Proof of concept
Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Lawrence Lee
Proof of concept
Whenever the subject of going totally fanless (including the PSU) is brought up the idea is more often than not met with a lot of sceptisism. "If you can't hear the fans, why not use them?" and "It is WAY to much trouble to build a fanless system for it to be worth it" is the two most used arguments against going fanless.
However, I've wanted to test that for a long time now and most recently frankgreen built an Atomsystem without using any fans at all. I've built one as well which I wanted to show. The system looks as it does to prove the concept, not to be safe/look good etc. I simply want to show that it works. When I've got my PicoPSU, my new case and my Vortex it'll be ready to show off:)
Please also note that I have taken the side off for the purpose of taking the photos, it really does run in a closed system.
The hardware used is
Zalman 400B (Placed outside of the case for the sake of concept, no fans)
Ninja 2
Windsor X2 4200+ 89W TDP
Seagate 120GB HDD
MSI Live Diva
Optical drive
Yes, it is a fan, but it is controlled with speedfan and set to start at 62 degrees. As soon as it does, the core temp drops 10 degrees in 5-15 seconds. I use it as a last resort. I can run the system totally fanless set to [email protected], but that last half multiplier and last voltage (to [email protected]) makes the fan spin sometimes when the system has been under full load for a longer period.
It is impossible to tell by the photo, but the fan is not moving.
And finnally a screen shot of speedfan or three.
Edit: It might be worth noting that the PC is under the desk, no more than 1m from the radiator in the other end of the desk. Ambient in the rest of the apartment is 24-25C, I do not know how warm it is under the desk.
However, I've wanted to test that for a long time now and most recently frankgreen built an Atomsystem without using any fans at all. I've built one as well which I wanted to show. The system looks as it does to prove the concept, not to be safe/look good etc. I simply want to show that it works. When I've got my PicoPSU, my new case and my Vortex it'll be ready to show off:)
Please also note that I have taken the side off for the purpose of taking the photos, it really does run in a closed system.
The hardware used is
Zalman 400B (Placed outside of the case for the sake of concept, no fans)
Ninja 2
Windsor X2 4200+ 89W TDP
Seagate 120GB HDD
MSI Live Diva
Optical drive
Yes, it is a fan, but it is controlled with speedfan and set to start at 62 degrees. As soon as it does, the core temp drops 10 degrees in 5-15 seconds. I use it as a last resort. I can run the system totally fanless set to [email protected], but that last half multiplier and last voltage (to [email protected]) makes the fan spin sometimes when the system has been under full load for a longer period.
It is impossible to tell by the photo, but the fan is not moving.
And finnally a screen shot of speedfan or three.
Edit: It might be worth noting that the PC is under the desk, no more than 1m from the radiator in the other end of the desk. Ambient in the rest of the apartment is 24-25C, I do not know how warm it is under the desk.
Last edited by Tobias on Sat Feb 14, 2009 4:25 am, edited 1 time in total.
In idle the core temp is 38-40 and with 2*Prime95 I flatten out at 61-62 degrees with [email protected] (the fan doesn't start with this setting) and if I go to [email protected] the fan kicks in after a while, run for 15-30 secs and is then idle for 10-15 minutes. I am currently running it at [email protected] because I do not ever do anything that stresses the CPU so hard that the fan starts.xan_user wrote:"cool" :8
Got some temps for the cpu under a little stress for a comparison?
Hows the HDD temps during heavy use ?
Are you running this with the case open?
I run this system in a closed case.
-
- Posts: 68
- Joined: Sun Aug 10, 2008 7:41 pm
- Location: Canada
i'm not too sure what to think of this.
all i see is a well thought out system using quiet parts, and a fan that is set to turn on at a certain temp- as opposed to increasing in speed (kinda like some notebooks).
the ultimate test for a passive system is to keep temps under control in a hot environment, whilst stressing itself.
adding a fan is the simplest solution, but you can hardly call the system passive at that point.
i mean- on a cold enough day, anyone with a lowish powered system and big heatsinks could probably stop their fans without any worries.
doesn't the fact that you use and arguably need the fan, support the common argument that "It is WAY to much trouble to build a fanless system for it to be worth it"?
"If you can't hear the fans, why not use them?"
since it's there already, spinning the fan at 3-400rpm should not add to system noise, and will give you better temps all round.
.....just some thoughts.
all i see is a well thought out system using quiet parts, and a fan that is set to turn on at a certain temp- as opposed to increasing in speed (kinda like some notebooks).
the ultimate test for a passive system is to keep temps under control in a hot environment, whilst stressing itself.
adding a fan is the simplest solution, but you can hardly call the system passive at that point.
i mean- on a cold enough day, anyone with a lowish powered system and big heatsinks could probably stop their fans without any worries.
doesn't the fact that you use and arguably need the fan, support the common argument that "It is WAY to much trouble to build a fanless system for it to be worth it"?
"If you can't hear the fans, why not use them?"
since it's there already, spinning the fan at 3-400rpm should not add to system noise, and will give you better temps all round.
.....just some thoughts.
-
- Posts: 68
- Joined: Sun Aug 10, 2008 7:41 pm
- Location: Canada
I'm with you on this one. My thoughts exactly.porkchop wrote:i'm not too sure what to think of this.
all i see is a well thought out system using quiet parts, and a fan that is set to turn on at a certain temp- as opposed to increasing in speed (kinda like some notebooks).
the ultimate test for a passive system is to keep temps under control in a hot environment, whilst stressing itself.
adding a fan is the simplest solution, but you can hardly call the system passive at that point.
i mean- on a cold enough day, anyone with a lowish powered system and big heatsinks could probably stop their fans without any worries.
doesn't the fact that you use and arguably need the fan, support the common argument that "It is WAY to much trouble to build a fanless system for it to be worth it"?
"If you can't hear the fans, why not use them?"
since it's there already, spinning the fan at 3-400rpm should not add to system noise, and will give you better temps all round.
.....just some thoughts.
That was generally the idea of this post To point out that it is possible, but also to point out that "cold enough" don't need to be so cold as everyone thinks. And that it is possible by using nothing but off-the-shelf products. The idea is not to show this off as a passive system, I don't have all the hardware I want to build the system I wanti mean- on a cold enough day, anyone with a lowish powered system and big heatsinks could probably stop their fans without any worries
I agree, at current settings I might actually need the fan sometimes. That is due to the fact that I'm to greedy;) I could well change the the multiplier from 11 (stock) to 10.5 and not need the fan.
I've meassured ambient where my computer is now, it is 28C. The CPU I use isn't exactly the coolest tool in the box, either. It is 3 years old rated at twice the TDP of say the 4050e-4450e (although at the same speed) and the case I use is an old Thermaltake case which don't exactly have silence in mind.
I'm obviously not Australian, so you will have to make a judgement yourself, if the conditions posted has any meaning for you. So far I've run 2*Prime95 for more than 25 minutes without the need of the fan @2,2GHz/1.1V in 28C ambient. I guess 28C ambient is pretty commonplace for you, it is not commonplace in Swedenthe ultimate test for a passive system is to keep temps under control in a hot environment, whilst stressing itself
EDIT: In the end it took 30 minutes for the fan to engage.
As for the fan actually being there. There are life buoys on all beaches here in Sweden as well, that doesn't mean you have to bath using floatation device (not yet at least...) That doesn't mean it is a bad idea to have life saving equipment handy As you say, there is a simple solution to a potential problem, why not take the precaution?
ntavlas: the Pico is ordered (I thought someone would pick up the meassured loads comming out of my PSU from the screenshot above ), and I'm eagerly awaiting the OCZ Vortex I got a NAS for storage allready:)
The case, well, it is an oooold Thermaltake case which I've used for atleast 5 different upgrades. The first iteration of this computer was a Duron 900 build and since then I've changed modo and CPU at least 5 times I have a new case on my shopping list as well