Computer fan performance

[Thunderchild]

Ok to be considered that what I sm saying is in the light of the fact that i work for an engineering company that designs and makes radiator cooling packs so fans are pretty much their thing. Now we had an issue of customer complaint where a 60 ish cm diameter fan was due to have a tip clearance to the fan housing of 2 mm. the result of our suppliers incompetence in xchaiving this meant that they lost the contract and we had some parts made by another comapny and built the rest ourselves.

Now I go and buy a cpu cooler with a nice 12 cm fan in it and guess what ? the tip to housing clearance is 3 mm surely this will impact hugely on this fans performance ?

Does anyone else have experience in fans ? I plan on taking it into work and asking one of the engineers tomorrow.

 

[DirtyLude]

Performance means what to you? CFM vs power consumption? Pressure vs Power Consumption? CFM/Pressure vs Noise level? MTBF?

There's lots of design criteria that go into a fan.

Then again, it just might be a cheap/poorly designed fan. As long as it works and the noise level is acceptable I don't think it's that big of a problem.

I'm definitely no fan expert. The guys at silentpcreview.com like to debate about computer fan design, though.

 

[Thunderchild]

well if you have a large tip to housing clearance then air will "leak" from the out side to the in side so less air (CFM) will be pushed through the fan as it would be like reducing the blade diameter, of course put under stress the air will take the easiest path which could be back to where it came from, if I can get the fan to perform better by closing the gap then a lower speed will move the same amount of air and so make the fan quieter.

At work we do buy some very exspensive fans and fan housings: both parts are cast and machine and the fan blades are about 40 cm in diameter but the clearance is under 1 mm these do cost over £ 1600 though and are used only on high performance applications, if you compare the [percentage of clearance to fan diameter between what i use at work and what i have in my computer the difference is quite a lot.

 

[dknguyen]

That's the thing about aerodynamics...it doesn't seem to be enough to use the "ratio of something" and just compare them if they are of different sizes because aerodynamics does not scale linearly. And depending on what the fan was used for, it may have mattered. I mean...how many people actually check the the efficiency and efficacy of their computer fan? Sure, people might say it, but how many people actually check it?

Not sure if this applies here...but I was reading on a plane forum that pulsejets does not work at small scales because the size of the engine is so small that the static boundary layer fills up a significant volume of the engine compared to a larger pulsejet. Maybe something similar is happening here with the computer fan (if anything is really happening at all). My guess is that it's just cheaper and no gives a crap enough to actually check in computer fans- everyone just checks temperature, and temperature is easy...just give the fan more power and the temperature will be lower.

 

[Thunderchild]

I think this is a case of cost and loose tolerances, i beleive that the fan was cheaply produced and therefore plenty of room for error was left

 

[Nigel Goodwin]

I think this is a case of cost and loose tolerances, i beleive that the fan was cheaply produced and therefore plenty of room for error was left

Pretty obvious really - it's a cheap fan, so it's made cheaply.

This doesn't mean it's not perfectly suitable for it's job, just larger than it could be otherwise - but would you sooner pay five times the price, for a fan just slightly smaller?.

 

[Thunderchild]

nope i would like the best possible performing fan of this size (12 cm) because the heatsink it goes on specificly requires a 12 cm fan. the better it performs the slower it will need to run and so the quieter it will be. I'm suprised at the apparent cheapness of it as it came from a good quality heatsink. at 1500 rpm (top speed) it pushes through no more air than an 8 cm fan at 3000 rpm. as it has twice the area it will push the same amount of air at half the speed.

 

[dknguyen]

When Nigel said "slightly smaller" I believe he was referring to the tolerance affecting the entire size of the fan, ever so slightly. Same fan- slightly smaller because of tighter tolerances. Not a fan that was actually specced at a different diameter.

 

[Thunderchild]

well actually the fan needs to be 4-6 mm larger in diameter thats an increase in surface area of 10 % which would mean (bearing in mind I'm not an expert) +10% more airflow just for the mere size and due to the losses because the air is feeding back you may be talking about another 10 % increae in airflow versus speed (and to some extent noise)

If we assume that due to the losses in air feeding from outlet to inlet and not going anywhere rendering say the last 3 mm of the tips ineffective that could mean a 10 % loss at the moment.

obiously I can't make the fan bigger but I can put some sort of spacer around the housing and close the gap so that i may eliminate the losses and recover 10 % in performance.

 

[Thunderchild]

When Nigel said "slightly smaller" I believe he was referring to the tolerance affecting the entire size of the fan, ever so slightly. Same fan- slightly smaller because of tighter tolerances. Not a fan that was actually specced at a different diameter.

i think Nigel meant that a better designed fan which is more efficient can do the same job but be smaller, naturally this would mean it would be more costly, however considering it cost £ 26 (the heatsink with fan)

I am familiar with its manufacturing process for the same money my company buys a whole airconditioning heater/condensing matrix built in a similar way to my heatsink but much larger, A slightly better fan would have not gone amiss and broke the budget.

the fans specs are in sone, what is sone ??? all other fans I have seen rated are in dB and this is not specifying speed either, the ultra quiet seems to be dissapearing on the horizon, it could have been ball bearings rather than the method used for another couple of pounds it could be a better fan