120mm Fan Testing on an MCR120 Radiator Round 6

[NaeKuh]

Martin i know it wouldnt be fair for the san ace, but they really change when placed in a shroud.

Have you tried them with shrouds?

I noticed the noise level is the first one variable that changes noticable.

OR are my fans just tripping out?

[OldChap]

Does Martin still have hair???

Can't help feeling that he will be pulling it out at the thought of re-testing with a shroud fitted

You make a valid point though ... when this is about sound and sound quality it is one possibility that must be investigated at least for a few of these even if only to prove or disprove the theory.

Shazza, good thinking there. Personally, I came to this wanting to know if there is a high speed (read performance) fan which would also be capable of being turned down when the rig is not being pushed hard. I am still looking for that one that overlaps the ap15 and Delta .....now just need to get another couple of fans to Martin.....and hope he is willing

[Martinm210]

OK ... I've been through all the data, looked at the charts, listened to the recordings, and trying to figure out what it all means

It's easy enough to point people to this thread for detailed info - but knowing there are folks who won't go through all the data, I've been thinking about how to explain this stuff in simple terms (and how I will use it myself).

Bear with me while I think as I type here - just wanting to make sure I'm on the right path.


Assuming I'm using a Swiftech MCR radiator with a fan controller. I want "good" performance, and I'm not looking for extremely high (or low) speed fans ... I've got a number of fans in mind, because I've used them and heard about others. So how do I use this data to help me decide? I've picked these out to examine -

1) Gentle Typhoons ... GT13, GT14, or GT 15 - to be determined (tbd)

2) Noise Blocker - but which one ... tbd

3) Zalman ZM-F3 - never used these, but they have previously received a lot of good feedback

4) Yate Loon - specific one tbd (have used them, and their price can't be beat)

5) Scythe S-Flex (picked these because I've used them and liked them)

6) Enermax Magma - have used them, and one of the few "red" fans without LEDs

7) Noctua - I like them on a heatsink - what about on a radiator?


Okay, 7 brands are more than enough for me to consider.

My next question: How much air do I need to move with these fans? After using the eXtreme PSU calculator, and some other estimator spreadsheets, I figure out I should be okay if I can run in the 1200 RPM range. Not too scientific, but my previous experience tells me this will work for me. But wait, I just read 15 pages about fans, and they aren't ranked by RPM. Ahhh ... I guess I should be more concerned about CFM. Looks like my previous fans have been in the 20 to 30 CFM range, so I'll start with that. Hey - I thought these fans were pushing 50 - 70 CFM

Sometimes I might want more airflow, and sometimes less, so I'd also like fans with a decent range.

And, noise - my hearing isn't what it used to be, so absolute silence isn't critical. But, I tend to use a lot of fans, so quieter is better. I wear headphones when gaming, so louder fans on occasion is also okay.

I'm going to start by looking at the chart that shows min and max CFM for my candidate fans:


Gentle Typhoons (minCFM - Max CFM) all readings rounded, since this isn't rocket science

GT13......................9 - 24
GT14......................8 - 28
GT15.....................13 - 35


Noiseblocker
M12-S3.................18 - 32
PL-2......................13 - 25


Zalman
ZM-F3....................15 - 37
ZMSF3....................16 - 30


Yate Loon
SL Petra...................15 - 30
SM Petra..................18 - 37


Scythe S-Flex
S-FlexF.....................9 - 31
KFlexH.....................12 - 38


Enermax
Magma.....................15 - 30


Noctua
P12...........................17 - 25
S12B.........................15 - 26


Not a lot of difference, but I'll rule out the ones that are below 30 on the high end of their CFM range, just because I want to have a little more flexibility at the high end, and I'm pretty sure airflow in a case environment is going to be less than in the test procedure. (edit ... just realized I should have kept the Yate Medium in for the next pass, but won't go back and add it now)


That Scythe KFlexH looks interesting (*goes back and reads the individual review) ... thumbs up from Martin, they're available from Amazon with prime shipping (just checking availability). I'll take this fan to the next round.

The Scythe GT 15 looks good ... on to the next round.

Same for Noiseblocker Multiframe M12-S1. Looks a little weird, but let's keep it under consideration.

And the ZM-F3 deserves another look,just because I've heard so much about it.


Now I've picked four fans to check out further. You can tell I'm biased to getting a "medium" speed fan that has more potential at the higher end. Simply preference on my part ... not meant to imply any real significant difference in performance.



Let's look at noise for my 4 picks to see if I can rule any out based on noise:


I looked at the summary chart in Post 1, and the bar graphs showing dbA @ 20CFM and at 30CFM. The dbA data at 20CFM and 30CFM, followed by Martin's subjective noise rating, shows the following:


KFlexH....................39 - 49 dbA, 8.5

GT15.......................35 - 43 dbA, 7.5

NB M12-S3..............39 - 49 dbA, 9.0

ZM-F3......................39 - 50 dbA, 8.3


I think any of these would work - now my choice is based on other factors like pricing, availability, and power. All else being equal, I'd go with the GT15s at this point - given the discernible difference to the human ear is about 3 dbA (as mentioned earlier by Martin).


But, I'd like to know if there are any big differences in Power requirements ... rather than looking up the specs (which may, or may not, be correct) I'm going to pick a number off of Martin's "CFM vs. Watts" chart. I'll pick 30 CFM as my comparison point:


KFlexH.........................0.70W

GT15...........................0.60 W

M12-S3.......................1.50 W

ZM-F3.........................1.10 W


Kind of interesting. I knew the GTs had low power consumption numbers, as I'm using them in a current build, and was surprised when I was doing my calculations for how many could fit on my fan controller.

When you have 12+ fans in a build, power consumption is worth considering.

A quick look at pricing (admittedly,this hasn't been my top criterion - but it's become more important since I've reached the age of living on a fixed income ):

The KFlexH is $12-14, the GT15 is about $16 when you can find it. The Noiseblocker is $20+, and the ZM-F3 can be had for less than $10.

(I still don't think you can beat the Yates for lower speed fans if you're looking for a bargain).

Some of us are particular about how a fan looks. For a quiet no-frills black fan, I'd be inclined to give the KFlexH a go. I might choose the Noiseblockers if the color combination fit with my build.

So ... that's how I'd use the data to pick a fan. Of course, I didn't have to do all of that, did I? The GT15 turns out to be my top choice, and I could have picked that up from the first chart.

But ... I had no idea the KFlexH even existed, so good learning there. I also now know to check my power setup if I pick up some Noiseblockers. And, I also understand where my existing Scythe S-Flex, Yates, and Enermax fans fit in terms of noise and CFM performance.

Whew ... I'm tired now. Good thing it's tail hour in Florida!

to Martin for all his hard work.

I like your workflow, that's a great decision process and example..

I tend to use the "Listen blindly by matching up FPM only" video approach as well. In the end if you're deciding between a few fans, just get a couple of windows going on the videos and match up similar air flow levels and pause each video. Then simply play each video back and forth and simply listen blindly (close your eyes) and go back and forth between fans. I tend to be one that doesn't trust meters or instruments very well. Noise level is one thing we can quantify with a tool, yet it doesn't at all fit what is annoying, etc. The good ole...trust your ears first...forget the numbers for a bit method..

Martin i know it wouldnt be fair for the san ace, but they really change when placed in a shroud.

Have you tried them with shrouds?

I noticed the noise level is the first one variable that changes noticable.

OR are my fans just tripping out?

No, I only tinkered a little with shrouds, but I plan to do some of that after I get all the main shroudless testing done. I suppose I could try a San Ace if that's what you guys want.

I only did one quick shroud vs. no shroud test with the yate, and couldn't really see any difference, but there are a few ways you could make a shroud. Do you bolt all the way through, or do you bolt the shroud to the rad, then the fan to the shroud to help isolate vibrations. Do you use grommets, or certain gaskets. The options are endless, and figured I'd save some of those tinker tests for last.

[shazza]

I am still looking for that one that overlaps the ap15 and Delta

+1 on that. That's really where my next interest lies as well. Just trying to make sure my simple mind can comprehend all the data.

I'm really grateful to Martin for doing this, but also don't want to burn him out! For example, it would be fun to see how some GT fans that have been painted compare to the originals - but that might drive the poor guy over the edge

[Martinm210]

+1 on that. That's really where my next interest lies as well. Just trying to make sure my simple mind can comprehend all the data.

I'm really grateful to Martin for doing this, but also don't want to burn him out! For example, it would be fun to see how some GT fans that have been painted compare to the originals - but that might drive the poor guy over the edge

I've could probably do that to my wonky GT15, but it's pretty inconsistent, so I could very well come back and tell you it's and improvement..

Vapor said he's ok with my keeping these, so maybe I can try that at the end. It'll probably be kawasaki green or something like that though...

[Martinm210]

FYI, these are next on the testing short list, thanks to Old Chap for two more Deltas and RatDog for the bottom three including some more PWM options...


I think some of these detlas should dial down better than the previously tested VHE that ruled the high speed area..

[penguins]

I tend to be one that doesn't trust meters or instruments very well.

This is Hilarious : )

@Shazza, what's a Bananas Tail? : )

and that's what I've been doing to figure out these fans. minus any sort of organization or thought process like you though lol.

this is coming along!!!

that scythe looks loud..

[Vapor]

I don't need any of the fans back

[Creekin]

@Shazza, what's a Bananas Tail? : )

@ censorship

[NaeKuh]

I only did one quick shroud vs. no shroud test with the yate, and couldn't really see any difference, but there are a few ways you could make a shroud. Do you bolt all the way through, or do you bolt the shroud to the rad, then the fan to the shroud to help isolate vibrations. Do you use grommets, or certain gaskets. The options are endless, and figured I'd save some of those tinker tests for last.

Anything to space the fan out a bit so you get the full vortex of the blades.

I noticed spacing them out, you dont hear that whine noise they make @ 7v.

Also i noticed a lot more air going though.

[OldChap]

So, how much space 25mm, 38mm, 50mm??? Yep I'm thinking of using fan frames.

Ultimately I suppose a duct from the rad with a fan at the far end.....

[Martinm210]

OK, I'll plan on doing a little shroud experiment run. I need to do one for the 140 vs 120+adapter controversy anyhow. I've got a few gutted 120 fans lying around, so I'll plan on running the San Ace120, yateSM with/without shrouds and maybe even try 25 vs 38mm or more depending on time.. Going to save that shroud work for later though, the new fans and PWM work is next in line.

[nex_73]

So, how much space 25mm, 38mm, 50mm??? Yep I'm thinking of using fan frames.

30mm is about right? I'm using old 25mm fan frames though
http://www.xtremesystems.org/forums/...d.php?t=193953
If you have the space for it:
push -> shroud -> rad -> shroud -> pull

High speed fans:
push -> shroud -> rad -> pull
push -> shroud -> rad

Low speed fans:
push -> rad -> shroud -> pull
rad -> shroud -> pull
--------------------------------
Looking forward to the test of the upcoming fans, think we will find the lower speed Deltas really closing the gap between the GT and the rest...

[jalyst]

Martin thank-you so much for all your amazing work, I'm in awe of the time you're able to set-aside to help the community!

Coming back to this for the 1st time in months, only just been advised by a local retailer (aust) I pre-ordered with, that my AP-15's are available.
Apparently they've been ready to pick-up for many weeks!

I plan to read this latest round in it's entirety tomorrow, but as it's now 320am I must get some shut-eye.
Just wondering if anyone can summarise, is the AP-15 still considered the noise/CFM king, or is there a new contender or contenders?

How does the 2150 & 3000 RPM compare to the 1850, are either a better alternative IYO?
Perhaps the 2150 is not a substantive enough improvement over the 1850?

I have some AP-15's ordered as rad fans, but if the CFM to noise ratio of the 2150 or 3000 is as good, & one can still make them comparatively quiet (as needed)...
...then perhaps they're a better option?

Any thoughts/experiences greatly appreciated, thanks/night!

[Martinm210]

Martin thank-you so much for all your amazing work, I'm in awe of the time you're able to set-aside to help the community!

Coming back to this for the 1st time in months, only just been advised by a local retailer (aust) I pre-ordered with, that my AP-15's are available.
Apparently they've been ready to pick-up for many weeks!

I plan to read this latest round in it's entirety tomorrow, but as it's now 320am I must get some shut-eye.
Just wondering if anyone can summarise, is the AP-15 still considered the noise/CFM king, or is there a new contender or contenders?

How does the 2150 & 3000 RPM compare to the 1850, are either a better alternative IYO?
Perhaps the 2150 is not a substantive enough improvement over the 1850?

I have some AP-15's ordered as rad fans, but if the CFM to noise ratio of the 2150 or 3000 is as good, & one can still make them comparatively quiet (as needed)...
...then perhaps they're a better option?

Any thoughts/experiences greatly appreciated, thanks/night!

Thanks!

GT's are still doing very well in the 1000-1850RPM range, although it does have some quirks. I've had two AP-15's with resonance issues depending on the radiator, so some sound quality issues to be aware of. Below 1000RPMs, I think they are all fairly close (within the 3dbA "barely perceptible" range, so plenty of options to choose from at the low end.) 1850+, some of the newer Delta's are also performing similarly like the VHE which came out a good 5dbA lower in noise level than the next best high speed fan. Unfortunately, the Delta VHE is pretty much high speed only and don't like turning down much. This is why there has been alot of interest in exploring some of the other Deltas....hoping to find one that bridges the gap a little better. Plan to test a few more this weekend...

Unfortunately, it seems you're still always best off picking a fan that can run at or near 12V at your desired noise level if possible. Same goes for the GT family, no sense buying GT15s and undervolting them to 7V all day...the AP13 or 14 performs slightly better at slower speeds and has less of an issue with resonance.

In the end, I try not to get hung up on any one fan...none of them are perfect, and as noise always goes...pretty subjective. I tend to be one not to trust numbers or meters anywhere near as much as my own ears, and that's why I've put much of my time into recording the whole process. Trying to record it in the best video/audio quality that I can so you can do you're own review.

My suggestion will always be...let YOUR ears do the review.. Compare multiple videos and match up like air flow numbers, and see which one "sounds less annoying" Just don't put too much emphasis in the noise level numbers, I've been tempted many times to completely remove the sound level meter and just have the audio to listen to. I'm still just not very happy with the common tools we have to quantify "annoying sound"...but it is what it is..

As far as the higher speed GTs...I have no idea. I know some folks on OCNs air cooling forums managed to find them by special order through some Japan sites, but it looks like they are paying $60+ a piece for them and their subjective tests seem to indicate they were fairly loud... So until they become readily available, not much sense in testing them.

But, I'm always willing to try one (if someone wants to sponsor) until the wife kicks me and my test rig out of the walk in..

[bing]

As far as the higher speed GTs...I have no idea. I know some folks on OCNs air cooling forums managed to find them by special order through some Japan sites, but it looks like they are paying $60+ a piece for them and their subjective tests seem to indicate they were fairly loud... So until they become readily available, not much sense in testing them.

Yes, they're damn expensive like this "MEAN"-Typhoon.



Shot with ordinary GT1850.

D1225C12BBZP-00 with rectifier ring at the blades for high speed.

Max speed : 5400 rpm , 150.1CFM , 1.14 Amp , 150 Pa / 0.6 inH2O , 50.5 dB,
Its the PWM version (4 wires) with minimum speed 1000 rpm at 0% duty cyle.

Non PWM part no is D1225C12BBAZ-00

It has the beastie CFM, imo one of the most powerful 120x25mm, 12volt fan and also has a very wide range from silent at 1000 rpm up to max speed at 50db that covers from mid to high cfm range while at 25mm thick.

[Church]

I somehow doubt in GTs still being best above range of common varieties of them as well. After all, specific shape of blades can work well at specific range of rpm-s. In pics it looks rather unchanged for those extreme GTs. GTs shine in 900-1800 range, what's wonderful that it's most commonly used range noise wise (900-1400), below there are more silent fans, and i'm guessing - above aswell. I don't expect miracle of it being best everywhere. Imho 2000 and above fans with more fit for high rpm blade shapes should show better cfm/db ratio. Of course, i might guess wrong and someone can prove by real testing of those extreme GTs.

[nex_73]

Yeah would be interesting to see if the high speeds D1225C might cope with the Deltas (AFB models).
I see B6=2150, B7=3000, B9=4250 & BB=5400

http://catalog.nidec-servo.com/digit...general/g_fab/

You might be able to get them trough:
http://www.cliffinc.com/products/fans/index.htm

[jalyst]

Yes, they're damn expensive like this "MEAN"-Typhoon.

...

Shot with ordinary GT1850.

D1225C12BBZP-00 with rectifier ring at the blades for high speed.

Max speed : 5400 rpm , 150.1CFM , 1.14 Amp , 150 Pa / 0.6 inH2O , 50.5 dB,
Its the PWM version (4 wires) with minimum speed 1000 rpm at 0% duty cyle.

Non PWM part no is D1225C12BBAZ-00

It has the beastie CFM, imo one of the most powerful 120x25mm, 12volt fan and also has a very wide range from silent at 1000 rpm up to max speed at 50db that covers from mid to high cfm range while at 25mm thick.

What, a 5400 version, I thought the highest GT was 3000? Crazy!
I love that it's still got great RPM variability, & is still only 25mm.

Tempted to get 2x D1225C12BBZP-00 instead of 2x AP-15.
Then again they're prolly tricky to get hold of, & as-you-say, pricey.

Where did you get yours from?

[jalyst]

Thanks!
GT's are still doing very well in the 1000-1850RPM range, although it does have some quirks.
I've had two AP-15's with resonance issues depending on the radiator, so some sound quality issues to be aware of.
<SNIP>
But, I'm always willing to try one (if someone wants to sponsor) until the wife kicks me and my test rig out of the walk in..

Thanks Martin,
I'm using the MCR-X20 rev.2, it came with my H20-220 Edge kit, hopefully the AP-15 has no issues with it?
I'd definitely be interested in hearing more about these potentially more flexible Deltas you speak of.

AP-15's it is then I think....
As I'd be fine with running at or near 12v most of the time, so long as they happily volt down a bit with a controller if/when needed.

I also want to find the cheapest price for a 3000 (or higher) GT, buy 1x, & send it to you.
So if anyone knows the best spots to buy, please let me know!

[jalyst]

I somehow doubt in GTs still being best above range of common varieties of them as well. After all, specific shape of blades can work well at specific range of rpm-s. In pics it looks rather unchanged for those extreme GTs. GTs shine in 900-1800 range, what's wonderful that it's most commonly used range noise wise (900-1400), below there are more silent fans, and i'm guessing - above aswell. I don't expect miracle of it being best everywhere. Imho 2000 and above fans with more fit for high rpm blade shapes should show better cfm/db ratio. Of course, i might guess wrong and someone can prove by real testing of those extreme GTs.

Good point, it remains to be seen i think, I want to find the cheapest place to buy a 5400 RPM GT, I will buy and send one to Martin

[Manicdan]

I somehow doubt in GTs still being best above range of common varieties of them as well. After all, specific shape of blades can work well at specific range of rpm-s. In pics it looks rather unchanged for those extreme GTs. GTs shine in 900-1800 range, what's wonderful that it's most commonly used range noise wise (900-1400), below there are more silent fans, and i'm guessing - above aswell. I don't expect miracle of it being best everywhere. Imho 2000 and above fans with more fit for high rpm blade shapes should show better cfm/db ratio. Of course, i might guess wrong and someone can prove by real testing of those extreme GTs.

i wouldnt care if you were proved right or wrong, i just want to see martin test a 5400rpm GT

[OldChap]

D1225C12B7AZ-00 is 3000 rpm GT and is about $28 plus shipping from Japan

D1225C12BBZP-※ 00 is the 5400 rpm pwm GT and is about $53 plus shipping from Japan

Just one source checked for this.

http://translate.google.co.uk/transl...58%26prmd%3Div

[jalyst]

Crap $53USD that is a lot for just one fan (I'm Australian), not sure if I can afford it...
Martin if you can give me your shipping address, I can work out how much extra shipping $ is required.

[bing]

jalyst, that came from the site pointed by OldChap, and unfortunately they ship only to Japan.