Here is what I think.
Most heatsinks are designed for blowing, if it was designed for sucking it would work as effectively.
Blowing air downwards onto a HS causes air to be recirculated and this reduces efficiency. If you have a back exhaust fan which is most likely the case, some of that recirculated air will be removed by the exhaust fan and PSU but not all. Some air will go elsewhere such as the same air being blown back down through the HS again. In suck mode the same applies but maybe less so due the ATX form factor where the air is now blown upwards right next to the exhaust fan which takes a bigger percentage of that air to be exhausted out (less recirculated air). This is why I think the Alpha HS works better in suck mode... Not only are they designed for suck mode but less recirculated air means better cooling.
Another words its not the blowing or sucking that makes a difference but how the air moves inside your case.
Ideally a HS should be ducted so cool ambient air goes through the HS and all escaping air is exhausted outside of the case. Its much easier to duct a HS in suck mode than in blow mode because the way most HS are designed. Imagine trying to capture all the exhausted air through a SLK-9xx in blow mode.
PSU makes no difference in suck or blow mode (to exhaust out of the case) because ~100% of air goes in one side and it comes out of the other. No matter which way you do it no air is being recirculated. Its all about airflow efficiency...
A good post I read was from the very knowledgeable dorothy who estimated about 15% of HS airflow is recirculated back to the HS and thus being less effiecient. It was a while back so the figures maybe slightly out.
Thats why Artic VGA cooler works well because no hot air is recirculated. No nit pick here, I know a tiny fraction of the heat is recirculated because its not completely sealed.
I can't understand why people stick to blowing CPU fans!!!
Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Lawrence Lee
-
MikeC
- Site Admin
- Posts: 12285
- Joined: Sun Aug 11, 2002 3:26 pm
- Location: Vancouver, BC, Canada
- Contact:
I agree with not putting it in the case, but then what I have to point out is that the manufacturers of these fanless PSUs are not intending they be used this way. They ARE sold as drop-in replacements for ATX PSUs in ATX cases. So they really need to work well as drop-in ATX PSU replacements. Putting one outside the case creates a host of issues that many people will not or cannot accept.Pirata wrote:@MikeC: I'd never use a fanles unit inside a case. They are silent. They have to be out of the case. BTW the Silver Stone fanless unit does have a lot of radiant surface, just like the Silentium. The SilentMaxx, the Yesico and the NC-Silence make up for that surface with big exterior heatsinks.
My point is this: If they are meant to be used outside the case, then they should be designed that way. Certainly the ATX PSU form factor is NOT ideal for a passive cooled PSU placed outside the case. The obvious options are to make it long and low (flat) for it to sit on top of a tower case, or to make a cooling chimney through it and have it hang off the back of the case.
1)
Mike, while it is very constructive to think about what manufacturers should do, it is realistic to think about what CAN BE DONE with what we have. The point is to find out the best way to use what there is.
Do you know the fanless Silver Stone? Have a look at it at www.dirkvader.de. Like the Silentium, its enclosure is a huge heatsink. It is ideal for putting it in the outside. I like it because it has active PFC, which prevents it from buzzing (Dirk says passive PFCs use a coil that makes the noise). This makes the Silver Stone completely silent (the Yesico and the SilentMaxx do buzz).
2)
You make some good points about the exoPSU form factor. That has made me remember one thing I've always wanted to comment to you. You see, though this is a big community, where you, Ralph and the others provide us with help, articles, and hardware reviews, and where people share their knowledge via the forums, I miss something.
Why isn't there some presence of the industry? Why can't we exchange ideas with the guys that design our PSUs, that cases, that heatsinks? Mike, we are a lot of people, even more if we contacted people down at www.dirkvader.com and other silent sites. Don't you think we could invite people at Seasonic, Thermalright, Aerocool, etc to devote some of their time to share ideas with the people down here in the trenche. They could benefit, we could benefit, don't you think? Maybe you, as founder of this site, could be the perfect spokeperson. Don't you think it's worth the try?
3)
I have a last question to ask you, Mike: I recently saw a thread where you showed your PC. You have suspended your HDs with a mod done by you. I find it is a cheap way to get more silent, and I'd like to do it myself too. Maybe there are more people out there who would like to do this too, and you have made other articles in the past explaining ways to undervolt fans and the like, so I thought you could make one more guide (a short one) showing how to suspend drives correctly, explaining what materials are needed, and how the procedure looks like. What do you think?
EDIT: does anybody have something to say about the idea of putting the HDs out of the case?
Mike, while it is very constructive to think about what manufacturers should do, it is realistic to think about what CAN BE DONE with what we have. The point is to find out the best way to use what there is.
Do you know the fanless Silver Stone? Have a look at it at www.dirkvader.de. Like the Silentium, its enclosure is a huge heatsink. It is ideal for putting it in the outside. I like it because it has active PFC, which prevents it from buzzing (Dirk says passive PFCs use a coil that makes the noise). This makes the Silver Stone completely silent (the Yesico and the SilentMaxx do buzz).
2)
You make some good points about the exoPSU form factor. That has made me remember one thing I've always wanted to comment to you. You see, though this is a big community, where you, Ralph and the others provide us with help, articles, and hardware reviews, and where people share their knowledge via the forums, I miss something.
Why isn't there some presence of the industry? Why can't we exchange ideas with the guys that design our PSUs, that cases, that heatsinks? Mike, we are a lot of people, even more if we contacted people down at www.dirkvader.com and other silent sites. Don't you think we could invite people at Seasonic, Thermalright, Aerocool, etc to devote some of their time to share ideas with the people down here in the trenche. They could benefit, we could benefit, don't you think? Maybe you, as founder of this site, could be the perfect spokeperson. Don't you think it's worth the try?
3)
I have a last question to ask you, Mike: I recently saw a thread where you showed your PC. You have suspended your HDs with a mod done by you. I find it is a cheap way to get more silent, and I'd like to do it myself too. Maybe there are more people out there who would like to do this too, and you have made other articles in the past explaining ways to undervolt fans and the like, so I thought you could make one more guide (a short one) showing how to suspend drives correctly, explaining what materials are needed, and how the procedure looks like. What do you think?
EDIT: does anybody have something to say about the idea of putting the HDs out of the case?
-
MikeC
- Site Admin
- Posts: 12285
- Joined: Sun Aug 11, 2002 3:26 pm
- Location: Vancouver, BC, Canada
- Contact:
Pirata --
1) To spell out exactly what's wrong about making fanless PSUs in ATX form factor and selling them as ATX PSUs...
A large portion of buyers don't read SPCR, and don't understand thermal issues in PCs. They see the product, the techno-promo info on the box or the website and say "Hey, this could really help me make my computer quieter, let's do it!"
If the product can't be used safely without special conditions, then those special conditions MUST be spelled out! Not doing so is bad misinformation. If the engineers of these prodicts don't understand that thier products cannot be used safely in a quiet low airflow PC, then SPCR has a responsibility to point this out to the potential buyers AND to the engineers -- at least that's my PoV as a reviewer.
And none of the fanless PSUs I have seen look like they could last at all in a PC that's thermally modest by today's standards -- say a P4-2.8C, which is the slowest coolest P4 Intel is making these days, and ~70W "thermal target." (which is hooey -- it gets hotter...)
In the context of discussion here in the forums, sure, OK, you can buy these things and put them on the outside of the case. It's a no brainer, but only a few people wll accept this. Lots more people will buy them and install them as ATX PSUs.
2) I am in contact with manufacturers constantly. I go to trade shows not only to find out what is new, but to present the quiet computing PoV, and ask questions from a quiet quiet computing PoV. Many engineers and product development engineers from PC component companies visit SPCR regularly. They troll the site for ideas. There are lots of products that have come on to the market as a result of direct SPCR influence, or been modified as a result thereof. So I think SPCR already has a fair amount of influence.
There are two core factors in product development:
-- Management/marketing generally drive product features. Invariably these are based on what they believe will sell better and more profitably.
-- Engineering do their best to deliver what management/marketing want while still conforming to safety and other industry standards. They generally do NOT have the opportunity to play with and mod the gear ad nauseum the way most of us amateur enthusiasts do.
It takes at least one champion with a fair amount of influence in any organization to make inroads into changing product development. It's high risk for them to try and drive such changes, due to the nature of corp. environments.
I doubt that a "consortium of silent PC enthusiasts" could really influence the industry any more than we are already doing. Maybe if we could organize a trade show or Silent PC "road tour" to demonstrate the large interest... But I am not doing it. Way too much work, too much risk,not enough guaranteed pay.
3) But this is so simple, and experimentation is key. Just one word of advice: The elastic for clothing is the best stuff I've used. It is really cheap and very effective. In fact, the original 2-drive suspension I made ~2 years ago is still working fine in my main rig! The key to effective vibration isolation is how soft the suspension is: As long as the drive can be stable, make the suspension as soft was you can -- ie, dont stretch it too tight.
1) To spell out exactly what's wrong about making fanless PSUs in ATX form factor and selling them as ATX PSUs...
A large portion of buyers don't read SPCR, and don't understand thermal issues in PCs. They see the product, the techno-promo info on the box or the website and say "Hey, this could really help me make my computer quieter, let's do it!"
If the product can't be used safely without special conditions, then those special conditions MUST be spelled out! Not doing so is bad misinformation. If the engineers of these prodicts don't understand that thier products cannot be used safely in a quiet low airflow PC, then SPCR has a responsibility to point this out to the potential buyers AND to the engineers -- at least that's my PoV as a reviewer.
And none of the fanless PSUs I have seen look like they could last at all in a PC that's thermally modest by today's standards -- say a P4-2.8C, which is the slowest coolest P4 Intel is making these days, and ~70W "thermal target." (which is hooey -- it gets hotter...)
In the context of discussion here in the forums, sure, OK, you can buy these things and put them on the outside of the case. It's a no brainer, but only a few people wll accept this. Lots more people will buy them and install them as ATX PSUs.
2) I am in contact with manufacturers constantly. I go to trade shows not only to find out what is new, but to present the quiet computing PoV, and ask questions from a quiet quiet computing PoV. Many engineers and product development engineers from PC component companies visit SPCR regularly. They troll the site for ideas. There are lots of products that have come on to the market as a result of direct SPCR influence, or been modified as a result thereof. So I think SPCR already has a fair amount of influence.
There are two core factors in product development:
-- Management/marketing generally drive product features. Invariably these are based on what they believe will sell better and more profitably.
-- Engineering do their best to deliver what management/marketing want while still conforming to safety and other industry standards. They generally do NOT have the opportunity to play with and mod the gear ad nauseum the way most of us amateur enthusiasts do.
It takes at least one champion with a fair amount of influence in any organization to make inroads into changing product development. It's high risk for them to try and drive such changes, due to the nature of corp. environments.
I doubt that a "consortium of silent PC enthusiasts" could really influence the industry any more than we are already doing. Maybe if we could organize a trade show or Silent PC "road tour" to demonstrate the large interest... But I am not doing it. Way too much work, too much risk,not enough guaranteed pay.
3) But this is so simple, and experimentation is key. Just one word of advice: The elastic for clothing is the best stuff I've used. It is really cheap and very effective. In fact, the original 2-drive suspension I made ~2 years ago is still working fine in my main rig! The key to effective vibration isolation is how soft the suspension is: As long as the drive can be stable, make the suspension as soft was you can -- ie, dont stretch it too tight.
MikeC, in all fairness to Pirata's idea, how is suggesting that users put their PSU outside the case (something that you yourself have done and demonstrated to others) going to cause any more "issues" than suggesting that users create "fresh-air PSU ducts"? I mean, extending/routing/sleeving cables isn't that big of a deal to me, whereas trying to create some kind of makeshift duct between the PSU and the front of my case and probably leaving a gaping hole in my front bezel does not exactly sound like it is devoid of issues.
On the other hand, I agree with you that companies producing fanless PSUs that are marketed as drop-in replacements for regular ATX PSUs are misleading the general public.
On the other hand, I agree with you that companies producing fanless PSUs that are marketed as drop-in replacements for regular ATX PSUs are misleading the general public.
-
MikeC
- Site Admin
- Posts: 12285
- Joined: Sun Aug 11, 2002 3:26 pm
- Location: Vancouver, BC, Canada
- Contact:
Fair enough. I am not dissing Pirata's idea or suggestion, just that given the $$$ charged for fanless PSUs, their being marketed as ATX PSUs is a bit of a joke -- and a travesty -- to the buying public, who cannot always be informed -- certainly not as well informed as SPCR regulars. I'll keep saying that -- until I find a fanless ATX PSU that really works as intended.Putz wrote:MikeC, in all fairness to Pirata's idea, how is suggesting that users put their PSU outside the case (something that you yourself have done and demonstrated to others) going to cause any more "issues" than suggesting that users create "fresh-air PSU ducts"?
No argument from me on that point! Imagine the users with no case exhaust/intake fans... if they bought one of those PSUs without knowing better... Well, their computers would be very silent shortly after that!MikeC wrote:Fair enough. I am not dissing Pirata's idea or suggestion, just that given the $$$ charged for fanless PSUs, their being marketed as ATX PSUs is a bit of a joke -- and a travesty -- to the buying public, who cannot always be informed -- certainly not as well informed as SPCR regulars. I'll keep saying that -- until I find a fanless ATX PSU that really works as intended.
EDIT: removed nested quote
@MikeC: if the industry is in here, it happens in such a silent way, that nobody notice it. I'd like to be able to talk to them or to listen to them. Something visible. It's just having one or two guys visiting the forums dayli or weekly. Every corporation could do it. It's just that they ar not used to listen the people that buy their products. That should change. We should change it. Specialized public opinion, that's us, should be one more strong driving element, and market staff should be a filter to that. It is not so these days.
That SPCR influence... I just can't see it.
Do you really think that presenting one product with mods suggested by SPCR could be such a terrible risk from an economical point of view? The question is whether we have ideas that are good and companies still don't realize they are good.
A road tour is not needed. It is happening right now, right here.
As to the suspending thing: what material do you use?
That SPCR influence... I just can't see it.
Do you really think that presenting one product with mods suggested by SPCR could be such a terrible risk from an economical point of view? The question is whether we have ideas that are good and companies still don't realize they are good.
A road tour is not needed. It is happening right now, right here.
As to the suspending thing: what material do you use?
-
peteamer
- *Lifetime Patron*
- Posts: 1740
- Joined: Sun Dec 21, 2003 11:24 am
- Location: 'Sunny' Cornwall U.K.
Exactly, that's the whole point of the site isn't it?@MikeC: if the industry is in here, it happens in such a silent way, that nobody notice it.
Proves just how much they have learnt from us..... just as MikeC said.
Back on subject though... Blow/Suck... Baloney !!!
Why compromise?.... Side draught (With a slight down) is the answer...
Discuss....
Pete
-
MikeC
- Site Admin
- Posts: 12285
- Joined: Sun Aug 11, 2002 3:26 pm
- Location: Vancouver, BC, Canada
- Contact:
"elastic for clothing" --
Go to a sewing supply shop. Where you can buy cloth, thread, etc. They have rolls of different kinds of elasticzed cord that is used in clothing -- like for waistbands, etc. There's different cross sectrion, thickness, stretchiness, color. The stuff I get is round cross-section, Costs ~CA$1/meter. Less if you buy lots. Good opportunity to chat up ladies that catch your eye: Men are automatically unusual and this interesting in such places by virtue of simply being there.
Go to a sewing supply shop. Where you can buy cloth, thread, etc. They have rolls of different kinds of elasticzed cord that is used in clothing -- like for waistbands, etc. There's different cross sectrion, thickness, stretchiness, color. The stuff I get is round cross-section, Costs ~CA$1/meter. Less if you buy lots. Good opportunity to chat up ladies that catch your eye: Men are automatically unusual and this interesting in such places by virtue of simply being there.
-
peteamer
- *Lifetime Patron*
- Posts: 1740
- Joined: Sun Dec 21, 2003 11:24 am
- Location: 'Sunny' Cornwall U.K.
'Across' the heat sink, through the fins. Not quite at 90 degs, with a slight angle 'down' towards the base plate."Side draught (With a slight down)"
Eliminates 'Dead Spots', reduces resistance to the flow.
Both of which mean you need less 'power' from the airflow, which obviously means more flow per fan V or greater cooling.
Use something (Thin cardboard in my case) to put a 'lid' across the top of the heat sink to form a funnel type affair and control/direct the airflow to maximum effect.
-
Ralf Hutter
- SPCR Reviewer
- Posts: 8636
- Joined: Sat Nov 23, 2002 6:33 am
- Location: Sunny SoCal
"Hey baby, maybe you'd want to come over one evening so I can show you my suspended Spinpoint with the 8MB cache".MikeC wrote:Good opportunity to chat up ladies that catch your eye: Men are automatically unusual and this interesting in such places by virtue of simply being there.
Now that'll make 'em weak in the knees!
Thanks for that helping help.
This elastic cord... will it endure the high temperatures without starting to smell weirdly, maybe fusing?
Edit: how thick is the cord you use? Can you name the material that cord is made of?
This elastic cord... will it endure the high temperatures without starting to smell weirdly, maybe fusing?
Edit: how thick is the cord you use? Can you name the material that cord is made of?
Last edited by Pirata on Sat May 29, 2004 7:25 am, edited 1 time in total.
-
MikeC
- Site Admin
- Posts: 12285
- Joined: Sun Aug 11, 2002 3:26 pm
- Location: Vancouver, BC, Canada
- Contact:
Ralf, Ralf, Ralf.... Didn't your mom teach you how to talk to the girls?Ralf Hutter wrote:"Hey baby, maybe you'd want to come over one evening so I can show you my suspended Spinpoint with the 8MB cache".
Now that'll make 'em weak in the knees!
Try: "Oh, um, do you know where I could find <insert choice of fabirc, fasterner, etc.> in this store? It's my first time in here and I'm kind of confused.." The beginning of the chit & followed by her chat which naturally you respond with your chit... and before you know it the two of you are buying meters of satin and lingerie fabrics to try out together.
... 
....-
MikeC
- Site Admin
- Posts: 12285
- Joined: Sun Aug 11, 2002 3:26 pm
- Location: Vancouver, BC, Canada
- Contact:
No guarantee at all.Pirata wrote:Getting back to reality (
It sounds to me like you haven't tried it. Just try it.
How to make it secure enough -- you'll figure it out once you start messing around with it. It's unlikely that you could "accidentally shake" the PC and have the drive fall out of an elastic suspension. An earthquake maybe.
Read this article (the 1st in the storage section!) again carefully and examin the pictures of the suspended drives.
And if you're so anxious about this, just BUY a NoVibes or similar...
If you are doing completely passive convection cooling, it would make sense to have everything possible (PS, HDDs, CPU, VGA ...) dumping heat into the same chimney shaft. The hotter the chimney air relative to surroundings, the stronger the convection effect. The power supply would add another 30% or so total heat to the heat generated by everything else.
That would really take dedicated case design though ... I'm imagining here a chimney shaped computer ... with a narrowed section to get all the air to flow through the CPU HSF ... or maybe just a 5 CFM fan here or there to mix air across various heatsinks.
It would be a little bit like the Zalman TNN case, but turned inside out - with the massive heatsinks inside the shaft.
By the way I also have to wonder why Zalman didn't try to make a chimney for the external heatsinks on that thing. Not worth it ...? The vertical fins make a sort of chimney anyhow ...?
the wesson
PS if you were concerned about the drives falling out of the elastic, you could always screw the drives to the elastics. Or sew them on ... grin ...
That would really take dedicated case design though ... I'm imagining here a chimney shaped computer ... with a narrowed section to get all the air to flow through the CPU HSF ... or maybe just a 5 CFM fan here or there to mix air across various heatsinks.
It would be a little bit like the Zalman TNN case, but turned inside out - with the massive heatsinks inside the shaft.
By the way I also have to wonder why Zalman didn't try to make a chimney for the external heatsinks on that thing. Not worth it ...? The vertical fins make a sort of chimney anyhow ...?
the wesson
PS if you were concerned about the drives falling out of the elastic, you could always screw the drives to the elastics. Or sew them on ... grin ...
-
Ralf Hutter
- SPCR Reviewer
- Posts: 8636
- Joined: Sat Nov 23, 2002 6:33 am
- Location: Sunny SoCal
I think Mike means this article.MikeC wrote:Read this article (the 1st in the storage section!) again carefully and examin the pictures of the suspended drives.
And, IMHO, you should read this one as well.
All 3 of the OEM P4 machines I've looked inside have this arrangement: duct the CPU HS to exhaust. There is one fan in the entire case besides the PS fan. They use one case fan on exhaust, with a curved plastic duct to the HS, to cool both the CPU and the case.
The HS involved is generic aluminum finned heatsink - tho fairly large. There is no fan on the HS.
So Dell and Compaq think that the CPU should be ducted out, despite the loss of HS efficiency. Why is that, do you think?
Maybe it's the best way to cool a computer with minimal fans and fan noise.
the wesson
The HS involved is generic aluminum finned heatsink - tho fairly large. There is no fan on the HS.
So Dell and Compaq think that the CPU should be ducted out, despite the loss of HS efficiency. Why is that, do you think?
Maybe it's the best way to cool a computer with minimal fans and fan noise.
the wesson
On the Poweredge700 we have at work, and the P4 machine I am getting at work, that happens to also be a dell, THen heatsinks are not just generic aluminum ones. THe one in the poweredge is quite tall and and has fins in a stack with 3 heat pipes coming out of it and ending at the top. THe dusting exhaust fan is a delta, that is rpm controlled through a thernal sensor controlled by the motherboard. That fan is capable of pulling 20watts!!! of power, which is an awful lot. It is 38m thick like the panaflo 120's are. I have not seen my new 8300 in the flesh yet, but I will post pics of either machine if requsted.