Some new test equipment

Panaflo FBA12G12H(-1A) results are up. Being a new page, I'll repeat the graphs here. I'll endeavour to split the graphs out in a later post.

The Panaflo is a strong performer, as you can see, and it in part achieves this by having the strongest RPM -> airflow performance of any fan tested to date (barring the Delta TFB). It really comes into its own above ~9v. Having said that, the SPL graphs aren't showing how absolutely annoying the sound the fan makes is. The bearings have a rumble to them that is amplified through the enclosure, and generate a medium-tone droning noise. There's also a buzzing that emits from the motor and provides yet another annoying sound, and this too gets amplified by the enclosure. Also, the shroud + enclosure would just generally excite and sing at numerous points across the rpm range. Quite frankly I don't know how an item can make so much noise, and yet still register a low SPL. I think it's a bit like TV commercials, how they "compress" the various frequencies such that a wide range of frequencies are all at the same volume, creating a "wall of noise" that is perceived as being MUCH louder than a SPL meter would indicate.

I think that the Panaflo exposes a real weakness in the SPL graphs, in that there is nothing at all pleasant about the noise it makes, but looking at the graphs, you'd never be aware of it. This is where a general frequency analysis would really help to sort the pleasant from the annoying.

Noise vs Airflow through PA120.1 radiator at 5cm

http://www.employees.org/~slf/curves...-noise-5cm.png



Noise vs Airflow through PA120.1 radiator at 1m

http://www.employees.org/~slf/curves...w-noise-1m.png



Ambient Noise vs Airflow through PA120.1 radiator

http://www.employees.org/~slf/curves...-noise-amb.png

Quote:

---

Originally Posted by Cathar

This is where a general frequency analysis would really help to sort the pleasant from the annoying.

---

That's what I was attempting to suggest earlier.
Record each fan and display it as a histogram.

Ya just like we talked about, SPL only shows so much.

Could I have an RPM of the Yate SL at 7 volts? Excellent work!

Quote:

---

Originally Posted by Cathar

<----- That's me. Yamaha R1, 2000 model

Next bike which is coming in a few months will be a Triumph 675.

Enjoy some on-board footage:

http://www.employees.org/~slf/R1/corners.gif

---

I'd like to see the same footage at the curve exit in the 10 grand RPM range...you know when you start seeing the sky instead of the trees ?

Quote:

---

Originally Posted by gabe

I'd like to see the same footage at the curve exit in the 10 grand RPM range...you know when you start seeing the sky instead of the trees ?

---

Sadly, I'm no wheelie monster yet. I had only just started getting comfortable with getting the front wheel up under power. I have some videos of some rather crap wheelies, but they're more embarrassing than anything to show off.

Quote:

---

Originally Posted by Cathar

Sadly, I'm no wheelie monster yet. I had only just started getting comfortable with getting the front wheel up under power. I have some videos of some rather crap wheelies, but they're more embarrassing than anything to show off.

---

Hope you ride with your helmet and body armor on! Nothing more fun than a bike but nothing more deadly. My lab abuts a large university hospital complex and there, they are known as 'Donorcycles' for good reason.

The BMI stuff we work on with humans has as thirty percent of its patients former Donorcyclists.

Jay

Quote:

---

Originally Posted by jayhall0315

Hope you ride with your helmet and body armor on! Nothing more fun than a bike but nothing more deadly. My lab abuts a large university hospital complex and there, they are known as 'Donorcycles' for good reason.

---

Being a rider for some 13 years, I'm well aware of the risks, and I've seen my share of bike accidents. Fortunately I've never witnessed a fatal accident. Been involved in a couple of serious accidents at the race-track though, which is why I'm currently laid up with a broken wrist + ankle.

In Australia it is illegal to ride a bike without a helmet. Thing here though is that extremely few people complain about the rule. Most here appreciate that even falling over from a standing position and smacking your unprotected head against concrete/bitumen can be (and certainly has been) fatal, let alone falling off some vehicle at speed without adequate head protection.

As for wearing adequate body protection, it would seem that many still don't understand what they're dealing with. Coming off at 35mph onto bitumen is about the same sort of deal as falling on top of a spinning bench grinding wheel. Same sorts of speed/abrasion involved. Ask people if they would willingly grab a full-speed spinning bench grinder and they'd call you insane, yet the very same people will ride around on a motorbike without gloves or adequate protective clothing.

What you see in my avatar is what I wear ALL the time. 1.4mm thick kanagaroo hide leather (about twice as strong against abrasion as cow leather, and lighter), full face helmet with 150G's transferrance impact loading (as opposed to the more inadequate Snell rated 250G's impact transferrance). Basically that means less likely to suffer brain damage. Carbon fibre protected gloves, with wrist straps to prevent them pulling off in an accident. Underneath the leathers is articulated spinal body armor with aluminium honeycomb mesh crumple zone protection to absorb impacts to the spine. Also molded plastic body part inserts to spread any impact loading against the torso and joints. Boots are full leather, and include hyperflexion/hyperextension protection, and after doing a 1-legged 85mph full body-weight somersault pivoting over my ankle and only suffering hair-line fractures through the talus, I'd say that the boots did their job. Without them I would've snapped the ankle for sure.

Yeah, I take protection seriously. You have to when you get around a track at speeds of up to 170mph. Anything less is folly.

Sorry for the OTT, but how is your shoulder healing? I know mine wasn't fully "healed" for two years, and has never returned to full strength, but I pray yours does. Fake stuff in that area for me never was just right, here's to hoping that it heals properly.

On topic, would the aluminum CMs be a good match for a BI GTS 240? I think my fans (CM white LED jobbies) are more than a bit underperforming...

Quote:

---

Originally Posted by [XC] Kayin

Sorry for the OTT, but how is your shoulder healing? I know mine wasn't fully "healed" for two years, and has never returned to full strength, but I pray yours does. Fake stuff in that area for me never was just right, here's to hoping that it heals properly.

On topic, would the aluminum CMs be a good match for a BI GTS 240? I think my fans (CM white LED jobbies) are more than a bit underperforming...

---

My shoulder is now almost wholly fine now, thanks for asking. It's been ~15 months now. In fact I've had to rely on it lately 'cos I had broken the right-hand wrist, so I was doing everything with the left shoulder that I had broken last year. I'd say that it's at about 98%, and that's where it's likely to stay. I can do push-ups, pull-ups, and move it to ~98% full range of my right shoulder. Overall I don't even notice it. On occasion it'll twinge a bit, or get a little bit stiff with certain actions, but it never really stops me.

Two years does seem to be about normal for "as good as it's ever going to get" recovery from what I've heard. I was lucky that I got a really good orthopaedic surgeon, and he showed me the CT scans after re-aligning the bones and he did an absolutely fantastic job. I'm significantly less than impressed with the job my current surgeon did on my wrist, but the surgeon who did my shoulder was overseas at the time of the accident, and I couldn't get him.

My only recommendation to you is to keep moving the shoulder, and in fact go a little hard on it. Don't baby it. You may be doing so without realising it. I jumped back on the bike 10 weeks after my shoulder op and started riding around fairly aggressively by 12 weeks. It gave me grief but my physio was always amazed at how quickly and well it was healing because I wasn't trying to nurse the injury. In hind-sight, nurse a serious injury for 12 weeks, but after that it really does pay to push it.

As for the aluminium CM fans, avoid them. They work well in open air, but against restriction they don't work well, they vibrate, and the GTX radiators are more air-flow restrictive than the PA's.

Any of the Adda based fans (ArcticCooling, Scythe), or the Zalman ZM-F3, would be excellent at operating against restriction. The Yate-Loon's from PTS are extremely strong performers too, offering less vibration to excite enclosures of any of the other fans. Hard to really split out a winner here. Basically if you pick a Scythe/Zalman/Yate-Loon, just tune it to whatever noise-level you are prepared to tolerate, and chances are it'll be delivering close to the best air-flow for that noise-level that you can get. The CM fan's definitely won't give you that.

I've got attendant neurological damage, so 'm walled at 60-70%, and greatly lessened manual dexterity. I was back to playing concerts in six months, but I still can't sit through more than an hour long concert unless it's something I sit in my lap.

Will the LL standard ADDAs work, or should I look specifically for those? That's no issue, but I have a lot of LL fans (a serious fan here) and if they'll work, they're in.

My best wishes for you that the neurological damage repairs itself somewhat. From what I hear, that can take ages, or it can never recover. It might just "come good" one day. Nerves are odd things. Hear all sorts of odd stories.

I haven't personally tried the Lian-Li Adda's, so I can't really comment. They do have different bearings to the Scythe's though, so I'd be wary of claiming that they'd perform equally. I imagine they'd do okay, just with more bearing/motor noise.

Oh, in case people weren't aware, the Acoustifan Dustproof (AFDP-12025) fan results are now in the graphs.

Overall, the fan placed about exactly where I thought it might, given how it behaved in free air. The motor has a muffled rumbling tick to it that is amplified by the shroud and enclosure. At various points in the RPM range it excited the shroud and enclosure quite noticably. About the only thing that surprised me was that it still beat the CoolerMaster aluminium fan. I was also certain that the CM would be quieter, but the CM has terrible rpm->flow efficiency through the radiator, and I guess that explains what pushed the Acoustifan ahead. In fact, looking at the results I gathered, rpm vs noise wise, the Acoustifan is worse than the Coolermaster, just that the Acoustifan pushes more air per rpm and this more than made up for its noisiness.

This is a little odd, 'cos the Acoustifan is an Adda rebadge, but it's one of Adda's environmental fans which has a semi-sealed motor. They are very expensive (~$35US for a single fan last I checked). I can only speculate that I have gotten a bad sample? Maybe the bearings on my sample are damaged? Some people swear by these fans, but the vibrations alone indicate to me that even if the motor noise was quieter, it'd still excite enclosures more than a number of other fans.

Just replaced my Yate Loon SLs with the ScytheS-Flex SFF21F's. So far I'm very pleased. At 100&#37; they are audible, but it's like hearing a fairly quiet refrigerator running in the next room. Easily forgotten. And the flow is much better as well.

I was pretty dissapointed in the flow of the YL fans. Not that they were underperforming, I'm sure they were working up to spec, but I wanted more flow since I'm cooling everything on one loop.

I'll have to wait for the heat tomorrow to tell for sure but so far it seems like I've dropped overall temps a bit, maybe as much as several C.

I should add that right now I'm seeing these temps at room 72-74c (with a currently poorly lapped CPU - could only find 600 grit - and with AS5):
CPU 37c-38c
GPU 34c steady
NB 41c

Load:
CPU 51-53c
GPU 48c

I'll be completing a CPU lap to 1200grit or less and applying Arctic MX-1 to the CPU. Will see what the temps are after it sits for a few days.

are there any results on the noctua?

Quote:

---

Originally Posted by santiagodraco

I was pretty dissapointed in the flow of the YL fans. Not that they were underperforming, I'm sure they were working up to spec, but I wanted more flow since I'm cooling everything on one loop.

---

Sounds like the straight-blades.

Quote:

---

Originally Posted by Duh

are there any results on the noctua?

---

No. I don't have any here, nor am I likely to test it unless someone wants to send me one. I'm not spending money on a fan that has currently proven itself to be an under-performer when matched with air-flow resistance (ref. Vapor's tests, and SPCR's followup articles about it).

Quote:

---

Originally Posted by Cathar

Sounds like the straight-blades.



No. I don't have any here, nor am I likely to test it unless someone wants to send me one. I'm not spending money on a fan that has currently proven itself to be an under-performer when matched with air-flow resistance (ref. Vapor's tests, and SPCR's followup articles about it).

---

These are curved (the YL). Curved on the leading edge.

Quote:

---

Originally Posted by santiagodraco

These are curved (the YL). Curved on the leading edge.

---

Got a pic? I'll verify for you. "Curved" is more referring to the transverse curvature, not the leading edge curvature.

Quote:

---

Originally Posted by Cathar

Got a pic? I'll verify for you. "Curved" is more referring to the transverse curvature, not the leading edge curvature.

---

Ah, got ya. I'll take a pic and post it. Now you have me curious :)

Here's a couple shots:

http://skeetersspot.smugmug.com/photos/175311329-M.jpg

http://skeetersspot.smugmug.com/photos/175311337-M.jpg

Those are definitely what Vapor and I are referring to as "straight blades".

The curved bladed Yates, which to my knowledge are only obtainable from Petra's, are the better performing Yate-Loon's.

The curved bladed Yate's have a much greater transverse curvature, probably around twice a curved as what you see in your pics, if that makes sense.

Quote:

---

Originally Posted by Cathar

Those are definitely what Vapor and I are referring to as "straight blades".

The curved bladed Yates, which to my knowledge are only obtainable from Petra's, are the better performing Yate-Loon's.

The curved bladed Yate's have a much greater transverse curvature, probably around twice a curved as what you see in your pics, if that makes sense.

---

It does. So basically the straight bladed YLs push less air. These were certainly quiet, silent really, but just weren't cutting the CFM mustard. So far I'm pretty happy with the Scythes.

Thanks :)

The Scythes are good fans to date in my testing.

Hey, here's a combination fan data graph.

Am after some feedback on what you think of it. Will provide one of these such graphs for each fan tested. It allows you to cross-reference all data points in one place.

If you want to convert voltage to airflow, choose a voltage and follow it straight down until you reach the blue line. Then go horizontally across until you hit the green line, and then follow than point where you hit the green line straight down to reference the air-flow rate at that voltage.

http://www.employees.org/~slf/curves.../flow-test.png

santiagodraco....that is what I would call a 'fake' straight blade. At this point I'm not sure if it's a fake, a discard, or what....but all the "Yate Loons" bought through a certain distributor (that EVERYBODY stateside but Petra uses) are very different fans from the direct-from-YL fans. Not in a good way either.

There was a REAL straight blade design from YL that was for the SH-speed fan and seemed to have some minor advantages at higher RPM, but for whatever reason, the whole line is now using a curved blade design....not sure if this is a temporary change or what.

I don't want to go on too much longer....but most of what I know and have said can be found here: http://www.xtremesystems.org/forums/...3&postcount=69

Cathar - not sure if I follow the way to find CFM at a certain voltage.....would it be 35CFM @ 12V on the Acoustifan sample you just put up?

Cathar, it looks like you're trying to map a one dimensional line through a 4 dimensional space (the four axes). Why don't you use (4 choose 2) = 6 lines to display the results? That way, only 1 lookup is needed per pair of dimensions.

Oh crap, I forgot about the various noise levels. I suppose you could make a graph with 12 lines so you only need 1 lookup per pair of dimensions. On second thought, the graph is fine as it is.

edit: What the hell was I thinking when I wrote this???