120mm Fan Testing on an MCR120 Radiator Round 6

[Martinm210]

Thanks!
Trying to plow through these 120's as fast as I can, starting to get a bit boring, but I'm looking forward to finishing so I can give the 140's a go. Then I want to maybe look at some pump noise related testing after that.

[Martinm210]

Sanyo Denki San Ace 120 9G1212H101 38mm

Special thanks to "The Finisher" for sponsoring this fan..

NOISE LEVEL PER CFM
Relative to the yate medium baseline fan, it's a little worse at lower volts and improves at it ramps up. Relative to the two other 3000RPM 38mm fans, it's better by about 2-3dbA. So noise level is actually very good from a metering standpoint at high RPMs and is now leading the way from about 42CFM to 55CFM.

NOISE QUALITY
This is again where the sound meter fails to report the whole story, noise quality is simply not captured in noise level.

Relative to the yate loon and most 25mm fans, very poor below about 11 volts. It has a very pronounced motor tick that's fairly harsh in comparison with the smoother 25mm fans. I've heard that PWM will make a worthwhile improvement to this problem, but I'm testing voltage control here and the fan is very ticky at lower volts similar to most 38mm fans. Unfortunately I don't have a good 3pin PWM controller that I can operate without a computer, or I would do a comparison for you. At this point in being consistent with the previous fan ratings, I'm simply going to say keep it at 11V or above if you want smooth sound. I would not recommend undervolting much unless you can some how cover up the motor noise to smooth it out somehow.

If I compare the yate loon medium at 8V in one video window and pause it, then load this fan at 6V pause it. The noise level meter and anemometer have them producing roughly the same noise level and air flow performance level. Yet to my ear, the yate loon is far superior in noise quality. With the yate, I hear air movement, with the San Ace I hear a motor. Perhaps some folks prefer the motor sound. Regardless, the videos are there for the end user to make their own judgements about noise quality.

INDUSTRIAL QUALITY
I think the industrial build quality is worthy of note, these fans are built like a tank and of very high quality. I suspect that is why many folks seem to like them so much, they do make your typical 25mm fan feel like a cheap piece of plastic and there is something to be said about appreciating something of high industrial quality.

COST
PPCS has these fans for $22.99, so the fan is in the "HIGH COST" category, but similar to many of the industrial grade fans, so not unheard of.

OVERALL
Very good noise level per CFM above 2500RPM, noise quality is poor below 11V but smooths out at higher voltage, cost is high, and build quality is excellent.

I'll give the fan a thumbs up for it's lower noise level at 12V compared to the other two ultra high speed fans.

VIDEO LINK
http://www.youtube.com/watch?v=Y0bvShNDT4M

[Church]

Martin: maybe you should double (or even quadruple?) resolution of united chart and use thinner lines? It's totally blured out in middle that's made up from most of fans

[Martinm210]

Martin: maybe you should double (or even quadruple?) resolution of united chart and use thinner lines? It's totally blured out in middle that's made up from most of fans

Yeah, I'll work something up when I'm done. Maybe break it down into batches of 10 or maybe some other logical order such as max CFM level, etc.

Been thinking of some sort of bar chart system to suppliment as well.

In the mean time, you have the individual chart relative to the yate medium as a clean comparison with every fan..

[Manicdan]

a bar chart at a certain voltage with the names in the bar so we dont have to match lines with labels

edit, how thin can a bar be until it gets called a line? i guess we will find out

[prtuc2]

Is weird at first when I had the San Ace H101 on my megahalems I undervolt to 7V didn't noticed any ticking sounds. Now recently I test it again the ticking sound is more noticeable (fan by itself) not sure what is the problem.

[Martinm210]

Sanyo Denki San Ace 109R1212H1011

Special thanks to Vapor for sponsoring this fan

NOISE LEVEL
Very good, and exceptional at high speeds. From a pure noise level meter numbers standpoint darn right impressive for the range it is covering. This fan also did a bit better than the 9G version by a small amount. It also has a high CFM level.

NOISE QUALITY
I'm going to say poor from 4-6V, average from 7-8V, and good from 9V and higher. Lower volts have that typical 38mm fan tick, but by the time you reach 2000+RPM levels that motor noise pretty well washes away in the air noise.

OVERALL
A very strong noise level fan of industrial quality. Noise quality is a bit poor from 4-6V (motor tick), average around 7-8V, and good 9V and above. I'll give the fan a thumbs up for having the lowest noise level from about 40 to 61CFM of the fans tested so far....

VIDEO LINK
http://www.youtube.com/watch?v=4AdL3NcP4AE

[Utnorris]

So I listened to the video and it's funny, while I hear the tick you are talking about, I have 6 of these running on two PA120.3's (old style) and at 5v I do not hear any tick from them. I have heard it before, but the majority of my San Ace fans that I have do not make that noise. In fact, at 5v it's barely noticeable of any fan noise to the point where I check to make sure they are all running. It will be interesting to see if all three of the San Ace fans you have with the same model number do this, assuming you are going to test all three.

[Martinm210]

So I listened to the video and it's funny, while I hear the tick you are talking about, I have 6 of these running on two PA120.3's (old style) and at 5v I do not hear any tick from them. I have heard it before, but the majority of my San Ace fans that I have do not make that noise. In fact, at 5v it's barely noticeable of any fan noise to the point where I check to make sure they are all running. It will be interesting to see if all three of the San Ace fans you have with the same model number do this, assuming you are going to test all three.

Good to know. Maybe I'm worrying too much about this fan tick or noise quality side of things. It makes sense to me that the more fans you have the less obvious that would be as they would all be sort of blending together.

Perhaps that's the difference or perhaps this H1 microphone and ultra quiet test scenario makes it all seem worse than it really is..?

The only experience I had was some folks convinced me to buy Panaflo Ms a while back when I first started water cooling. I changed out my yate slow speeds for Panaflo Ms. They were too loud for my liking at 12V, so I tried to undervolt them to an acceptable noise level.

Even with 6 of those fans running on two MCR320's...the ticking on those fans drove me nuts. It wasn't much, but it was annoying to me and I much preferred the smoother sound the 25mm yates gave me and eventually pulled the panaflos out and went back to the yates..

Needless to say, I wasn't too happy I just wasted my money on 6 brand new panaflos on what was supposedly the much better fan per the forum recommendations. That's probably also a pretty good indicator that I'm easily annoyed..

I will try the others, but it'll be a while. Heading out of town here for the holidays and weekend, so no testing for a bit.

Anyhow, the 109R-H seems to be the lower noise level of all the 38mm fans tested so far, that's a really good indicator..

[Martinm210]

FYI,
Tried reducing the line weights to 1PT to see if this helps looking at the charts any...not sure it does, but here they are:

First one is all fans at all levels, second is reducing the axis limits to get a better look at the lower speeds:
ALL

All at lower levels only

[Lu(ky]

Hey Martin great job thank you for taking the time to do this. There are so many new fans out there now this really needed to be done.
Question about the Scythe GentleTyphoon 1850 fan. If this fan was in a 120mmx38mm frame would it push more CFM? Or have better pressure..
Thanks

[zalbard]

FYI,
Tried reducing the line weights to 1PT to see if this helps looking at the charts any...not sure it does, but here they are:

First one is all fans at all levels, second is reducing the axis limits to get a better look at the lower speeds:
ALL

All at lower levels only

Thanks, it helps a lot. On a side note, I think your GT14 info on the legend is wrong (should have squares on the line, I believe).

[asura]

Martin, I personally find the graph with thinner plot lines much easier to read, so that's one person you've helped out at least.

It's interesting the wobble that the GT's have, presumably it's at the same point as the wobble Nidec Servo report in their curves for the GT's

Lu(ky if it's static pressure we're aiming for, one of these should do the trick, twenty six times the static pressure of a GT, for only thirteen times the apparent loudness of course that's a subjective measurement... which just feels dirty

[SpuTnicK]

@Martin

Sanyo Denki San Ace 109R1212H1011
VIDEO LINK

are those golden screws that hold the fan? fancy looking

I was waiting for San Aces. I have about ten of H1011. And have listen to every one at 5V. I could not hear the motor noise so noticeable in your video.
But maybe you are right:

this H1 microphone and ultra quiet test scenario makes it all seem worse than it really is

Anyways, thank for results. Keep up the good work

[Church]

Martin: how about swapping noise / airflow axis, so that best fans in chart are actually "on top" , also extend with making airflow axis hmm .. 4-10 times more sparse. Of course it will result in very "high" image - but as it's stretched vertically, it can be easily scrollable up-down. I mean something like 1900x4800 or even 1900x9600? Should fit well width-wise in fullscreen on FullHD screens (leaving 20pt for scrollbar )

[Martinm210]

Thanks for the ideas and thoughts. It seems several people think the videos and audio are more sensitive than their own experiences. I'm not sure what to do about that other than refraining from commenting about noise quality. Personally I thought the audio matches really well to my own when sitting there doing the actual tests...so I' thinking it's simply due to the ambient noise level. If I listen to the fans in a normal room with normal ambient noise levels than the details wash away.

Perhaps testing in complete silence is the wrong idea?

Anyhow, I think I'll give up on trying to give you any of my own perceptions and simply run the recording.

I may even just kill the whole charting idea again...just record the videos and let you decide from that?

What do you all think?

[Martinm210]

Thanks for the ideas and thoughts. It seems several people think the videos and audio are more sensitive than their own experiences. I'm not sure what to do about that other than refraining from commenting about noise quality. Personally I thought the audio matches really well to my own when sitting there doing the actual tests...so I' thinking it's simply due to the ambient noise level. If I listen to the fans in a normal room with normal ambient noise levels than the details wash away.

Perhaps testing in complete silence is the wrong idea?

Anyhow, I think I'll give up on trying to give you any of my own perceptions and simply run the recording.

I may even just kill the whole charting idea again...just record the videos and let you decide from that?

What do you all think?

[ggdh]

Very impressive work

I'm looking forward for 140mm fans

[MagisD]

Martin keep what your doing the same. Those noises are still there no matter anybodys ambient which is different for ever single user. You method is a level playing feild. And quite frankly nobody can touch you when it comes to acurate consistent fan test.

Great job and keep up the good work.

If somebody wants it done differently let them build a setup collect the fans test them there way and process the work.

Sent from my X10a using Tapatalk

[zalbard]

Perhaps testing in complete silence is the wrong idea?

No way.
It's the worst case scenario. It is very accurate and easy to reproduce.
Plus it is exactly what I encounter when I go to bed and leave my PC running.

[Church]

To me chart is THE source of results, at least for first rough selection of fans that seem interesting, and only then i'd check recording of those select few. I have no slightest intention to read description for all 20-30 fans while listening to recordings for all of them and then have it all blended in mind by then. My short term memory is inadequate for that . Line chart imho simply is best & most compact way to represent results for actual choice of fan selection/purchase, though hindered a bit by large number of tested fans. But that shouldn't mean that it needs to be dropped totally because of this minor shortcoming with replacing it with something like videos-only that is usuable for comparing only very small number of fans.

[BrokenArrow]

Agree with above. Keep the charts.

[Martinm210]

Alright, I'll keep them, but I want to somehow incorporate the noise quality component.

This is sort of what I was thinking to help separate things out a bit..

The idea is to pick out the dbA numbers off the chart for each CFM bracket, then include some subjective ratings for quality. Noise quality ratings would be something like low/medium/high with an overall rating at the end of some sort.

[Martinm210]

Sanyo Denki San Ace 9G1212M101 38mm

Special thanks to the finisher for sponsoring this fan..

NOISE LEVEL
Relative to the yate medium, it's up to 2dbA louder at the lower end, about the same in the middle, and about 1dbA better at the high end and extends a fair amount beyond reach of the yate. Relative to the other fans tested, it was louder than average from about 4-6V, about average 7-9V, and slightly better than average from 10-12V

NOISE QUALITY
Very much like the 9G high speed, has a fairly pronounce motor noise at lower volts that doesn't really clean up until about 10-11V. While this is very subjective and extreme silence based testing, I have been recording every fan at the same Zoom H1 100% manual, low cut on, 12" distance. This fan has a more pronounced motor noise at lower volts, and I wouldn't recommend running under about 10V, it is really best suited running at 12V if possible. I do think the San Ace motor noise is generally smoother than that of the UK3 or PanoU. The UK3 has much more of a "Tick" where the San Ace is more of a motor bump as it turns. A few folks have expressed their own personal experience as one that they can barely here this noise I'm talking about.... So...take the noise quality notes with a grain of salt and listen to the videos with the understanding that this is worst case scenario testing here (absolute silence with an ultra high sensitive microphone) These types of noises are much less noticeable in a system where there are a multitude of there masking noises or ambient noises. As always, in order to split hairs, the testing sensitivity is turned up a notch. My suggestion is to find a similar fan you have on hand, and set the volume level to match what you are used to hearing, then start playing the videos and listen in..

OVERALL
Variable noise level and noise quality, with top notch build quality. Regardless, I much prefer the 9S, 25mm variety over these 38mm variants. The 9S models have exceptionally good noise quality in comparison. If you happen to be a San Ace fan "Fan", I would highly recommend looking more at the 25mm San Aces. While they did produce a higher noise level, I think their noise quality is far superior. The 9SH model also produced more CFM at 12V than this 9GM model did..

VIDEO LINK
Not yet processed at YT, but this link should work when it's ready:
http://www.youtube.com/watch?v=UaH1_qjMU5o

NOTE: I was not able to record RPM data in this test. I recorded what I could.








COMPARED TO ALL FANS TESTED SO FAR

[Utnorris]

Yeah Martin, don't take my comments as being I think they are wrong. Noise is subjective and I like worse case scenarios better than a "Rosy" picture. Everyone's hearing is different and while some might find one thing annoying others wouldn't be bothered by it, just look at how often people state pumps make a noise that bothers them. Anyway, I love the work, still trying to digest the data and figure out which would be the best rad fan. I am lazy in that respect which is why I want a list based on performance.