Great news - really happy these are working for you mate Nice to see some results! Non-bollocks test data is always my favorite (to much marketeering guff these days) Interesting results on the fans - very impressed with your BDF! I'm guessing that it is fairly noisy compared to conventionally made fans? I'm guessing that at low RPM's the forces involves are so small that blade design doesn't really matter too much. Possibly true - however against a heatsink the main factor is going to be the pressure drop. The faster you try and push the air through the fins, the higher pressure you are going to need to do it. I would also imagine usual fluid flow rules apply here, to get 2x the flow you need 4x the pressure
@Bloody_Pete yep I'm planning to do a round with the same fans at 1,650rpm, which is the top speed of the Arctic P12 on the heatsink, and another round of testing at the A12's maximum speed. The thing to bear in mind with the A12 is that it was not designed first and foremost as a "performance" fan but rather as a "low noise" fan, so my prediction is that it'll always be trailing on the performance graphs. Sadly for Noctua, there are many cheaper fans which are also pretty good when it comes to keeping the noise down. As @cobalt6700 says the forces at play in this scenario are very low. When the fans are pushing air through the heat sink at such low speed, the "measurable" static pressure is probably around 1 Pa or 0.1mmH2O, which is basically nothing. The business of "airflow optimised" fans being sold as alternatives to "pressure optimised" fans is mostly marketing bollocks, because all fans are optimised for airflow. The key here is I've shown maximum values for static pressure (e.g. zero airflow), and it's obvious that these numbers don't translate to actual performance on the heatsink. For the record, the Thermalright heatsink I'm using (Ultra 120 Extreme) has very densely stacked fins, and it scales very well as fan speed is increased - this was my main reason for choosing it. I'll do a run of tests on the box to compare how static pressure is affected when there is airflow, and hopefully then we'll get a clearer picture of what's going on! Cheers for all the comments; I hope this is of some benefit and interest to the community.
Awesome! I'm just thinking that covers both bases for the A12 then I used to love my TRUE120 back in the day! I used to use it with one of the first generation Noctua fans and it was still great. This is more than useful, I don't think anyone's done proper fan reviews for a long time!
OK, so I had a bit of time over lunch to run some of the fans on the box with the dampers open, which means there is a fair bit of airflow. Curiously, the leader of the pack (the Delta) falls to the bottom, more or less alongside the Noctua, and the Gentle Typhoon shows its true colours! I'm aware that it's more complicated than "just" static pressure, but what's interesting to note here is that the Gentle Typhoon is considerably ahead of the other two fans in this scenario. I left the original graph in the background so you can compare them. The BDF was virtually identical in performance to the Delta, so I omitted it from the graph. I'll do some 2,000rpm tests on the heatsink to see how they all scale.
The GT is nice, but the Noctua seems a fair bit quieter. If I get a chance this evening I'll measure both with the Tenma. I expected the Noctua to trail behind the others even at higher speeds, but it turns out the GT and A12 are at level pegging. No point making a graph as I only tested four fans, so here are the results with each fan set to 2,000rpm: (Max load temperature above ambient) Noctua NH-A12x25: 30.9°C Gentle Typhoon: 30.9°C Delta AFB1212H: 30.0°C Nidec G1238B: 29.7°C The Delta pulls ahead despite the far lower static pressure, and it's no surprise that the G1238B takes the lead. It's also worth mentioning that the Arctic P12 put in a very strong performance at its max speed of 1,650rpm, achieving a delta temp of just 31.9°C. I still have several fans to test and will probably acquire more as time goes on. Here are a couple of pics of the setup. The white wire in the first pic is the probe from one of the thermometers. I measured ambient temperature in three places and averaged them out for an average max ambient reading. The hottest ambient temp (unsurprisingly) is at the intake of the fan, and it's typically 1 or 2 degrees above the other measurements. Most fans seem to cycle a lot of the air that's being exhausted from the heat sink, with the exception of the G1238B which blasts the exhaust air into another dimension even at only 2,000rpm.
Interesting! I assumed that A12 would perform well as it look so similar to the GT. Looking forwards to the rest of the results!
Here's a quick sound level video I did with the Noctua, Gentle Typhoon and the P12 on the Thermalright heat sink. The Noctua is really, really quiet.
Video stops after 2:55 but runs to 6:50. I really dislike the Gentle Typhoon sound at 1600rpm and 2000rpm and would probably sound a lot worse on a radiator. That's what the some of the original poorer quality ones used to sound like with a high pitched background noise overriding the air noise. Fortunately for every poor one there were around 4 good ones. Unfortunately I sold most of my good ones when I switched to Noiseblocker PL2s. The Noctua does it for me with sound at all speeds. I'd like to hear that one on a rad. I'm currently using 14 x Noiseblocker PL2s, 1500rpm (running around 500rpm) on my main system. I believe I have one or two spare, would you like me to send one for testing?
Yeah that is meant to be the end of the video after the P12... whoopsie! The GT is definitely a good bit louder than the Noctua. The thing with the Noctua is that it also has a smoother, softer sound as well as being objectively quieter. It's hard to describe. I'll put the fans on the rad to see how much of a difference there is. And yes, I'll definitely take a PL2 for testing - I've never heard a Noiseblocker in person but they seem to be popular. The eLoops look really interesting too.
I hadn't thought this through. I'm still being careful as I'm classed as extremely vulnerable so would prefer to avoid the Post Office if at all possible. I'll try and make a package that will fit in a Post Box. Are you still at Gourdie Road? It's not the loudness of the GT it's the high pitched noise that overrides the sound of the air noise that irritates me. I hate to say this but I replaced my original GTs for the Noiseblocker PL2s purely for their looks in my CaseLabs case. They looked awful.The GTs I didn't sell were the ones that suffered from the high pitch noise. Like a fool I sold the good ones.
@Big Elf don't worry about it - definitely not worth going to the PO for the sake of a fan which I can easily obtain myself! And yes the GTs have a frequency about them. I found that I had to run mine at 1,000rpm otherwise the sound got to me. To be honest, in a very quiet room most fans will still be audible at low speed unless they are >1m away, or inside a case with some sort of damping or filters. I also need to test some sound levels with the Silverstone filter I got.
I must have been really lucky as my 3 never had this problem! I remember hearing it mentioned a lot, I think they did a refresh of them and it was the refreshed ones that suffewrred from this.
Haven't done a whole lot on the testing front but I have plans (and fans...). I'm trying to better understand the physics behind airflow and it's heavy stuff, but I'm getting somewhere -- particularly with the difference between static pressure and velocity pressure, and how pitot tubes do their thang. I also put together a model of the G1238B in 3D so that I could have a close look at the shape of the impeller blades and how they interact with the vanes at the back of the frame, and it's just madness how cool this fan is. The impeller seems to take some design cues from the turbofan on a jet engine and the clearances are incredibly tight for a "measly" 120mm square fan.
Hey folks, I've been working on a video of the airflow chamber and it'll take some time to get it finished. I've got yet more fans coming this week, most notably a Silverstone Air Penetrator from yesteryear. I'm expecting it not to be very good, but gimmicky fans are always welcome when testing! The 3D side-project continues, and I wondered what might happen if Nidec and Noctua were to collaborate... well, I'm guessing Nidoctua would be a fair bet. The rubber thingys really help reduce the noise.
Back on topic... I got an odd NMB fan from ebay last week. I think it might be an older version of the 4715KL, because it's really hard to find and has a unique impeller design with petal-shaped blades (unlike anything I've seen before). I wondered how this might fare on the test chamber and (surprisingly) it's actually quite weak for such a powerful fan... which might explain why it's not so commonly encountered. Anyways, I'll be doing some more "intermediate" videos as soon as I have some free time.
