Tinker tinker little hs

[walterm]

Found an article comparing AMD heatpipe heatsinks. The older version outperformed the newer and neither are as good as they seemed to me.
Home Improvement time..
Since I have three of these heat sinks, the 70 mm fans remove easily, and I have a selection of fans..
I've purchased an assortment of fans, and some 80mm to 120mm adapters.
The heatsink fins are spaced close together, my understanding an increase in pressure rather than CFM is the solution.
An 80mm X 38mm high rpm would probably work but noise would be unacceptable.
IF I didn't already have them it would be silly, may be silly anyway.
Will try various Yate Loon fans, though may have to switch to others.
Opinions comments Knowledge welcome.
The adapters taper 80 to 120mm in one inch, which should not hamper pressure flow, and additional air should cool motherboard.
And yes I now know the Cooler Master Hyper 212 plus is on Amazon for $28. Among other choices. Now.
Adapters were intended to aid air flow in older cases where possible. Found two versions at about $6 shipped (one has 80mm and 92mm mounting holes).
So daffy as I am after this will try to improve cooling on Arctic Cooling Freezer PRO 7 Rev 1 & 2 oooah bboy.:wall::wtf:rant:

[MrWicked1968]

wow, your stream of consciousness posts are a lot of fun.

[mdk777]

limiting factors are not fans.

1. only two heat-pipes (wrapped around twice on stock AMD) Since it is the water evaporation inside the heat-pipes that moves the heat away from the CPU.
2. Copper plate on stock AMD (not direct contact to heat-pipes..limits rate of transfer to heat-pipes and fins)
3. fin surface area is limited

Increasing the fan speed and pressure only addresses the last limiting factor.

Consequently, on Heat-pipe coolers, improved performance is achieved by :

1. direct contact heat-pipes (fastest way to remove heat)
2. More heat pipes, 4, 5 and more See one and two on the AMD best Coolers on this list http://www.frostytech.com/top5heatsinks.cfm
3. for fin surface area (larger fins and higher density)

After doing these three things, you can adjust the fan to suit your demand for cooling verses silence. However, it is the least important factor of the three.

But I guess the point of your post was that you were going to confirm what the entire heat-sink industry has confirmed over the last 10 years by personal experimentation?

[walterm]

sheesh, crazy may be, willfully stupid NOT.
my surmuse was heat sinks rated for 125 watts should function well at 65 or 80 watts, or even 95.
The article I saw proclaimed the stock heat sinks marginal and loud on 125 watt and not much better on 95 watt CPUs.
They did find that on 4 of 5 of the aftermarket heat sinks tested (including a $20 Rosewill) increasing static pressure by changing fans yielded improvement with less noise.
I have Semprom 140 (which I will try to unlock and overclock) that came with a meager cooler, an Athlon II x2 245, bought used without heat sinks, a new Phenom II x2 545 oem, an Athlon II x3 435 with dinky cooler, and I'm contemplating a Athlon II or Phenom II 4 core as my primary build.
Silly but I need a hobby distraction just now, and Ive never found differences between gestalt immersion learning and more formalized study.
Formal study builds in a logical manner, directives till you have experience.
Gestalt has more oh no, do it over , what did I do moments. For me its more fun, I get to play, and while it won't be a formal education, it'll get me where I want to go.
Gah, too much ..
I do not expect miracles, I just feel, for the cost of some dremel work I can create, or at least improve, what I got.
Fifty years ago some smart mechanics gave Factory design teams fits in motorcycle racing.
Recently, check the career of Valentino Rossi. Yamaha hired him (he insisted he bring his tuner), They turned a technically leading bike into a winner. Wasn't luck. They did the same for Honda first.
No I'm not one of them.
I'm close but never finished an engineering degree.
And I have seen and experienced designs that could work better (who put a 70mm fan on these things, lordy me oh mi).
Blasted know it alls.
uuMMM, of the coolers tested as delivered the the heat pipe thru a base outperformed the heatpipe contact coolers. just cause....
And if any of those Frosty Tech coolers are in the $20 to $40 range tell me which.
I got these new $18 ea shipped. Hey, didn't know beter at time.

[mdk777]

And if any of those Frosty Tech coolers are in the $20 to $40 range tell me which.
I got these new $18 ea shipped. Hey, didn't know beter at time.

You take a $20 cooler and add a $20 fan and you get a $40 cooler that is limited by its core design.

Better to buy the $40 to $60 cooler and get the best core design. Then your modifications of fan speed and pressure will result in the greatest benefit.

You are just throwing good money after bad. It is not an efficient design process.:D

Here in the states they have riding lawn mower races. Sure it can be done, but for half the price you could build a much better, faster, lighter, go-cart from scratch. Your analogy of motorcycle racing is correct, you start with the best technology and tune it to make it better. You don't start with a known limited technology and spend a great deal of time and money to bring it up to adequate. :)

[Digerati]

my understanding an increase in pressure rather than CFM is the solution.Can you have one without the other? Perhaps in a sealed unit, but I don't know of any case that is sealed.

I think it is all about basics - materials, surface area, and air movement.

They did find that on 4 of 5 of the aftermarket heat sinks tested (including a $20 Rosewill) increasing static pressure by changing fans yielded improvement with less noise.Sorry - you (or that article) lost me. Static to me means stationary - not moving. Static pressure in a duct (between the fins) is a bad thing - and a force the fan must overcome. You want dynamic pressure, no? You want to move that heat away.

you start with the best technology and tune it to make it better.As long as we remember that the better makers have engineers with advanced degrees and adavnced engineering tools designing products on the cutting edge of technology. Do we really have the qualifications to make it better? You don't start with a known limited technology and spend a great deal of time and money to bring it up to adequateFledgling antique or classic car collectors may disagree. ;)

[mdk777]

Do we really have the qualifications to make it better?

tune it to your specific CPU(die size), OC, case layout, required cooling verse noise tolerance, etc. etc. etc. :D


Sorry - you (or that article) lost me. Static to me means stationary - not moving. Static pressure in a duct (between the fins) is a bad thing - and a force the fan must overcome. You want dynamic pressure, no? You want to move that heat away.

Fans are designed two ways.

1. to move the greatest amount of air (case fans)
or
2. to push air through an obstruction (restriction of fins)

What the OP is describing is the type of fan designed to maximize the pressure against a restriction. (type 2 fan)

He is correct, but as I mentioned before, it is the final optimization, after you have the best design otherwise.

[Digerati]

tune it to your specific CPU(die size), OC, case layout, required cooling verse noise tolerance, etc. etc. etcRight. Or, you can buy the right one in the first place. ;)

Fans are designed two ways.

1. to move the greatest amount of air (case fans)
or
2. to push air through an obstruction (restriction of fins)

What the OP is describing is the type of fan designed to maximize the pressure against a restriction. (type 2 fan)I agree with that too. I guess it is just a matter of semantics. When I think of something being "static", it is being still, not moving. So by changing the fan dynamics (rpm, length, width, depth, pitch, etc.) you change both the static and dynamic pressures on and through the obstruction. But it is increased dynamic pressures you seek when the objective is to move heat away the heatsink.

[mdk777]

I agree with that too. I guess it is just a matter of semantics.

NO.

It is a standard measure in the industry.
See any detailed fan specification.
Different fans will develop different back pressures depending on the fan blade.
Some fan blades will move air efficiently at a given RPM with low back pressure.
Others will move air efficiently at a given RPM with higher back pressures.
Through-put and back pressure, (or static pressure) are the way fans are described.
The analogy would be the difference between horse power and torque.
Both describe engine performance, but describe difference aspects of that performance.

It is not only semantics. It is that you don't understand the technology, or how the engineering specifications are used.

[Digerati]

The only semantics is that you don't understand the technology, or how the engineering specifications are used.
Let's not turn it personal, okay? I've studied aeronautics - I understand how fans works. I am afraid you don't understand what I am saying - perhaps that's my fault.

My comments referred to CPU coolers - not fans in general as your last post suggest. And specifically I was referring to walterm's comment, "They did find that on 4 of 5 of the aftermarket heat sinks tested (including a $20 Rosewill) increasing static pressure by changing fans yielded improvement with less noise."

I did not disagree with anything you said, mdk777. What I am saying is cooler makers seek dynamic pressure, not static pressure. They select a fan that will achieve the necessary dynamic pressure needed to move the heat away. Static pressure in a cooler is unwanted. Fan makers may try to achieve static pressure, but cooler makers need the air to flow through, so they seek dynamic.

[mdk777]

OK,
Without getting into the specifics of cooler design, what the OP said is still correct. He was describing changing fans on a specific cooler.

increasing static pressure by changing fans yielded improvement with less noise."

You said you didn't understand.

Sorry - you (or that article) lost me. Static to me means stationary - not moving. Static pressure in a duct (between the fins) is a bad thing - and a force the fan must overcome. You want dynamic pressure, no? You want to move that heat away.

Now you say you understand, but were talking about something entirely different.

My comments referred to CPU coolers - not fans in general as your last post suggest.

OK, so in summation;
1. the OP was correct.
2. You said you didn't understand the term static pressure(in relationship to the fan change the OP referenced)
3. now you say you understand, but were not in fact responding to the specific wording in the OP post; the very wording you in fact quoted in your own response!! (TWICE NOW)

"They did find that on 4 of 5 of the aftermarket heat sinks tested (including a $20 Rosewill) increasing static pressure by changing fans yielded improvement with less noise."

You either now understand how static pressure is measured or you don't.
You really need to either research what you are going to talk about, or at least admit when you learn something new.

[Digerati]

You said you didn't understand
now you say you understandYou have said nothing I didn't already know or understand. I understand how fans works. I understand how HSF assemblies work. What I have said multiple times now concerns what the OP related from the article he mentioned.

You really need to either research what you are going to talk about, or at least admit when you learn something new. I did research static and dynamic pressures before I commented to verify what I thought to be true. Did you? Static pressure through a heatsink is unwanted resistance. And thanks again for keeping it professional and avoiding the personal attacks. :rolleyes:

[mdk777]

Static pressure through a heatsink is unwanted resistance.

1. again, not what the OP was talking about.
2. incorrect.

in-order for heat transfer to take place, the air has to be in contact with the fin. A certain amount of static pressure is required to allow this contact time.

If the air flows non-laminar, it will reduce the static pressure, but also reduce the contact time. Maximum heat exchange efficiency is not achieved by the lowest resistance.

There is no way to have a professional discussion with you. You lack a basic understanding of the topic.

[Digerati]

There is no way to have a professional discussion with you. You lack a basic understanding of the topic. Professional? There's no place for personal affronts in professional discussions, yet that's your typical approach. Bash anyone who has a different view point than you! What's up with that? It's really sad.

You don't want contact time. Heat sinks are "radiators" - they radiate the heat [to cooler surroundings] whether there is air flow or not. But if there is no flow, the surrounding air gets hot, thus reducing the effectiveness of the radiator. The purpose of the fan is to bring in cool air and move the hot air away from the radiator so more heat can jump into the cool air. You want to reduce and/or overcome any static pressure to allow the dynamic pressure to move the heat out. The dynamic pressure used to compensate for the static pressure is lost efficiency in the cooler.

If you disagree with that, then I guess we just disagree. But stop attacking the person because you don't agree with them. It really is not cool.

[mdk777]

You don't want contact time.

Yes, you have shown that you have done absolutely no research on this topic.

Every word of your last post is 100% wrong.

Please post again when you have bothered to read something on the topic.

[Zero82z]

You don't want contact time. Heat sinks are "radiators" - they radiate the heat [to cooler surroundings] whether there is air flow or not. But if there is no flow, the surrounding air gets hot, thus reducing the effectiveness of the radiator. The purpose of the fan is to bring in cool air and move the hot air away from the radiator so more heat can jump into the cool air. You want to reduce and/or overcome any static pressure to allow the dynamic pressure to move the heat out. The dynamic pressure used to compensate for the static pressure is lost efficiency in the cooler.
You do not understand what static pressure is. It has nothing to do with resistance from the heatsink fins (that is drag and has no direct relation to the state of the air itself).

Yes, you do want contact time, because in order for energy to be transferred from the heatsink to the air, they must be in contact. The amount of heat transfer from one body to another depends on a number of factors, one of which is time. The longer the two bodies are in contact, the more heat is transferred from the hotter one to the colder one (up until they have reached thermal equilibrium). Radiation is a very small component of the total heat transfer from a computer heatsink to the air, since it is very inefficient and slow compared to convection. That's why we use fans rather than passive cooling.

[walterm]

What I am trying to say, apparently badly, is fan size, rpm and blade design effect air flow THRU the heat sink. Dense, by which I mean closely spaced fin design, impedes the amount of cooling air available exiting the heatsink.
The closer the pressure out of the fan assembly to the pressure out of the heat sink the more affective the cooling.
IT appears that in balancing costs/performance/appearance/size/CFM ratings some heat sink fans are less capable than ideal.
It is interesting that people who find most power supplies misrepresented and or poorly engineered expect much better of heat sinks.
Especially as case cooling changes dramatically and test procedures vary widely.
As with psus finding excellent top line units is much easier than evaluating low level units.
I can't locate the article I'm using for reference, I have been to frostytech, hardware secrets, toms hardware, anandtech, bjorn3d, madshrimps, and god only knows where else.
One fact is VERY clear, performance tests vary hughly and year by year.
Appearance sells coolers as often as performance if not more so.
I began this adventure believing (silly me) in the year of our lord 2010 that market forces consumer protection and consumer reviews could guide me.
WOW. hIPPY tIME, dON't trust anyone without proof and check your facts. Oh yeah.
Sappy time there are a couple genetic flows in my family hit me late in life age 50 about, I did not handle this well (denial you know). I'm STILL adjusting thanks.
Back on track, heatsinks are designed, sometimes as compromised as psu's, from where I now STAND//.
Time and cases Change their effectiveness.
It is like trying to determine, using the parts I have acquired, the best builds worth completing.
Tell me..
Which of these would you chose and why new EA500D's btwn $20 and $35 no warranty, refurbished ocz600sxs one year, Zalman ZM460-APS $34 no warranty, ZM500-RS $40 3 year warranty, cx400 new $40, cx400 $26 almost new, NEO ECO 520 $40, used EA650, used TruePower Trio 480, and these are the good choices.
FWIW the AMD AM2 cooler with copper base and 60mm fan worked better rhan the newer cooler with 70mm fan. Both went long term without updates which implies changes May be worth while.
The question here is the ability of the fans to push air thru the heatsinks,
IF I understand the reviews in question the heatsink is overdesigned for the fan supplied.
Which is not uncommon, see several "referrence fan" comparasions.
Going to a 92 mm fan is probably an adequate noise reduction performance solution.
Going to a 120 is a what the hell, should help with case limitations experiment.
I expect to do a high end (for me) build eventually. I have spent my budget on bits and pieces about 30% or more wasted out of about two grand.
I hope to budget $100 to $200 a month if I can to my hobby. Right now I need operating systems, no bargains there.
SD, cheap tricks, this is a hobby, for knowledge.
I will end up with ten or twelve computers to sell, trade, or most likely give away.
There is a movie, called "Indian" I think, about an Australian tinker going to speed week.The fact is I want to be able make reasonable decisions and part selections,and approach builds with confidence.
This is working.
Taking classes at the moment would be difficult. and more expensive,
Can you ask yourself, what if I..
And I may have to use a Noctua FAN but I expect benefits,
Stubborne I am
The Force Be With Us.

[Zero82z]

The thing is that the heatsinks you're dealing with are designed to cope with the heat from the CPU they are packaged with when it is running at stock speed, and for that purpose, they are perfectly adequate. They may not produce the lowest temperatures, and they may not allow for much or even any overclocking, but that is not their purpose. If you want something with better performance, then you should be looking at a heatsink designed to perform better, and there is no shortage of those on the market.

[walterm]

Like MAN, i'm playing with what i got to see what we will see.
I will NOT be using these to their design limits, The design uses small fans at high speed to achieve max cooling.
Exactly why should using a large slower fan upset you so?
These things are sold, and were sold, for use with in restrictive (read old) cases on 125 or more watt CPUs.
I have no intent on exceeding this, nor purchasing more of these (other than for 65 or 80 watt CPUs.
lIke they placed a 120mm generic fan (that was used to compare heatsinks performance rather than heatsink Assemblies) in front of an Arctic Cooling Freezer PRO 7 Rev 2 (not practical) but as with 4 out of 5 other units got better cooling.
I'm prepared to fail.
Like you can't launch a decent model rocket without risking losing it.
Like you're gonna tell me intel push pin mountings are like optimal instead of cheap? Harder to change fans there. Price them from Hong Kong. gOT THREE at the price of ONE amd unused new $18.
Lke Prolemart er whatever costs what?
Remember multiple builds here just to compare and practice..:beer:
Idle hands or active mischief, hey hey..

[Zero82z]

Who says I'm upset? I'm just saying that you shouldn't expect to obtain impressive performance from these heatsinks regardless of the kinds of modifications you make. That's all.

[walterm]

Acceptable, is fine, superior to stock in cooling or noise worth while. Which seems likely.
Looks like I can remove four plastic clips and glue adapter to plastic fan holder.
DARN. TOO EASY.

[walterm]

Picking a cpu fan from among all the case fan specifications more dubious than I thought.
I have it turns out FOUR pwm fans to choose from, a NEXUS, an AVG, a Metronic, and a Delta. Metronic and Delta are 32 AND 38 MM thick and heavy.
I have an Arctic Cooling 92 mm rated at 37 CFM max.
The stock 70 mm maxes at 76 CFM AT 5000 rpm. According to specs.

[walterm]

Update, so far I have mounted (glued) an 80 mm to 120 mm adapter to the the plastic frame that originally held the 70mm OEM AVC fan.
FWIW the "ducting effect" seems to in no way interfere with adequate air flow. The "better" fans [although all chosen for their specified ability to move air fins] simply can accomplish more at less RPM and more quietly.
I am using a stock heat sink that "bundled" with 140 watt CPUs. Using a 95 watt CPU, is not straining cooler performance. Haven't tried core unlock or over clock yet.
Definitely NOT a reasonable project:
HEATSINK $18, 80 to 120 mm adapter $ 4 to $6, Nexus fan $16. [didn't have to count glue, a friend does some plastic repair at work].
Cost about $40, so..
Cooler Master Hyper 212 Plus is about $ 26 and change shipped from Amazon..
I did get a "lightly used" OCZ Vendetta 2 for $11.99 on epay [seller says cost that to ship, I believe him]. Waiting on a Xigmatech (maybe lucky twice?) too..
BUT I do have a horizontal cooler that works, possibly better, that I can run with led fans. And so far,I like it, ha;
Good horizontal coolers are EXPENSIVE.

Hmmm wonder if I could rig an electrostatic dust collector.. nah.. wweell,,, nooo.