PDA

View Full Version : flow directing shroud



iandh
04-03-2010, 04:54 PM
This is an idea I'm toying with, I'll probably order some material on monday to make up a prototype. This is basically a ripoff of the scoops you see on some of the high speed delta fans.

I think there is some optimization to be done on the blade shape/count, but this is a good start.

The problem is, with traditional shrouds, the air coming off the blades isn't really traveling straight, in fact from simulations I've seen the blades create vortices.

When the fan is closer, this is less of an issue because the air is scooped directly into the fins, but then you are faced with a dead spot. When the fan is spaced away with a shroud, you have a reduced dead spot, but you have more room for turbulence to happen.

With this I'm hoping to get the best of both worlds. If I get the design right, it should be possible to straighten out the airflow before it enters the rad, and it may even improve the performance of low static pressure fans on radiators.

That would be pretty cool if all of a sudden fans like the slipstream that weren't even an option before started to perform well...

Thoughts?

http://www.ihacc.com/pics/shroud1.jpg
http://www.ihacc.com/pics/shroud2.jpg

overclocking101
04-03-2010, 04:58 PM
nice! looks like a sick idea. hope it works out to, at least the shroud will have a function and be worth the cost of buying one.

Boulard83
04-03-2010, 05:21 PM
I dont think its going to help ... the basic idea of a shroud is to equalize the flow over the radiator area.

This shroud may lead to worse result than without it cause it throw the air on fewer spot on the RAD.

Testing need to be done but i dont think this can really help. The idea isnt bad but ...

eternal_fantasy
04-03-2010, 06:04 PM
Silverstone's airflow concentrating design:

http://tv.hexus.net/show/2010/03/CeBIT_2010_It_s_not_size_that_matters_with_the_Sil verstone_SG07/

http://www.techpowerup.com/img/10-03-06/silverstone1.jpg

AFAIK the 3 large fans on the bottom of the Silverstone Raven 2 case had a fan grill with the same goal but the design was a thick honey combed plastic shroud to direct the airflow's path and avoid the air vortex.

eternal_fantasy
04-03-2010, 06:14 PM
http://www.silverstonetek.com/images/products/ff121/FF121-PACKAGE.jpg

Introduction
Finding a quiet and efficient cooling fan is the ultimate goal for every
enthusiast. The FF121 is a fan grille designed for easy installation with any
120mm fan. The honeycomb pattern on the grille not only reduces wind shear
noise but also channels the airflow into more focused beam. Its integrated
filter also provides reduced dust buildup to extend the life cycle of
components. All in all, the FF121 is an excellent upgrade for any 120mm fans
in your system.

Special Feature
Honeycomb grille increases airflow focus
Integrated filter provides class leading dust reduction
Highly compatibility with all 120mm fan


http://www.silverstonetek.com/products/p_contents.php?pno=FF121&area=

masska
04-03-2010, 06:26 PM
+1

Nexus (http://www.nexustek.nl/NXS-nexus-beamair-airguider.htm) also has a similar design and I think were the first to come out with this concept. A shroud body with these designs and even a frame similar to the NB-Multiframe fans would work.

eternal_fantasy
04-03-2010, 06:35 PM
+1

Nexus (http://www.nexustek.nl/NXS-nexus-beamair-airguider.htm) also has a similar design and I think were the first to come out with this concept. A shroud body with these designs and even a frame similar to the NB-Multiframe fans would work.

Yeah one of the website reviewing the Raven 2 linked that Nexus shroud as a comparison... though couldn't find the link searching for it on google earlier :(

shazza
04-03-2010, 06:38 PM
Very interested to see how this turns out - keep us informed, please.

lowfat
04-03-2010, 06:39 PM
+1

Nexus (http://www.nexustek.nl/NXS-nexus-beamair-airguider.htm) also has a similar design and I think were the first to come out with this concept. A shroud body with these designs and even a frame similar to the NB-Multiframe fans would work.

I've never been able to find these in stock anywhere.

I was unaware that Silverstone made a similar product though. Might have to grab a few of those.

masska
04-03-2010, 07:26 PM
Review Link 1 (http://www.dragonsteelmods.com/index.php?option=com_content&task=view&id=9939&Itemid=1) 2 (http://www.dreamwarecomputers.com/reviews_2008/nexus_beamair/index.htm)

iandh
04-03-2010, 09:12 PM
I dont think its going to help ... the basic idea of a shroud is to equalize the flow over the radiator area.

This shroud may lead to worse result than without it cause it throw the air on fewer spot on the RAD.

Testing need to be done but i dont think this can really help. The idea isnt bad but ...

If it "didn't help", why would Delta incorporate this on a line of fans they've been selling for years?

How would it put air on fewer areas? The entire underside of the shroud is open and this would do nothing to reduce static pressure, in fact it would likely increase it.


http://www.silverstonetek.com/images/products/ff121/FF121-PACKAGE.jpg

Introduction
Finding a quiet and efficient cooling fan is the ultimate goal for every
enthusiast. The FF121 is a fan grille designed for easy installation with any
120mm fan. The honeycomb pattern on the grille not only reduces wind shear
noise but also channels the airflow into more focused beam. Its integrated
filter also provides reduced dust buildup to extend the life cycle of
components. All in all, the FF121 is an excellent upgrade for any 120mm fans
in your system.

Special Feature
Honeycomb grille increases airflow focus
Integrated filter provides class leading dust reduction
Highly compatibility with all 120mm fan


http://www.silverstonetek.com/products/p_contents.php?pno=FF121&area=


+1

Nexus (http://www.nexustek.nl/NXS-nexus-beamair-airguider.htm) also has a similar design and I think were the first to come out with this concept. A shroud body with these designs and even a frame similar to the NB-Multiframe fans would work.

I've seen all of these products, none are ideal in design. None help to eliminate the dead-zone, one of the key reasons for using a shroud, the silverstone is too restrictive, and the Nexus works but is not an optimal design. Both the silverstone and Nexus designs are for redirecting the air flung outwards by an unrestricted fan outlet.

My design, along with the Delta, actually help to scoop more air from the fan blades (the silverstone "air concentrator" does to some degree as well, but it also is restrictive). They aren't simple flow redirectors. The main advantage of my design over all three, is that it actually attempts to correct the dead zone, which is the whole point of using a shroud in the first place.

Any gains seen by the Nexus would be offset by losses at the dead zone... maybe not so much on air coolers, but definitely on radiators. It works perfectly fine for airflow concentration in a case, such as in the test that Masska posted. edit: In fact, the Nexus would likely make the dead zone worse if used on a radiator. Same goes for the silverstone.

Might I remind everyone, the silverstone design is just a hyped adaptation of the Delta fan design which has been around for years and years. That isn't "Silverstone's air concentrating design", it's a design they ripped off from Delta, just like I did. Delta ripped it off from jet impellers.

The Nexus design is no different than the filter outlets on some laminar flow benches, a design that has been around for decades. Hexagonal arrays have been used for airflow aligment for some time. In fact, I think I've even seen fan test benches using them prior to the release of the Nexus product. edit2: In fact, I just remembered the origins of the Nexus design. SPCR I believe used a bundle of straws as an airflow redirector in their test rig, and it eventually evolved into the Nexus design. Either way, the Nexus is totally different design, meant for a totally different purpose. On top of that, the laminar bench at my work uses a hex matrix on the blower outlet, and it's older than the personal computer age. I think my old boss bought it in the sixties.


Up until this point, nobody has designed a product that corrects both airflow pattern AND dead zone. That is the whole point of my shroud design.

Boulard83
04-03-2010, 10:30 PM
Shrouds help for noise a lil bit and they mostly kill the deadspot in the WC world. Im using 50mm shroud with my Gentle Typhoon on a TFC360 and i can tell you that i have nearly no dead spot on the rad.

I how delta that your refering to but they are only directing the air flow and not filling the deadspot.

http://electronics.toosharp.com/images/delta9b.jpg

And the honeycomb thing reviewd wont help for deadspot either...

eternal_fantasy
04-03-2010, 10:34 PM
Your design doesn't change the fact that the airflow will still be flung towards the outside of the shroud, only that they will have a greater velocity and pressure away from the fan. Unless you have fins or design that direct air from the fins of the fan to the center "dead spot", your design will not work as you intend it to. If you somehow combine your design with the design on the first picture I posted, where it tries to redirect the airflow towards the center, you might have something on your hands.

Boulard83
04-03-2010, 10:40 PM
Like the Delta fan i just posted ;)

iandh
04-03-2010, 11:39 PM
Shrouds help for noise a lil bit and they mostly kill the deadspot in the WC world. Im using 50mm shroud with my Gentle Typhoon on a TFC360 and i can tell you that i have nearly no dead spot on the rad.

I how delta that your refering to but they are only directing the air flow and not filling the deadspot.

http://electronics.toosharp.com/images/delta9b.jpg

And the honeycomb thing reviewd wont help for deadspot either...

You just repeated the same thing I said in the post I made above yours? :confused:


Like the Delta fan i just posted ;)

Yes, the same fan I mentioned in the very first sentence in the first post of this thread. ;)


Your design doesn't change the fact that the airflow will still be flung towards the outside of the shroud, only that they will have a greater velocity and pressure away from the fan. Unless you have fins or design that direct air from the fins of the fan to the center "dead spot", your design will not work as you intend it to. If you somehow combine your design with the design on the first picture I posted, where it tries to redirect the airflow towards the center, you might have something on your hands.

The airflow won't be "flung", when it hits the vanes, it will be diverted to an angle perpendicular to that which it impacts.. both inwards and outwards, so in effect will be injected into the dead zone. The portion that is deflected outwards will still have residual outwards momentum, but that is irrelevant. The portion which is deflected inwards will actually serve to inject more airflow into the center, not reduce it. Look at the vanes more closely, and try to imagine what would happen to the air as it hits them. Part of the purpose of the vanes is to break up the outward momentum and convert its direction. Also, don't discount the importance of pressure buildup inside the shroud. There's a reason why shrouds become less important with lower velocity fans.

I worked for a decade at an aerospace contractor, with windtunnel testing experience, and have an at least rudimentary knowledge of aerodynamics... I'm not just guessing here. ;)

The first design you posted only makes the airflow more laminar, it does nothing to correct the deadspot. The only reason the deadspot would eventually dissipate is because of pressure from the outer portion of redirected air. My guess it it wouldn't be until at least 5-6 inches away from the fan. The deadspot would still be very pronounced at normal shroud distances. It is inferior to my design (edit: for radiator use. It is fine for focusing airflow inside a computer case, which is exactly what it is intended to do). It is no more effective than Delta's solution, it just performs the job in a smaller space at the cost of restrictiveness.



You guys do understand that the top picture is the side that would be facing away from the radiator, right? The bottom picture is the side facing the radiator. The fan would be mounted in a push configuration on the top of the shroud. Other than that, I can't see how one could possibly confuse the functionality of my design with these others posted.

Vampiyer
04-03-2010, 11:41 PM
iandh is no clown, and that render of his is obviously rough. He stated that the blade design would need to undergo optimization.

I suppose the Delta design can be used to straighten most of the flow, but perhaps a deviation in blade curvature is needed toward the mid-inner portions to direct flow into the dead spot (or this could be addressed throughout the entire flow-directing blade - instead of directing air straight, it could be slightly inclined toward the center).

I'd want to use a 10mm blank shroud in front of the directing shroud (to reduce cavitation and turbulence noise from having the fan blades so close to the flow-directing blades (which must obviously have an opposing geometry)). However, that might defeat the purpose of directing the flow somewhat...

Hopefully you're able to run some CFD simulations or something... not sure how involved the project is.

Edit: And that Silverstone Fan actually MUST have a deadspot for that balloon to remain in the air above it. How else would it be held in place? Pressure differential it seems...

iandh
04-04-2010, 12:25 AM
iandh is no clown, and that render of his is obviously rough. He stated that the blade design would need to undergo optimization.

I suppose the Delta design can be used to straighten most of the flow, but perhaps a deviation in blade curvature is needed toward the mid-inner portions to direct flow into the dead spot (or this could be addressed throughout the entire flow-directing blade - instead of directing air straight, it could be slightly inclined toward the center).

I'd want to use a 10mm blank shroud in front of the directing shroud (to reduce cavitation and turbulence noise from having the fan blades so close to the flow-directing blades (which must obviously have an opposing geometry)). However, that might defeat the purpose of directing the flow somewhat...

Hopefully you're able to run some CFD simulations or something... not sure how involved the project is.

Edit: And that Silverstone Fan actually MUST have a deadspot for that balloon to remain in the air above it. How else would it be held in place? Pressure differential it seems...

That render is *rough as hell*- it is a concept drawing only. The actual blade shape will be a much more sweeping curve... think reverse gentle typhoon. This design has been in my head for months... that drawing was in my CAD program for about five minutes. I have more homework to do before I determine the exact ideal blade shape.

I should be able to minimize the noise issue as long as the geometry at the edge is correct. Keep in mind, the fan will be in a push config sitting on top of the side of the shroud pictured in the first image, so it will aready have a built in 5mm spacer due to the hub supports. :)

I do have access to FEA/CFD software, but it is slow going because I have to arrange for it. The majority of testing will be done with about five different inital prototypes to give me a starting point.

As far as that silverstone fan goes, I'd bet my sig rig there is a deadspot @ 1". If there wasn't, it would be flying in the face of physics. The only reason it would be minimal for the fan size is because that fan has a proportionally tiny hub.




I hope I'm not coming off as being Mr. Know-it-all or anything, but I NEVER publicly post an idea unless I've thought it through first... the only question for me is how well will it work, not if it will work. ;)

Alexandr0s
04-04-2010, 12:45 AM
My first remark on your design is that the blades are straight. I'm not sure if you plan on changing that, but as you can see on the Deltas, the directional blades are curved and placed in opposite direction of the actual blades themselves (the blades go clockwise, the directional blades go anticlockwise). Maybe that's a change you could incorporate into your next design.

Besides that, it's an awesome idea, and I'd love to see how it works out! :D

iandh
04-04-2010, 12:47 AM
My first remark on your design is that the blades are straight. I'm not sure if you plan on changing that, but as you can see on the Deltas, the directional blades are curved and placed in opposite direction of the actual blades themselves (the blades go clockwise, the directional blades go anticlockwise). Maybe that's a change you could incorporate into your next design.

Besides that, it's an awesome idea, and I'd love to see how it works out! :D


That render is *rough as hell*- it is a concept drawing only. The actual blade shape will be a much more sweeping curve... think reverse gentle typhoon.


:)


The drawing is only representative to show approximate dimensions and positions of major structures. The final design will be different. I just wanted to get enough on paper to discuss.

Boulard83
04-04-2010, 10:38 AM
With a good design and blades like the DELTA are using this thing can work. but how much gain for the pain ? ....

gmat
04-04-2010, 10:47 AM
Someone posted the Gentle Typhoon technical papers here, they designed their blades and frame in order to eliminate any blade tip vortex. It doesnt eliminate dead zone though. You might want to take a look at if for some inspiration too (well especially the frame design)...

iandh
04-04-2010, 12:23 PM
With a good design and blades like the DELTA are using this thing can work. but how much gain for the pain ? ....

That is what I am worried about... it may work very well, or it may not be worth the trouble.

The only way to find out at the moment is tests.


Someone posted the Gentle Typhoon technical papers here, they designed their blades and frame in order to eliminate any blade tip vortex. It doesnt eliminate dead zone though. You might want to take a look at if for some inspiration too (well especially the frame design)...

They did one hell of a job too... I definitely will be taking cues from that design, to some degree at least.

naokaji
04-05-2010, 02:26 AM
I say go ahead and make a prototype, test it and take it from there, the idea to "fix" shrouds is certainly interesting.
If it ends up working I'll definitly be in for some.

gmat
04-05-2010, 03:28 AM
At the moment, no one is making fan shrouds, with open corners, and price equivalent to the low grade fans we currently kill in order to use their frame... So there would be a small market, i guess.
Integrated decoupling would be a huge plus (a bit like what Noiseblocker are doing on their multiframe series).

Utnorris
04-05-2010, 07:34 AM
I think it's a great idea Iandh and I am glad someone is looking at this. We all know shrouds do improve the concentration of the air flow on a rad, the key is what you are working on, eliminating the dead space. I doubt you will see much gain temps wise, but rather that you can run the fan at a slightly slower speed cutting the noise but getting the same results as running them at a higher speed. I have seen this design on Blade servers recently, so I know it works, otherwise why would the do it. It reminded me of a jet engine. I will take a pic later at work and post what I am talking about, maybe it will help with the design.

relttem
04-05-2010, 08:05 AM
what CFD package are you planning on using? The only package I know of that is powerful enough to evaluate a fan would be FLUENT. If you use that you will have to do a moving/sliding mesh configuration - which is a beast in fluent. At the same time, that is a great thing to know how to do in Fluent. I'd be more interested in the increase of static pressure on the blade itself. You could look into superchargers and turbochargers as well.

iandh
04-05-2010, 10:19 AM
I think it's a great idea Iandh and I am glad someone is looking at this. We all know shrouds do improve the concentration of the air flow on a rad, the key is what you are working on, eliminating the dead space. I doubt you will see much gain temps wise, but rather that you can run the fan at a slightly slower speed cutting the noise but getting the same results as running them at a higher speed. I have seen this design on Blade servers recently, so I know it works, otherwise why would the do it. It reminded me of a jet engine. I will take a pic later at work and post what I am talking about, maybe it will help with the design.

This is my goal; I share the same suspicion.


what CFD package are you planning on using? The only package I know of that is powerful enough to evaluate a fan would be FLUENT. If you use that you will have to do a moving/sliding mesh configuration - which is a beast in fluent. At the same time, that is a great thing to know how to do in Fluent. I'd be more interested in the increase of static pressure on the blade itself. You could look into superchargers and turbochargers as well.

The analysis would be done by a friend of mine down at APL/JPL. I do believe FLUENT is what he would use, although I am a novice as far as that type of simulation software goes. I am more of an expert in programming and image processing. He comes to me for advice on IDL code, and I go to him for simulations.

I've done enough work in this type of field where I can pretty much do a rough analysis in my head of flow patterns and come out damned close to the sim. It's just something you get a feel for after doing enough engineering. However, the more complex the design, the farther away from accurate my guess becomes.



As mentioned above, my main motivation is increased static pressure... as long as I design the vanes to minimize noise, there is a chance we could reduce the fan RPM needed for a given cooling capacity, and thereby reduce noise without hurting performance.

relttem
04-05-2010, 10:55 AM
a concern would also be vortex shedding off the edge (top) of the blade - as the air is rushing along the face of the blade due to the motion and instead of being forced forward it will actually roll off the edge and create a vortex...but, as you mentioned, the CAD drawing is rough. And, that is something that FLUENT is good for..you can do simple mods and see what happens.

Blueking
04-05-2010, 12:01 PM
So how effective are these shrouds for fans on rads ?

NaeKuh
04-05-2010, 12:56 PM
So how effective are these shrouds for fans on rads ?

my guess is almost no difference.
The reason i say that is because if we saw a better improvement from shaping the flow, rads fins would be designed like air heat sink fins.
However they arent, they are slitted to take air from chaotic directions.
(i think)

So my inital guess is ur our not going to get much improvement if any over a standard shroud.

Iandh u need to make a techbox shroud type thing for us, a universal 120x3 shroud kinda like how the old PA's used to have to fit on any radiator. :\

Church
04-05-2010, 01:07 PM
NaeKuh: rather then some smart modular uni-shroud, so that from parts of two types (ends and sidewalls) can be assembled shroud for any rad .. shipment size would be smaller too.

NaeKuh
04-05-2010, 01:13 PM
NaeKuh: rather then some smart modular uni-shroud, so that from parts of two types (ends and sidewalls) can be assembled shroud for any rad .. shipment size would be smaller too.

yes that would be awesome...

I also was wishing for a shroud which featured a removable filter that u could clean and replace with ease.

Church
04-05-2010, 01:27 PM
NaeKuh: for filter - shallow cutout along sidewall(? line? don't know the right english word), where filter can slide in from one end of shroud. Though imho even more interesting looked those magnetic filters, that were THE simpliest to mount or dismount (IIRC by Deciflex?). For such no cutout in shroud needed, just metallic top side so that they can stick like stick to metallic case ..

iandh
04-05-2010, 01:29 PM
a concern would also be vortex shedding off the edge (top) of the blade - as the air is rushing along the face of the blade due to the motion and instead of being forced forward it will actually roll off the edge and create a vortex...but, as you mentioned, the CAD drawing is rough. And, that is something that FLUENT is good for..you can do simple mods and see what happens.

My first plan is to make about five different vane shapes, and test them with two different fans (Yate "standard blade", and GT "swooped blade").

After that I'll get an idea of what direction I need to tweak in, and I'll hand that new drawing over to my contact to play with in software.

Once I have a vane profile, I'll have my CNC shop make up a punch press die from the results to form the blade profile.

What I'm hoping is that if we get the spacing and profile right, we can actually catch/prevent that vortex and transfer it mostly into linear motion. The problem is, I will have to optimize for a given flow rate. I'm going to target around 35 CFM, as that is the range that most silent cooling folks run their fans at.


my guess is almost no difference.
The reason i say that is because if we saw a better improvement from shaping the flow, rads fins would be designed like air heat sink fins.
However they arent, they are slitted to take air from chaotic directions.
(i think)

So my inital guess is ur our not going to get much improvement if any over a standard shroud.

Iandh u need to make a techbox shroud type thing for us, a universal 120x3 shroud kinda like how the old PA's used to have to fit on any radiator. :\

The reason for radiator construction isn't because it is the best aerodynamic design, that would be pretty unmanufacturable. It's the best balance between optimal function and manufacturability.

The reason for the slitted fins in radiators is to create turbulence at the contact point, but helping the air get into the fins in the first place could create higher air velocity > more turbulence > better performance

The problem with not correcting the flow out of the fans before they hit the rad is that the fan ends up fighting against its own turbulence to force the air into the fins... so we end up relying mainly on static pressure for radiator performance. With a proper flow correcting shroud, static pressure would still be in the equation, but flow would be introduced more as well.

This could only improve performance vs. a standard box shroud... I would have to try pretty damn hard to design it so poorly that it wouldn't make any improvement. Even manufactured exactly to the rough drawing would be likely a measurable improvement.


FYI I will be making a standard box shroud as well, because this flow directing shroud would only work on the push side... putting one on the pull side would be a waste of money vs. a standard box shroud.

Church
04-05-2010, 01:55 PM
iandh: I'm just vary of one thing. Especially for silent/slow fans big part of noise is made not by engine or bearing, but by air movement itself. Can it happen that flow will be faster even at slower rpm, but it hitting such shaped shroud may add additional noise to cancel out gains? Of course it all is purely speculation/guess of mine ..

Sadasius
04-05-2010, 02:08 PM
Interesting. Would love to see what would happen. Build one and test it.

eth0s
04-05-2010, 02:55 PM
This is a good idea iandh. (And I like the new logo, btw.) Now, if I understand you correctly, what you are trying to do here is to improve the efficiency of any fan, by reducing the vortices produced by the spinning fan blades. If this design is successful, it will allow any fan to operate more efficiently, meaning that for the same amount of energy input, more air will be moved over the radiator, because less will be lost to "waste" created by vortices. Also noise will be reduced, since noise is really just the byproduct of the wasted energy embodied in the vortices. As such, this design will increase the efficiency of any fan. But perhaps the better the fan, the less improvement will be gained from this design. However, eliminating the dead-spot is a valuable improvement for any fan, although that is really just a function of the height of the shroud, as proved by Cathar so many years ago. However, I guess you would argue that you are trying to achieve the same results with a shorter shroud because the blades are redirecting the airflow to eliminate the dead-spot. As such, I would think that you would want to eliminate the center "plug" altogether in your design, but maybe I am missing something. All in all, I would like to see how your prototypes work, and what results you get. I would tend to think that they will work as intended, and that they will have the greatest benefit on the loudest and least efficient fans. And of course, if it works as intended, it should improve the performance of any fan. Which means that you will get more performance with less fan, or that is to say that you will no longer need to go deaf to get high airflow over your radiator.

JasonDTM
04-05-2010, 03:18 PM
Great idea, but the problem with bending airflow to ones will is it will create another problem that will need to be solved and so on. :(

iandh
04-05-2010, 03:36 PM
iandh: I'm just vary of one thing. Especially for silent/slow fans big part of noise is made not by engine or bearing, but by air movement itself. Can it happen that flow will be faster even at slower rpm, but it hitting such shaped shroud may add additional noise to cancel out gains? Of course it all is purely speculation/guess of mine ..

That is where vane shape gets very important, especially the shape of the leading edge. A gently rounded edge should be the best for noise, but worst for scooping air.


This is a good idea iandh. (And I like the new logo, btw.) Now, if I understand you correctly, what you are trying to do here is to improve the efficiency of any fan, by reducing the vortices produced by the spinning fan blades. If this design is successful, it will allow any fan to operate more efficiently, meaning that for the same amount of energy input, more air will be moved over the radiator, because less will be lost to "waste" created by vortices. Also noise will be reduced, since noise is really just the byproduct of the wasted energy embodied in the vortices. As such, this design will increase the efficiency of any fan. But perhaps the better the fan, the less improvement will be gained from this design. However, eliminating the dead-spot is a valuable improvement for any fan, although that is really just a function of the height of the shroud, as proved by Cathar so many years ago. However, I guess you would argue that you are trying to achieve the same results with a shorter shroud because the blades are redirecting the airflow to eliminate the dead-spot. As such, I would think that you would want to eliminate the center "plug" altogether in your design, but maybe I am missing something. All in all, I would like to see how your prototypes work, and what results you get. I would tend to think that they will work as intended, and that they will have the greatest benefit on the loudest and least efficient fans. And of course, if it works as intended, it should improve the performance of any fan. Which means that you will get more performance with less fan, or that is to say that you will no longer need to go deaf to get high airflow over your radiator.

The center piece sits right against the fan hub, so has no effect whatsoever on airflow.

I do have another idea (from a suggestion via PM) to add an upside down cone at the hub. It would prevent a vortice being formed as the air passes by the corner of the hub.

Serpentarius
04-18-2010, 09:08 PM
so you're adopting the contra-rotation theory (http://en.wikipedia.org/wiki/Contra-rotating_propellers)? it'll work ... but the trick is, you'll have to get the static (non-moving) blades closer to the rotating blade ... the closer you get, the more static pressure you'll produce (lil' side effect the noise)

it's something like scooping upon scooping the air flow

before you work it out with acrylic, try it with some aluminum drinks can + superglue

mlwood37
05-08-2010, 06:56 AM
I&H Please contact me ref the kill coils i have bought from you that you have not sent. I have sent several private messages to you and you have not responded.

iandh
05-08-2010, 10:26 AM
I&H Please contact me ref the kill coils i have bought from you that you have not sent. I have sent several private messages to you and you have not responded.

As I told you at least twice in our previous PM's, email is the best contact method. You have not sent me a single email to date, even though that is the communication method I requested. I'm not upset, but it was explained to you that PM is not a good way to contact me if you are in a hurry for a response.

As mentioned previously, I do not always respond to PM's quickly becuase I only visit XS every few days, and sometimes am away for a week or more.

All of my email accounts are connected to my cellphone, and I can be reached instantly that way. I respond within 24 hours or less.