TRUE and Megahalems: Performance with different fan configurations

[muddocktor]

@burebista, thanks for the link. It looks like a damn nice heatsink. I have been wanting to test a Noctua heatsink as I've heard that their workmanship is great and they aren't any slouches in the engineering department either.

@bing: I would love to test it out, but it depends on finances, since I don't get free stuff from the heatsink world but have to buy my stuff. Oh, Jab-tech has thrown me a few discounts on some things in the past and he has also given a few items to help me out, but he is just one small business and can't afford to be giving out heatsinks and such. Things like that would have to come from the manufacturers and I figure I'm not on their radar since I'm not much for writing front page articles. Plus, I would always wonder if they were cherry picking review items if they were sending them to me too.

 

[EarthDog]

Thanks for the info man! Great little review!

You should use graphing on that data too!

 

[muddocktor]

OK, got in from the rig and the testing will continue shortly. And with some more fans added to the mix too. I just ordered 2 Scythe S-Flex SFF21G fans, 2 Scythe Gentle Typhoon 57 cfm fans and another Panaflo L1BX so I can test all these in 1 and 2 fan configurations. While waiting for them, I will continue my testing of the new style TRUE mount.

And EarthDog, I will graph everything later to make it nice and pretty.

 

[mattspalace]

Next you'll have to buy 8 different varieties of thermal grease and see how each positively or negatively affects temps.

My bottom line - if the cooler will hold two fans, go with two fans.

 

[neonblingbling]

And EarthDog, I will graph everything later to make it nice and pretty.

I was bored (and just saw this post ) :

 

[neonblingbling]

My reply to the data: just use one fan. Cheaper, quieter, and cools better in one case.

Personally, I'd just use a medium Yate Loon, or maybe under-volted high Yate Loon. Couple reasons:

• It's cheaper than a San Ace ($20 vs $5)
• It's quieter than a San Ace (39 dBa vs 33 dBa)
• It cools better than the Noctua (10C in worst case)
• Dual fans only cool about ~3C cooler at best for 'twice the cost' and more noise.

 

[muddocktor]

Thanks for the quick graphing, neonblingbling. Testing is my thing more than graphing it out.

 

[neonblingbling]

If you want, just let me know when you have more data and I'll throw it in. I personally don't have the patience for lab work.

 

[muddocktor]

OK, I'm getting ready for round 2 of my fan testing series on these heatsinks. The new fans to be tested are:

Panaflo - L1BX (by NMB-MAT) - Model # FBA12G12L (X2)
Specs

• Fan Size: 120mm x 120mm x 38mm

• Speed: 1700 RPM

• Airflow: 68.9 CFM

• Noise: 30 dBA

• Static Pressure: 32.4Pa (3.30mm H2O)

• Bearing: Hydrowave bearing (essentially the same as the Sony FDB bearing)

• Power: 2.16w W

Scythe - S-Flex SS21G (X 2)
Specs:

• Fan Size: 120mm x 120mm x 25mm

• Speed: 1,900 RPM

• Airflow: 75.0 CFM

• Noise: 35.0 dBA

• Bearing: S-FDB (Fluid Dynamic Bearing by SONY Corporation)

• Power: 2.88 W

Scythe - Gentle Typhoon D1225C12B5AP-15 (by Nidec Servo Corp) (X 2)
Specs:

• Fan Size: 120mm x 120mm x 25mm

• Speed: 1850 RPM

• Airflow: 58.3 CFM

• Noise: 28 dBA

• Static Pressure: 20Pa (2.06mm H2O)

• Bearing: Double ball bearing

• Power: 4.32 W starting, 1 watt running

Late addition Fan
Panaflo - M1A (by Matsushita Electric) - Model # FBA12G12M (X2)
Specs

• Fan Size: 120mm x 120mm x 38mm

• Speed: 2100 RPM

• Airflow: 86.5 CFM

• Noise: 35.5 dBA

• Static Pressure: 47.1Pa (4.80mm H2O)

• Bearing: Hydrowave bearing (essentially the same as the Sony FDB bearing)

Audio impressions of these fans:

Gentle Typhoon - slight high pitched but very smooth sound to them. No perceptible bearing noise and would be very easy to live with in a push-pull config. The highly swept 9 blade rotor configuration really does a good job of cutting down tip noise from the blades.

Panaflo L1-BX - Not as highly pitched as the Gentle Typhoon, but more tip noise. You can hear a slight raspyness to the rotor noise, but the overall sound is both low and smooth. Another fan that wouldn't be hard to get used to. The 7 blade rotor definitely makes more noise, but then this fan also pushes more air than the Gentle Typhoon also, with significantly higher static pressure. No perceptible bearing noise.


S-Flex - These are the first of the "G" models I've ever owned and tested and they were actually noisier than I expected. The sound from them is lower pitched than the Gentle Typhoon but much raspier sounding from tip noise. They move much more air than the Gentle Typhoon and it's hard to tell if they move less or more air than the Panaflo fans. They are definitely noisier than the Panaflo fans and if you are very sensitive to noise I don't think these would make a good choice for you unless you hook them on a fan controller. No perceptible bearing noise.

Also, I would like to thank Gary at Sidewinder Computers for donating a Panaflo L1BX for this test. I had one already but he donated a brand new one along with the other that I ordered so I could test with 2 brand new Panaflo fans for the push-pull tests. Thanks very much Gary, it is much appreciated!

EDIT: I also want to note that the Gentle Typhoon fans I received are essentially closed corner fans, which will need to be opened up for use with the TRUE wire clips.

EDIT 2: The Panaflo M1A sounds much like it's lower speed brother except it's slightly higher pitched due to the faster rpm is spins and it's also louder. These are not fans to use if you have to have absolute silence. I would classsify them as in the same loudness class as the Scythe S-Flex G fans (which is actuallly reflected by the manufacturer's dB ratings they give for them).

 

[Zonnza]

Since I still havnt purchased my cooler, Im gonna hold up and wait to see how your 2x s-flex G fans run.

 

[ciku]

I was using CM R4's 2000rpm push config then exhaust... then today replaced with Scythe Slipstream 1200rpm same config, 1 fan as push then same fan for exhaust and my results are exactly the same.

Ambient Temp: 76c-77c
Idle Temps: 33c-35c
Load Temps: 42c-44c

Load temps meaning with wow, vent, rawr, coretemp, messenger, utorrent and vlc running.

 

[muddocktor]

Testing is still ongoing, and I decided to go ahead and add the Panaflo M1A fans into the test, since I have them already. So far I have tested the Gentle Typhoons, S-Flexes and am finishing the last run with the Panaflo L1BX fans on the Megahalems. Hopefully I will wrap up with the M!A fans this evening on it and then swap out to the TRUE before going to bed tonight. Tomorrow I can hopefully get all the fans tested on the TRUE, but that might be hard to do since it's my wife's birthday tomorrow too.

EDIT: One thing I ran across yesterday when researching the cfm and static pressure on the Gentle Typhoon fans caught my attention. As stated on the box, these are actually built for Scythe by Nippon Servo (Nidec) and going to their website and checking, I found that these are actually a product they were already manufacturing and not specially built for Scythe (just rebadged). There is nothing at all wrong with this and it's done all the time, but Nidec also makes a version of the Gentle Typhoon that Scythe doesn't carry, which spens at 2150 rpm and rated at 69 cfm and 30 dB. I would love to try this model out since I think that with their blade design the sound will still be very reasonable and the extra cfm and pressure will make them work better on a heatsink. Here is the page with the specs for the Gentle Typhoon fans on the Nidec website.

 

[Zonnza]

Mudd, if u get a chance, can u post or PM me what temps u had with 1 and 2 s-flex "G" fans on the mega? I want to make the purchase for my mega this weekend so it arrives before thanksgiving this next week and that would mean I would have to send the order out soemtime before tomorrow night. Having the info would really help me estimate how much the "F" version fans would give and I would be able to show it to my brother since he has been nagging me to get different fans all this time.

 

[muddocktor]

Fan testing, Round 2:

Thermalright Ultra Extreme:

TRUE with 2 M1A fans in push-pull config:
Max temp - 72.5
Min temp - 32
CPU1 max temp - 72
CPU2 max temp - 72.5
CPU3 max temp - 65
CPU4 max temp - 65

TRUE with 1 M1A fan in push config:
Max temp - 72
Min temp - 37
CPU1 max temp - 72
CPU2 max temp - 72
CPU3 max temp - 65
CPU4 max temp - 65

TRUE with 2 L1BX fans in push-pull config:
Max temp - 71
Min temp - 34
CPU1 max temp - 72
CPU2 max temp - 71
CPU3 max temp - 63
CPU4 max temp - 63

TRUE with 1 L1BX fan in push config:
Max temp - 73.75
Min temp - 35
CPU1 max temp - 73.75
CPU2 max temp - 73
CPU3 max temp - 66.25
CPU4 max temp - 66.25

TRUE with 2 S-Flex "G" fans in push-pull config:
Max temp - 71
Min temp - 35
CPU1 max temp - 71
CPU2 max temp - 71
CPU3 max temp - 65
CPU4 max temp - 65

TRUE with 1 S-Flex "G" fan in push config:
Max temp - 73
Min temp - 38
CPU1 max temp - 73
CPU2 max temp - 73
CPU3 max temp - 67
CPU4 max temp - 67

TRUE with 2 Gentle Typhoon fans in push-pull config:
Max temp - 73
Min temp - 36
CPU1 max temp - 72.25
CPU2 max temp - 73
CPU3 max temp - 65.75
CPU4 max temp - 66

TRUE with Gentle Typhoon in push config:
Max temp - 74
Min temp - 37
CPU1 max temp - 74
CPU2 max temp - 74
CPU3 max temp - 68
CPU4 max temp - 68


Prolimatech Megahalems Round 2 test results:

Megahalems with 2 M1A fans in push-pull config:
Max temp - 73 C
Min temp - 34 C
CPU1 max temp - 73 C
CPU2 max temp - 73 C
CPU3 max temp - 66 C
CPU4 max temp - 66 C

Megahalems with 1 M1A fan in push config:
Max temp - 74 C
Min temp - 34 C
CPU1 max temp - 74 C
CPU2 max temp - 74 C
CPU3 max temp - 67 C
CPU4 max temp - 67 C

Megahalems with 2 L1BX fans in push-pull config:
Max temp - 74 C
Min temp - 35 C
CPU1 max temp - 74 C
CPU2 max temp - 68 C
CPU3 max temp - 74 C
CPU4 max temp - 68 C

Megahalems with 1 L1BX in push config:
Max temp - 75 C
Min temp - 37 C
CPU1 max temp - 74 C
CPU2 max temp - 75 C
CPU3 max temp - 68 C
CPU4 max temp - 68 C

Megahalems with 2 S-Flex "G" fans in push-pull config:
Max temp - 72.25 C
Min temp - 33 C
CPU1 max temp - 72.25 C
CPU2 max temp - 72.25 C
CPU3 max temp - 66 C
CPU4 max temp - 66 C

Megahalems with 1 S-Flex "G" in push config:
Max temp - 75 C
Min temp - 36 C
CPU1 max temp - 75 C
CPU2 max temp - 75 C
CPU3 max temp - 68 C
CPU4 max temp - 68.5 C

Megahalems with 2 Gentle Typhoon fans in push-pull config:
Max temp - 81 C
Min temp - 37 C
CPU1 max temp - 80.5 C
CPU2 max temp - 81 C
CPU3 max temp - 73 C
CPU4 max temp - 73 C

Megahalems with Gentle Typhoon in push config:
Max temp - 83 C
Min temp - 39 C
CPU1 max temp - 83 C
CPU2 max temp - 83 C
CPU3 max temp - 77.5 C
CPU4 max temp - 77 C

This is posted after getting the Megahalems tests done and before I've run the tests with the TRUE. I will edit those results in after I get done with the TRUE

 

[CualvezAsteria]

Interesting the that Panaflo L1BX performs sooo much better than the GT, 7C better in push-pull, 8C better in push only! I have both and am beginning to think the GT is not worth the money.

 

[muddocktor]

Actually, I'm not surprised at all with the performance delta between them. I think that these fans that use a 9 blade impeller design lose out on max static pressure and the specs I dug up on the GT tend to prove it out, besides what I see in actual performance testing. I think the highly swept back feature of the blades on the design also play into this. The swept back design delays and mitigates transonic flow noise for a given tip speed, but at the expense of max static air pressure. I think that if Scythe would buy and market the 2150 rpm version of this fan from Nippon Servo, it would perform as well as the L1BX. As to the noise factor, who knows. I will say this for the GT; it definitely performs better than the Noctua fan for heatsink use. But I have the feeling that the lower rpm GT's will be much like the NF-12P in performance.

 

[neonblingbling]

Here is a graph for round two (as of now). Please note that these temperatures are the average max of the cores; I do not believe it is proper to convey the single highest temp.

 

[neonblingbling]

My interpretation: Interesting how a low speed high pressure (Panaflo L1BX) does not benefit from a push-pull setup (0.25C difference). Maybe at that pressure, any more airflow is not beneficial. Every other fan setup experiences from a 1C to a 3.25C difference. Then again, I would probably only use one fan in any case to cut costs/noise.

 

[JamesMore]

I know with out the correct equipment its hard to know noise levels but can I get your comments on how loud some of these fans are in the push pull setup.

 

[Zonnza]

This definitely made my order for a megahalem rev b with 2 SFF21F fans and MX-3 arctic cooling thermal paste, grand total of 113 bucks including shipping from sidewinder