What a waste of time!

[Enigmo]

I've just swapped my Sunon 60mm 33db fan for a Papst 80mm 21db using a 60-80mm adapter and the noise levels are exactly the same. As I'm now using a 4 pin power cable rather than 3 my BIOS thinks my fan is running at 0 RPM because the motherboard detector has been bypassed so I get a system critical exclamation mark whenever I power up even though the fan is spinning as it should. This is inevitable but still very annoying. The CPU is still running at 59 degrees as before.

What the hell was the point?

 

[RalfHutter]

I don't know why this isn't discussed more, but you've found the dirty little secret of fan adaptors: they don't really work. I think it's because of the extra back-pressure due to the air hitting the beveled sides of the adaptor. This creates more noise and more turbulence. The noise is self-evident, the turbulence keeps the larger fan from working as well as it would if it was mounted normally.

Do yourself a favor, get an Alpha PAL 8045 and put a quiet 80mm fan on it, that's the best way to get good, quiet cooling.

 

[Witchfinder]

I never understood the Papst Hype, if you check the specs the CFM output is very low on the quiet ones, you may as well run a Panaflo or YS-Tech at 7V or through a voltage dropper and save money with the option of re-running at full 12V in the middle of a heatwave..

Papst fans USED to be the absolute best (I`m talking the old Mainframe days here!) but now I feel that they`re riding on their name, they certainly aren`t made as well as the old aluminium cased ones but still cost..

 

[Enigmo]

That's exactly what I've been wondering myself Ralf. People come along to this forum and ask "how do I make my PC quieter?" and the same stock answers appear instantaneously, one of them being to get a larger fan and a fan adapter. It's automatically assumed that these things actually work.

Based on this disinformation I purchased an Alpha 6035, a fan adapter and a Papst 80mm fan. I've spent a small fortune on this little setup and I'm no better off now than I was when I was using a bargain basement, bog standard heatsink and fan. The noise levels are about the same and the cooling efficiency is the same (which wasn't really adequate to begin with - an average of 59 degrees).

Thanks for the suggestion but I really don't want to splash out on any more cooling stuff when I've only just bought an Alpha 6035.

Personally I think the CPU manufacturers should be doing more to address this problem. I mean isn't it ridiculous that these processors are designed in such a way that they get so hot in the first place? Surely with the technology available today a solution to this problem could be found? Or is it that the manufacturers want these chips to self destruct so you have to keep replacing them before they are so outdated you would normally be upgrading anyway?

Witchfinder - I didn't buy a Papst fan based on personal experience. My decision was based on the general consenus of opinions posted here. In any case if the Papsts are no better than Panaflos I'm not all that concerned since I didn't pay over the odds for it - it cost me about £11 - about the same price as Panaflos I think.

 

[punkrawket]

<< What the hell was the point? >>

you never know until you try

 

[Mikewarrior2]

You're running 59C "cpu" temp, right?

Well, let's have some details, such as:

Motherboard
CPU type(XP or T-bird)
Vcore
Heatsink
Ambient system temp.

Depending on various factors, you may not be able to improve your "cpu" temp until you work on case cooling, etc.


Mike

 

[Crucial]

I hardly think cpu mfg's want the chips to self destruct. This is a fact of life when you are running these chips as fast as we do.

I have a Thermalright AX-7 with the adjustable Enermax fan. I can dial in as much noise/cooling power as I want. I went from the other "king" of cooling, the sk6, to the AX-7 and my cpu temps went down about 7deg C at idle and full load. All in a cooler that didn't need to bolt to the motherboard. total cost. $36

As for fan adapters. Phooey! Every review I have read says they don't work as advertised. The concept is flawed. The more backpressure you put on a fan the noisier and worse it performs. Just mount the thing using a fan grill or some other means.

 

[Enigmo]

Thanks for the replies everyone

punkrawket - I agree with that approach normally, but not when it involves throwing hard earnt money down the drain or risking the health of an expensive PC.

Here are my 'puter spec...

AMD Thunderbird 800
Gigabyte 7ZM motherboard
Alpha 6035 heatsink (supplied with a Sunon 60mm 23.5 cfm fan but replaced with an 80mm Papst 27cfm fan)
What's a Vcore?
I'm running motherboard monitor which reports CPU temp at mostly between 58 and 62 degress celcius. The other two censors report temps of 23 and 32 degrees celcius - one of these will be system temperature but I don't know what the other one is.

I know it's a conspiracy theory but I wouldn't be surprised at all if these CPUs had a predetermined life cycle. I mean this is true of cars isn't it. They'd last forever if the manufacturers wanted them to, but this is clearly bad for business.

Anyway, your alternative mounting idea sounds promising. Have you got any links to articles which explain more about this? For instance would I have to get myself an 80mm grill or would my now superfluous 60mm grill be OK? I thought I was asking for trouble as soon as I saw the adapter and realised how far away from the heatsink the fan is positioned. That can't help cooling efficiency at all.

 

[Crucial]

here is an example of what I was talking about. 2cooltek

 

[orthancstone]

nah, CPUs can last a long time. my parents still have an old Pentium 75 sitting around (rarely used, but it works!).

considering this site is very descriptive on all hardware stuff, i'll assume you know enough about why they are hot.

the real problem is AMD has to work hard and quick to spit out new, faster chips. look at intel...their chips run pretty nicely in the temperature department. why? well, intel typically has been working on those chips for a long time and have been working to make it the kind of chip that can last a long long time (which means temp needs to be lower). AMD usually has to play catchup.
there is also the fact that AMD processors (especially the XP) typically do more per clock cycle, driving up the workload and the temp. it is the sacrifice you take...faster proc, more heat. if it sucks so much for you, buy a intel chip. if you don't mind the temp and you want great performance for the lower price, get an AMD and beef up the cooling (hey, a great cooled system plus an AMD is still probably cheaper than a P4...unless you spend big bucks for cooling).

 

[Crucial]

<< nah, CPUs can last a long time. my parents still have an old Pentium 75 sitting around (rarely used, but it works!).

considering this site is very descriptive on all hardware stuff, i'll assume you know enough about why they are hot.

the real problem is AMD has to work hard and quick to spit out new, faster chips. look at intel...their chips run pretty nicely in the temperature department. why? well, intel typically has been working on those chips for a long time and have been working to make it the kind of chip that can last a long long time (which means temp needs to be lower). AMD usually has to play catchup.
there is also the fact that AMD processors (especially the XP) typically do more per clock cycle, driving up the workload and the temp. it is the sacrifice you take...faster proc, more heat. if it sucks so much for you, buy a intel chip. if you don't mind the temp and you want great performance for the lower price, get an AMD and beef up the cooling (hey, a great cooled system plus an AMD is still probably cheaper than a P4...unless you spend big bucks for cooling). >>



Say What? What has AMD playing "catch up" have to do with heat? And besides AMD hasn't bee playing "catch up" for quite some time now. My athlons have been outperforming peniums for a couple years now. If anything Intel has been trying to "catch up".

Instead of trying to bash AMD why not try and find out what is wrong with his system and give him some pointers.

 

[WarCon]

Well I use to run an Alpha PAL 6035 with the delta screamer (hated it), but was wondering one thing. Everyone here is talking about the fan waffling in back pressure by hitting the sides of the adapter. Well one thing is the PAL is designed to draw air off the heatsink (thats the purpose of the metal shroud on the sink so it draws air from low on the heatsink). If your drawing air, you shouldn't get addition turbulence and if you seal it all up real good (I used electrical tape), then the added cfm (assuming several things like pitch on the blade) should be seen. If your using a shallow pitched 80mm (21db is most likely shallow pitched), then you lose alot of ability to push or pull against a load (which is exactly what a heatsink is). A bigger heatsink (more surface area) is going to be required for quiet cooling. Alot of fans CFM rating is based on no load condition. Unless a fan is horribly designed, a general rule is noise = ability to push against a load (look at vacuums, if it was possible to push/pull against a load and be quiet you would see quieter vacuums)

As far as fan adapters go they shouldn't be used to blow onto the heatsink, but the PAL isn't designed for being blown on anyway. I think if your wanting to quiet a 45db delta by drawing using a 35db 80mm with the same or more cfm, then adapters are a way to go. Otherwise put a nice quiet fan on your heatsink and get 2 big quiet 120's and mount them on the door right above the heatsink and get your extra cooling that way. Quiet 120's make a lower pitched sound that is usually much more tolerable/noticable.

...................edited..................
Added more info.

 

[Crucial]

good point. i wonder if enigmo has the fan blowing or sucking?

 

[WarCon]

If you really want to blow some air though, assuming you don't mind removing your mobo, then you could always rig a mounting (using the holes in the mobo) that will take a 92mm fan (maybe a 120 if you don't mind bending the bolts alittle) right over the heatsink. Tilting the fan slightly helps. Hold a fan put your hand in front of it and feel where the point of most pressure is. It is not directly in front of it. It usually blows out at an angle.

If you can find them, I recommend using all plastic nuts and bolts and be careful about components on your mobo around those holes.

Just a thought

 

[Enigmo]

Thanks for the link csaddict. The proceedure seems straight forward enough. I just need to get myself an 80mm grill now. If it doesn't work at least I've only lost a pound or two. That's actually the site where I bought my Alpha 6035 from but I missed that article completely.

Thanks for all the info everyone. I'm trying my best to absorb it all. I assume the fan is blowing rather than sucking but I couldn't say for certain. I attached the fan to the adapter with the central label facing outwards just like the one I removed.

If this helps the fan is a Papst 8412 NGML 80mm fan (4 pin) (FG-000-PA) and you can find the specs here: http://www.overclockers.co.uk/acatalog/Online_Catalogue_Papst_Fans_61.html

I got these slightly wrong. I should have said the noise level was 19db (allegedly, ho ho) and the CFM 26.5. Not a huge difference but I thought I'd clear that up anyway.

 

[WarCon]

Just put you hand over it and see if air is going in or out.......

If its blowing, then turn it around and see if its quieter and/or works better.

 

[bozo1]

<< I assume the fan is blowing rather than sucking but I couldn't say for certain. I attached the fan to the adapter with the central label facing outwards just like the one I removed. >>


If the label is facing outwards, then your fan is sucking.

I think alot of people miss the point with the adapters or swapping a 60mm for an 80mm. Many do it simply to get away from the high pitched hair dryer whine of the 60mm Delta screamers. There is cooling loss which can be compensated for by using a higher CFM fan that the Delta. Swapping a 38CFM Delta screamer for a 42CFM or higher 80mm fan will give you about the same amount of cooling without that high pitched whine. Yes you can hear it but it's a much nicer noise.

 

[SWScorch]

Okay I just have to ask one thing: Are you sure its the Papst fan you're hearing? At 21 decibals that would be absolutely silent to my ears. Are you sure you arent hearing some other fan; say your power supply, or even your hard disk spinning. Hard disks are pretty loud; I would think they would overpower a 21 decibal fan.

If this is not the case, and you are certain that it is the fan you are hearing, then just feel free to ignore my post

 

[orthancstone]

<< Say What? What has AMD playing "catch up" have to do with heat? And besides AMD hasn't bee playing "catch up" for quite some time now. My athlons have been outperforming peniums for a couple years now. If anything Intel has been trying to "catch up".

Instead of trying to bash AMD why not try and find out what is wrong with his system and give him some pointers. >>



you misread my point. before you accuse me of trashing AMD, check out my rig

intel spends more time then AMD in developing chips. so, whatever chip intel designs for 10 years from now, AMD will not be designing for that same time period as early as intel. this occurs because intel has TONS of damn money for development.
anyway, my point in mentioning that is intel has more time to be concerned with heat management. (i also gave another reason earlier but i'm not gonna reiterate myself).

 

[Enigmo]

<< Okay I just have to ask one thing: Are you sure its the Papst fan you're hearing? At 21 decibals that would be absolutely silent to my ears. Are you sure you arent hearing some other fan; say your power supply, or even your hard disk spinning. Hard disks are pretty loud; I would think they would overpower a 21 decibal fan.

<<

Thanks for the suggestion, but I'm certain it's the heatsink fan making the racket. Last week I replaced the fan in my PSU with a 12db Papst fan and before I realised how hot my CPU was getting I tried running one of the same 12db fans on my heatsink. Noice was eliminated almost totally but my CPU temperature went through the roof so I had to switch back to my Sunon 60mm fan ASAP. I've got 2 HDs but they were inaudible and I've got no other case fans or GFX card fans.

WarCon - I'll whip the case off tomorrow and see if switching the fan the other way round makes any difference. I could be wrong but I would have thought noise levels would be lower if the fan was sucking because there would be no buffeting of air hitting the adapter/heatsink?

bozo1 - I've grasped the concept of CFM/size/noise relations through reading posts on this forum but it doesn't seem to apply to my situation for some reason. I've gone from a 33db fan to a 19db and not noticed any reduction in noise.

 

[vicwang]

"faster proc, more heat. if it sucks so much for you, buy a intel chip... there is also the fact that AMD processors (especially the XP) typically do more per clock cycle, driving up the workload and the temp"

True, AMD has faster procs than Intel, but that doesn't mean they produce more heat. When comparing Athlon XP's to Pentium 4's (.18 micron to .18 micron), AMD chips produce significantly LESS heat than P4's at comparable performance levels. And doing more work per clock cycle doesn't make Athlons hotter, if anything it makes them COOLER because there aren't so many wasted clock cycles where heat is being generated but very little work is being done (a la the P4). Plus the fact that non-Northwood P4's are about TWICE the size of Athlon XP's has a lot to do with the P4's higher heat output...

Here's a good link. Notice the max output of the two CPU's:

Athlon XP vs P4 '

And here's another good site with thermal specs. Note that the non-Northwood P4's are MUCH hotter than Athlon XP's.

Sandpile.org

 

[vicwang]

"could be wrong but I would have thought noise levels would be lower if the fan was sucking because there would be no buffeting of air hitting the adapter/heatsink?"

I would think so too, but I would also think that the important thing here is the "noise-to-perfomance" ratio, not "noise" in and of itself. What I mean is, you might get less noise by having a particular fan suck instead of blow, but you could get just as much noise and better performance by using a lower-RPM fan and blowing... In other words, at a given performance level, having the fan suck could very well be MORE noisy due to the higher RPM it would require to achieve the same level of performance.

I'm also wondering just how loud this 19 dba Papst could possibly be... Since the decibel scale is logarithmic, isn't 19 dba less than ONE TENTH the noise level of a 33db fan? Have you tried running the fans by themselves outside of your case to compare the noise levels? I 've read several articles stating that sounds in the 20 dba range are virtually silent. I wonder if the fan you received is defective, or if it's causing unwanted vibrations and resonance. You might try using rubber washers/grommets on all your fan screws, if you haven't already, to make sure there's no resonance problems (I got some really good ones from Lowe's for something like 20 cents each)

 

[Enigmo]

Well I've just flipped the fan around the other way and the temperature has come down to 48 - 53 degrees celcius so that's quite a good drop. I'm still not sure whether it's sucking or blowing because the air flow felt the same when I put my hand next to it both ways round. It's still very loud which I'm not too happy about since this was the whole purpose of the exercise but at least my CPU won't melt now.

Thanks for the grommet tip. I'll look out for those next time I go into town. I don't know about it being defective. It seems to spin as it should, there's no stuttering and the noise is constant rather than variable so that's less annoying than it could be I suppose.

 

[Enigmo]

Update: the temp has now increased to 59 degrees so it looks like I'm back to square one. I just hope it levels out there.

 

[Crucial]

if you look at the side of the fan casing ther should be little arrows showing airflow and fan direction. Sorry to hear about your temps.