is the underlying science that defines nature, while mathematics is the tool we use to describe it. Engineering is the use of mathematics to describe the way physics applies to real world situations and machines. ;)
Now for some ideas you can test as you explore the fascinating rabbit hole you have discovered. I personally have worked with pump curves, which, while similar, are also very different, because most liquids are essentially non-compressible. One similarity between pumps and fans is that the designer must choose between maximum flow and maximum pressure. That is something you can test, and is a place that will clarify for you the difference in the language chosen by the engineer.
Situation: One fan might produce 60 CFM with low resistance, but drop off to 10CFM with high resistance, while another fan might produce only 40 CFM with low resistance, but only drop to 25 CFM in the high resistance situation.
Hypothesis (i.e.Theory to test): The higher pressure fan would probably be a great choice for a CPU tower cooler, where the fins create a lot of back pressure, while the high flow fan would be a good choice for a case fan, where there is less back pressure.
Situation: Some builds have a lot of wires that will create turbulence and resistance
Hypothesis: A higher pressure fan would work better in situations with lots of wires
Situation: A tower cooler could have a sandwich, with a fan in front of the cooler and one behind, while another might have a push fan, and yet another might have a pull fan
Hypotheses: A push fan is more effective than a pull fan. Adding a pull fan to the back does not increase the flow so much as it reduces the back pressure seen by the push fan, allowing you opt for a higher flow fan as the push fan, rather than a higher pressure fan. Note that this might also apply to case fans in general. In a case with a single fan, a push fan may be more effective than an exhaust fan, while if you use one of each, you might prefer higher flow fans to higher pressure fans.
reply
thepuff791
For once, I can help with the testing methodology I think. I have just under 10 years of aerospace engineering background specifically on low to high pressure ECS, so take my advice as you will based on that:
1. I think the way you described the flow system was fine. Any engineer with a nonzero amount of brain cells would know exactly what you meant. Almost none would correct you in any way, in fact, I think a lot would describe pressure drop on relation to flow and it's restriction within a system like you did honestly. I would probably say With restriction X, a pressure drop of Y is required to yield flow Z.
2. This might be nit-picky, but I prefer inW or inWc over inAq. Also, when describing pressure drop on the charts, you need to put Pd or whatever variant of that you want to define that inWc is pressure delta. If you measured in psi, you would say psid to distinguish that it is a pressure delta to be specific so the same should be done for inWc.
3. You should publish all of your governing theory on the GN site. We use W = CdAKPN/(T) for our flow equation which is mass based, but obviously you are using volumetric, so seeing the theory and how you get your data in general would be nice.
4. If your equations use absolute pressure, you ABSOLUTELY need to put your barometric pressure recorded at the time of testing. That would completely change the values.
I hope that helps. In short, I think you guys did pretty well honestly. I know you want to present the data more from a scientific POV than a reviewer, but I think all the data you presented was sound. I would prefer some charts on non-linear charts if you think it is digestible to a regular audience too maybe on the website. Anyways, love to see the fan tester up! Awesome day for GN.
reply
For once, I can help with the testing methodology I think. I have just under 10 years of aerospace engineering background specifically on low to high pressure ECS, so take my advice as you will based on that:
1. I think the way you described the flow system was fine. Any engineer with a nonzero amount of brain cells would know exactly what you meant. Almost none would correct you in any way, in fact, I think a lot would describe pressure drop on relation to flow and it's restriction within a system like you did honestly. I would probably say With restriction X, a pressure drop of Y is required to yield flow Z.
2. This might be nit-picky, but I prefer inW or inWc over inAq. Also, when describing pressure drop on the charts, you need to put Pd or whatever variant of that you want to define that inWc is pressure delta. If you measured in psi, you would say psid to distinguish that it is a pressure delta to be specific so the same should be done for inWc.
3. You should publish all of your governing theory on the GN site. We use W = CdAKPN/(T) for our flow equation which is mass based, but obviously you are using volumetric, so seeing the theory and how you get your data in general would be nice.
4. If your equations use absolute pressure, you ABSOLUTELY need to put your barometric pressure recorded at the time of testing. That would completely change the values.
I hope that helps. In short, I think you guys did pretty well honestly. I know you want to present the data more from a scientific POV than a reviewer, but I think all the data you presented was sound. I would prefer some charts on non-linear charts if you think it is digestible to a regular audience too maybe on the website. Anyways, love to see the fan tester up! Awesome day for GN.
reply
Violet-ui
I hope you guys don't lose sight of the big picture.
The P/Q curves themselves are not as accurate as actual real world results. Yes, you can mostly estimate real world performance based on the PQ curve, but it's less accurate than straight up measuring the real world performance.
For example if you want to know how a fan performs on a radiator, put it on a radiator and measure dT water over ambient with a constant load. This directly measures the heat dissipation. Instead of using a PQ curve to guess how mow airflow would go through the rad and then assume that airflow is proportional to heat dissipation (which it mostly is but not accurate), which is how someone else with that fan tester suggests doing it.
Also at 10:43 mounting it like that is not accurate if I am seeing that correctly. It looks like the fan would still be able to draw air through the grid from above and below. In a case with 3 front intake fans, they share the front panel. So each fans has effectively like a third of the area. Setting it up like you did there would make the fan use more of the grill than it could in a real world use case. I think this is why you're still seeing good airflow even with the very low porosity. It would not be like this in the real world
It also wouldn't be accurate for when you're using 1 front intake fan, since the air from above and below would be in big parts recycled case air, unless it's a case where the other fan slots are blocked when not in use.
reply
I hope you guys don't lose sight of the big picture.
The P/Q curves themselves are not as accurate as actual real world results. Yes, you can mostly estimate real world performance based on the PQ curve, but it's less accurate than straight up measuring the real world performance.
For example if you want to know how a fan performs on a radiator, put it on a radiator and measure dT water over ambient with a constant load. This directly measures the heat dissipation. Instead of using a PQ curve to guess how mow airflow would go through the rad and then assume that airflow is proportional to heat dissipation (which it mostly is but not accurate), which is how someone else with that fan tester suggests doing it.
Also at 10:43 mounting it like that is not accurate if I am seeing that correctly. It looks like the fan would still be able to draw air through the grid from above and below. In a case with 3 front intake fans, they share the front panel. So each fans has effectively like a third of the area. Setting it up like you did there would make the fan use more of the grill than it could in a real world use case. I think this is why you're still seeing good airflow even with the very low porosity. It would not be like this in the real world
It also wouldn't be accurate for when you're using 1 front intake fan, since the air from above and below would be in big parts recycled case air, unless it's a case where the other fan slots are blocked when not in use.
reply
Hyperion64
It would be interesting to understand the impact of obstacles on noise-normalized thermals from a sound perspective since mesh can direct noise away from the measured surface and filters slightly absorb noise. A sound source with fixed emissions, e.g. a fan with controlled RPM, will then provide a baseline against which different types of obstructing surfaces can be evaluated.
Adding a steady heat source would allow the comparison between noise increase compared to temperature increase by installing obstructing surfaces.
Regarding the presentation of information: Viewers may struggle to grasp the differences between air flow and thermals. From a technical point of view, air flow measurement reduces the complexity of the system to get accurate and consistent results, but the presentation of the scientific results should translate into applicable data.
One option is the calculation of thermals based on air flow in a hypothetical environment. The irony of simulating practical impact is not lost on me. However, cross-validation on a reference system allows for the most accurate and isolated base results made accessible to a broader audience by providing relevant metrics without having to perform the same test in multiple environments.
Think of a spreadsheet that calculates thermals based on air flow under preconfigured conditions for each column.
reply
It would be interesting to understand the impact of obstacles on noise-normalized thermals from a sound perspective since mesh can direct noise away from the measured surface and filters slightly absorb noise. A sound source with fixed emissions, e.g. a fan with controlled RPM, will then provide a baseline against which different types of obstructing surfaces can be evaluated.
Adding a steady heat source would allow the comparison between noise increase compared to temperature increase by installing obstructing surfaces.
Regarding the presentation of information: Viewers may struggle to grasp the differences between air flow and thermals. From a technical point of view, air flow measurement reduces the complexity of the system to get accurate and consistent results, but the presentation of the scientific results should translate into applicable data.
One option is the calculation of thermals based on air flow in a hypothetical environment. The irony of simulating practical impact is not lost on me. However, cross-validation on a reference system allows for the most accurate and isolated base results made accessible to a broader audience by providing relevant metrics without having to perform the same test in multiple environments.
Think of a spreadsheet that calculates thermals based on air flow under preconfigured conditions for each column.
reply
lexzbuddy
I'm an engineer and I understand the focus issue. When you have multiple projects, time constraints, meetings, people and a plethora of other things then you need to do your job too, focus can be a challenge.
As you aren't an engineer, I imagine there are some additional issues for you. You don't have a formal, basic education in fluid mechanics and thermodynamics. That will make thing even more challenging for you. I can only imagine the challenge you have given yourself.
I don't know if you have considered it but perhaps you should consider hiring an actual engineer. That would be a challenge too as we are all pretty specialised and the longer we are away from university the less likely we are to change path. Hiring a graduate without experienced engineers to guide them would be a challenge too as they may not integrate well without appropriate guidance. So, how do you find someone suitable Not impossible but not easy either.
So, well done taking on such a challenge. Good luck in the future and I hope that you make this work for you in a way that's meaningful.
reply
I'm an engineer and I understand the focus issue. When you have multiple projects, time constraints, meetings, people and a plethora of other things then you need to do your job too, focus can be a challenge.
As you aren't an engineer, I imagine there are some additional issues for you. You don't have a formal, basic education in fluid mechanics and thermodynamics. That will make thing even more challenging for you. I can only imagine the challenge you have given yourself.
I don't know if you have considered it but perhaps you should consider hiring an actual engineer. That would be a challenge too as we are all pretty specialised and the longer we are away from university the less likely we are to change path. Hiring a graduate without experienced engineers to guide them would be a challenge too as they may not integrate well without appropriate guidance. So, how do you find someone suitable Not impossible but not easy either.
So, well done taking on such a challenge. Good luck in the future and I hope that you make this work for you in a way that's meaningful.
reply
lrmcatspaw1
I am so scared of this.
I have been reading old magazines trying to identify what testing methood was the most reliable for testing Fan real world effectivness for a bit of time (20 years) and there had been not that much information about how a testing methood was better or worse.
The big bag of air speed to fill up test. The throu radiator air pressure test. The air speed in a tube test.
And I usually went by the average of those to pick something that did not have a high pitch sound while in use.
But I always thought to myself yes, this test helps but... There was always a but. The amount of variables was just too much to consider everything.
(Oh, and some fans actually look kinda cool so that was something to consider as well).
When it comes to charts, to me, the most important part was always performance of the fan (throu radiator and without it) at a set db (usually a rather quiet one, but I will be honest: Not only do I not remember the actual db nor I full am aware of how much db is what).
reply
I am so scared of this.
I have been reading old magazines trying to identify what testing methood was the most reliable for testing Fan real world effectivness for a bit of time (20 years) and there had been not that much information about how a testing methood was better or worse.
The big bag of air speed to fill up test. The throu radiator air pressure test. The air speed in a tube test.
And I usually went by the average of those to pick something that did not have a high pitch sound while in use.
But I always thought to myself yes, this test helps but... There was always a but. The amount of variables was just too much to consider everything.
(Oh, and some fans actually look kinda cool so that was something to consider as well).
When it comes to charts, to me, the most important part was always performance of the fan (throu radiator and without it) at a set db (usually a rather quiet one, but I will be honest: Not only do I not remember the actual db nor I full am aware of how much db is what).
reply
lexzbuddy
Final comment. You are new to this, you will make mistakes, that is normal and perhaps one of the best ways to learn. It's complicated and takes a long time to get good at. Good engineers are made by learning from their mistakes and oversights. A good engineering manager will let them make mistakes and learn. Don't punish yourself for not being perfect. Just learn from your experience and remember to take the next step, then the next one, then the one after that. It's a journey not a destination, a bit cheesy but mostly true. What you don't know, ask. It's ok to not know. Be prepared to accept that, it takes a lot of time to get somewhere, if it was easy, everyone would be doing it.
The language, you'll learn. It's all extremely specific and folks will definitely let you know, I promise you. Don't let it get to you, it's the nature of the beast. Just remember to have some fun too, that matters.
reply
Final comment. You are new to this, you will make mistakes, that is normal and perhaps one of the best ways to learn. It's complicated and takes a long time to get good at. Good engineers are made by learning from their mistakes and oversights. A good engineering manager will let them make mistakes and learn. Don't punish yourself for not being perfect. Just learn from your experience and remember to take the next step, then the next one, then the one after that. It's a journey not a destination, a bit cheesy but mostly true. What you don't know, ask. It's ok to not know. Be prepared to accept that, it takes a lot of time to get somewhere, if it was easy, everyone would be doing it.
The language, you'll learn. It's all extremely specific and folks will definitely let you know, I promise you. Don't let it get to you, it's the nature of the beast. Just remember to have some fun too, that matters.
reply
gamersnexus
Important to consider those TRC variables, I assume you already know this from those books -
In an ideal gas, if you keep the volume constant, the pressure is directly proportional to the absolute temperature (as stated by Gay-Lussac's Law). This means that if you raise the temperature (measured in Kelvin), the pressure will increase, and if you lower the temperature, the pressure will decrease.
However, if the gas is not confined to a constant volume (for example, if it’s allowed to expand), the relationship is governed by the ideal gas law, pV = nRT. In that case, changes in temperature can also lead to changes in volume, and the effect on pressure will depend on how the volume adjusts.
In summary, under constant volume conditions, temperature and pressure change in tandem; but in more general situations, other factors (like volume and the amount of gas) must be considered.
reply
Important to consider those TRC variables, I assume you already know this from those books -
In an ideal gas, if you keep the volume constant, the pressure is directly proportional to the absolute temperature (as stated by Gay-Lussac's Law). This means that if you raise the temperature (measured in Kelvin), the pressure will increase, and if you lower the temperature, the pressure will decrease.
However, if the gas is not confined to a constant volume (for example, if it’s allowed to expand), the relationship is governed by the ideal gas law, pV = nRT. In that case, changes in temperature can also lead to changes in volume, and the effect on pressure will depend on how the volume adjusts.
In summary, under constant volume conditions, temperature and pressure change in tandem; but in more general situations, other factors (like volume and the amount of gas) must be considered.
reply
itguydave2164
I think the main point you need to get in the charts is something the consumer can relate to. It's fantastic to see real figures, but in the shoes of a consumer looking to purchase a cooling setup, it means nothing to me. What you need is some sort of line on the chart that represents ideal - and the closer the results are to that line, the better. The actual results from the fan tester might need to be calculated alongside noise and effectiveness at reducing the case temperature. In the case of radiators, your entire testing regime will need to consider the radiator type in addition to the fan, so there needs to be some way of explaining how static pressure impacts cooling with various radiator types. All in all, maybe think about how you define ideal, and everything references back to that. Without it, it's just too much data for the average consumer to interpret.
reply
I think the main point you need to get in the charts is something the consumer can relate to. It's fantastic to see real figures, but in the shoes of a consumer looking to purchase a cooling setup, it means nothing to me. What you need is some sort of line on the chart that represents ideal - and the closer the results are to that line, the better. The actual results from the fan tester might need to be calculated alongside noise and effectiveness at reducing the case temperature. In the case of radiators, your entire testing regime will need to consider the radiator type in addition to the fan, so there needs to be some way of explaining how static pressure impacts cooling with various radiator types. All in all, maybe think about how you define ideal, and everything references back to that. Without it, it's just too much data for the average consumer to interpret.
reply
AndriusKamarauskas
13:40 I have a suggestion for an orifice name - as in analog photography, a small aperture is called a pinhole. Still, then you would be creating slang that would be understandable just in your community. I strongly recommend aligning yourself with scientific literature naming since you are doing research. As a teacher in this field, you have a responsibility to teach them right from the start.
It reminds me of the situation when I was a kid and asked my mom in which street we were living and she responded with random name - 55
so when I was referencing other people's houses in the same street I always said random name - 55 and then number of other house. I was at least 12 when I understood that there was no number in our street name. Steve don't be like my mom
Another suggestion, make a segment of fan testing named - only fans
reply
13:40 I have a suggestion for an orifice name - as in analog photography, a small aperture is called a pinhole. Still, then you would be creating slang that would be understandable just in your community. I strongly recommend aligning yourself with scientific literature naming since you are doing research. As a teacher in this field, you have a responsibility to teach them right from the start.
It reminds me of the situation when I was a kid and asked my mom in which street we were living and she responded with random name - 55
so when I was referencing other people's houses in the same street I always said random name - 55 and then number of other house. I was at least 12 when I understood that there was no number in our street name. Steve don't be like my mom
Another suggestion, make a segment of fan testing named - only fans
reply
lexzbuddy
Fluid mechanics, it's pretty interesting stuff and most people only have a basic understanding. Even as a basic engineering graduate, your knowledge would only be fairly fundamental. By fundamental I mean, understand most of everything in those books you have but still, fundamental believe it or not. It's a pretty deep rabbit hole and you can spend a lifetime as an engineer focussed on the subject and not know it all.
Air is compressible. That means you are on the more complicated side of the topic. I do hydraulics which are incompressible fluids. I do a lot of control stuff. Air, compressible fluids, a whole level up in terms of complexity. You've bit off a huge piece of pie. I think you're a bit mad . Good for you!
reply
Fluid mechanics, it's pretty interesting stuff and most people only have a basic understanding. Even as a basic engineering graduate, your knowledge would only be fairly fundamental. By fundamental I mean, understand most of everything in those books you have but still, fundamental believe it or not. It's a pretty deep rabbit hole and you can spend a lifetime as an engineer focussed on the subject and not know it all.
Air is compressible. That means you are on the more complicated side of the topic. I do hydraulics which are incompressible fluids. I do a lot of control stuff. Air, compressible fluids, a whole level up in terms of complexity. You've bit off a huge piece of pie. I think you're a bit mad . Good for you!
reply
alphameric
An interesting test could be something like oversized fan cooling performance on a variety of CPU coolers (or a standardized one for reference).
Main reason being you have some coolers that use 90, 120, or even 140mm fans that have unused surface area which could be capitalized on by sizing up to a non-standard fan.
For example I use a Deepcool AN600 with Noctua NF-A14x25r G2 and it's much quieter than the standard 120mm fan and even NF-A12x25 at comparable temperatures.
I WISH I had a tiny ITX case or cooler with a use case for a 200mm fan.
I just know there are a ton of coolers that could be easily buffed with oversized fans, but I don't have the time/resources to document it.
reply
An interesting test could be something like oversized fan cooling performance on a variety of CPU coolers (or a standardized one for reference).
Main reason being you have some coolers that use 90, 120, or even 140mm fans that have unused surface area which could be capitalized on by sizing up to a non-standard fan.
For example I use a Deepcool AN600 with Noctua NF-A14x25r G2 and it's much quieter than the standard 120mm fan and even NF-A12x25 at comparable temperatures.
I WISH I had a tiny ITX case or cooler with a use case for a 200mm fan.
I just know there are a ton of coolers that could be easily buffed with oversized fans, but I don't have the time/resources to document it.
reply
xypha85
Ok, i love this stuff. Like i would come work for you doing it lvl of love this stuff.
My rando question/thought. Its relatively easy testing,expensive and obviously lots of learning, and needs thw tools, buy with equipment like you have which i think any case designer should have or at least know a std tested table of say porosities. Why in the f, do case makers not know this, 50% porous seems so far quiet consistently a good numbrer, how is it 50 odd years of cases are cases not always 50% pourous panels by now Would seem like a std choice and any deviation would need marketed justification
reply
Ok, i love this stuff. Like i would come work for you doing it lvl of love this stuff.
My rando question/thought. Its relatively easy testing,expensive and obviously lots of learning, and needs thw tools, buy with equipment like you have which i think any case designer should have or at least know a std tested table of say porosities. Why in the f, do case makers not know this, 50% porous seems so far quiet consistently a good numbrer, how is it 50 odd years of cases are cases not always 50% pourous panels by now Would seem like a std choice and any deviation would need marketed justification
reply
GeckoDeLimon
It wouldn't be uncommon for someone discussing these results to say that there wasn't a significant drop off in performance until the 32.6% panel. Considering your MMAQ / CFM graph (28:30), what would the graph look like with the Y axis inverted Place your zero reference at the top and the water column pressures increasing toward the bottom. Then poor performers would be at the bottom of the graph, and the good performers would be seen to NOT drop off and stay closer to the ideal reference.
I could be wrong, though. I haven't put in the work that Steve has. :D
reply
It wouldn't be uncommon for someone discussing these results to say that there wasn't a significant drop off in performance until the 32.6% panel. Considering your MMAQ / CFM graph (28:30), what would the graph look like with the Y axis inverted Place your zero reference at the top and the water column pressures increasing toward the bottom. Then poor performers would be at the bottom of the graph, and the good performers would be seen to NOT drop off and stay closer to the ideal reference.
I could be wrong, though. I haven't put in the work that Steve has. :D
reply
henrysalayne
What the hell is cfm Centifaradmeters The world has developed a pretty conclusive set of acceptable units called SI derived units and I would strongly recommend to use it for clear communication. The same goes for mmAQ. It's supposed to be called mmH20. Lower and upper case letters matter here, since M and m are off by 9 orders of magnitude! Even better would be the use of Pascal (but Noctua do use the unit mmH20 themselves). Since you export everything into an Excel sheet it could automatically handle any unit conversion.
reply
What the hell is cfm Centifaradmeters The world has developed a pretty conclusive set of acceptable units called SI derived units and I would strongly recommend to use it for clear communication. The same goes for mmAQ. It's supposed to be called mmH20. Lower and upper case letters matter here, since M and m are off by 9 orders of magnitude! Even better would be the use of Pascal (but Noctua do use the unit mmH20 themselves). Since you export everything into an Excel sheet it could automatically handle any unit conversion.
reply
UberLopez
Hi Steve and team excellent work and glad to see some initial results, especially such a clear point of diminishing returns re: front panel porosity.
Testing I would like to see is around the airflow of fans mounted on the bottom of a case (like in Coolermaster NR200P) and how clearance from the table/floor affects that airflow. I imagine you could construct a vertical plate setup which you could offset by common case foot heights. Perhaps there will be an optimal/no further improvement point as with panel porosity.
reply
Hi Steve and team excellent work and glad to see some initial results, especially such a clear point of diminishing returns re: front panel porosity.
Testing I would like to see is around the airflow of fans mounted on the bottom of a case (like in Coolermaster NR200P) and how clearance from the table/floor affects that airflow. I imagine you could construct a vertical plate setup which you could offset by common case foot heights. Perhaps there will be an optimal/no further improvement point as with panel porosity.
reply
Crazyneo2917
As a small SI that started from passion with the focus on doing the right thing and with a few thousand pcs sold over the last few years, i can understand the feeling and the effort going behind this. Once we grow, we hope to have such tools to measure and make or design our own products rather than depend on 3rd party products that don't meet our requirements. Until then, we will piggyback on your results to make our decisions more accurate rather than depend solely on our standard bench tests
reply
As a small SI that started from passion with the focus on doing the right thing and with a few thousand pcs sold over the last few years, i can understand the feeling and the effort going behind this. Once we grow, we hope to have such tools to measure and make or design our own products rather than depend on 3rd party products that don't meet our requirements. Until then, we will piggyback on your results to make our decisions more accurate rather than depend solely on our standard bench tests
reply
alang3079
I think it would be cool for you to make a fluid mechanics 101 video for all the stuff that you learned along the way to understanding how this machine works. It would be great for you to show that you understand everything and can explain it in a manner that (mostly) everyone can understand, and it would be great for us as the audience to know the concepts that affect the results we might see. Looking forward to all the upcoming content!
Also, the orifice plate should be named Scotty.
reply
I think it would be cool for you to make a fluid mechanics 101 video for all the stuff that you learned along the way to understanding how this machine works. It would be great for you to show that you understand everything and can explain it in a manner that (mostly) everyone can understand, and it would be great for us as the audience to know the concepts that affect the results we might see. Looking forward to all the upcoming content!
Also, the orifice plate should be named Scotty.
reply
gamersnexus
Steve, I have been very excited for this as I’ve been hoping to see tests of lower speed server fans (ie Delta AFB1212HHE) to see how they perform noise-normalized at low speeds on dense radiators.
There’s a whole market of gaudy, expensive consumer pc fans, but the server stuff is cheap and bulletproof with unbelievable performance at high speeds.
I’d like to verify that the gamer fans do offer better noise normalized performance vs the real fans.
reply
Steve, I have been very excited for this as I’ve been hoping to see tests of lower speed server fans (ie Delta AFB1212HHE) to see how they perform noise-normalized at low speeds on dense radiators.
There’s a whole market of gaudy, expensive consumer pc fans, but the server stuff is cheap and bulletproof with unbelievable performance at high speeds.
I’d like to verify that the gamer fans do offer better noise normalized performance vs the real fans.
reply
Add a review, comment
Other channel videos
