Water + Heat Pipes (MOD pics)

You guys all need to settle down. We can easily start handing out vacations if you don't wish to play by the rules.
 
you know I think you guys are ripping on this guy a little hard.

"no one is THAT stupid"

you'd be stupid not to figure this out: air cooling = norm, water cooling = good, water + air cooling = better. He's after the right concept and the execution certainly could be worse.

Now, IMHO, to get better temps, you first gotta take off the stock air hs and re-apply your tim. The stock stuff has never been anything better then "nasty".

Now If you really wanna take this seriously, you gotta modify that stock heatsink to conduct heat to the top better. I would suggest you fit a couple pieces of thick sheet copper between the stock copper fins. apply a little bit of sauder the bottom of this copper plate so that it contacts the base of the stock heat sink. file/sand until its flat and shiny. Keep in mind thats gonna take several hours.
 
Why do you seem to think that he has to prove anything? Its one of thousands of mods here, and far as Im concerned, one of the more interesting ones. Trashing his idea over and over is not making your E-penis any bigger.
I mean really kid, you went on other forums and made threads trashing him.

He is doing it, he is using the tools at hand to show results. If you dont like those results, then you reproduce the experiment yourself. Put up, or shut up.

People use what they have on them. I have a similar meat thermometer and its accurate within 4 degrees C at 500 degrees.
I thought, ah heck, I have heatpipes and a similar NB setup to his, Ill take my meat thermometer and check different spots on my NB heatsink under load. You know what? Im getting 1-2 degrees difference. He is getting 30 degrees change. With screens, pics and temps, its acceptable proof for me considering the tools he has on hand. Like I said, put up, or shut up.
Cheers

Sorry that's not how this kind of thing works. You don't make a claim and then force everyone else to prove you wrong, that's the total opposite of the scientific method. A theory is not true until proven wrong, you have to prove that your idea is right especially if it goes against the accepted thinking. This logic doesn't hold anymore water than his original idea.

And as far as his results being accurate, a meat thermometer is NOT meant to accurately measure in this kind of environment. IT makes more contact with the air than the heatsink itself, giving you a mixed reading. As far as screenshots, he posted ambient case temps... what does that have to do with anything? If you need proof that a meat thermometer is a bad way to measure this sort of thing, look at the ambient case temps he posted and the temperature that his heatsink is supposedly at... his case is roughly 15degrees HOTTER than the northbridge..... you cannot cool anything to 70c with a waterblock, when the air your cooling the water with is HIGHER than 70c. His external radiators are going to be sucking in air that is at most 84c and at least the temp in his room. To achieve a block temp of 70f with that loop he'd need to be pulling frigid air through his radiators.
 
Sorry that's not how this kind of thing works. You don't make a claim and then force everyone else to prove you wrong, that's the total opposite of the scientific method. A theory is not true until proven wrong, you have to prove that your idea is right especially if it goes against the accepted thinking. This logic doesn't hold anymore water than his original idea.

Unfortunately for you, yes that's exactly how the science community works. One person or group publish the idea, the test procedure and the gained results. Another group (or many groups) will follow the procedure given to see if the results are reproducible. If they aren't, they'll try to find out why. Maybe they'll find that the measuring devices aren't accurate and reliable. Maybe they'll find the original methodology at fault. But they DO try to prove it wrong. It's called peer review. You don't just sit back and say "Prove it, prove it, do this, do that," So if you're so concerned about his results setup an experiement yourself and prove him wrong. And you're wrong about theories. A theory can never be proven true, just acceptable considering the given environment. However, theories are proven wrong all the time when an anomaly pops up that the theory can't explain away.


And as far as his results being accurate, a meat thermometer is NOT meant to accurately measure in this kind of environment. IT makes more contact with the air than the heatsink itself, giving you a mixed reading. As far as screenshots, he posted ambient case temps... what does that have to do with anything? If you need proof that a meat thermometer is a bad way to measure this sort of thing, look at the ambient case temps he posted and the temperature that his heatsink is supposedly at... his case is roughly 15degrees HOTTER than the northbridge..... you cannot cool anything to 70c with a waterblock, when the air your cooling the water with is HIGHER than 70c. His external radiators are going to be sucking in air that is at most 84c and at least the temp in his room. To achieve a block temp of 70f with that loop he'd need to be pulling frigid air through his radiators.

Who's to say that the ambient sensor is even remotely accurate? or that it's not postitioned near a heat producing component? You're being entirely too critical of the man's setup.
 
Unfortunately for you, yes that's exactly how the science community works. One person or group publish the idea, the test procedure and the gained results. Another group (or many groups) will follow the procedure given to see if the results are reproducible. If they aren't, they'll try to find out why. Maybe they'll find that the measuring devices aren't accurate and reliable. Maybe they'll find the original methodology at fault. But they DO try to prove it wrong. It's called peer review. You don't just sit back and say "Prove it, prove it, do this, do that," So if you're so concerned about his results setup an experiement yourself and prove him wrong. And you're wrong about theories. A theory can never be proven true, just acceptable considering the given environment. However, theories are proven wrong all the time when an anomaly pops up that the theory can't explain away.

I have proved it wrong, with logic. He has provided nothing that could even be considered proof. Peer review assumes that the originator of the idea has some reason for believing it to be true, which is not the case here. He has no reason why this should work, and no proof that it does. I'm not asking him to dismantle the system because it's dumb, I'm just asking that he back his claims regarding it's performance (and I'm sorry but a meat thermometer is by no stretch of the imagination accurate). If he has some kind of reason for me to think it works, I would gladly test it... but it's tough to swallow mutilating a good block for seemingly no reason.

Who's to say that the ambient sensor is even remotely accurate? or that it's not postitioned near a heat producing component? You're being entirely too critical of the man's setup.

He's claiming his northbridge is at 21c... that in itself should raise some questions as to the accuracy of his measuring device. Positioning of the thermometer, as well as the amount of contact it has are probably the BIGGEST factors effecting his temps.

The 100c test has "good" contact, at the base of the heatsink. The 70c test has terrible contact at the top. The combination of location and contact is gonna give a HUGE difference in temps.

Oh and let me restate... 21c northbridge.. on water.
 
I have proved it wrong, with logic. He has provided nothing that could even be considered proof. Peer review assumes that the originator of the idea has some reason for believing it to be true, which is not the case here. He has no reason why this should work, and no proof that it does. I'm not asking him to dismantle the system because it's dumb, I'm just asking that he back his claims regarding it's performance (and I'm sorry but a meat thermometer is by no stretch of the imagination accurate). If he has some kind of reason for me to think it works, I would gladly test it... but it's tough to swallow mutilating a good block for seemingly no reason.

My god you act like he's killing his baby's momma. The block's surface can be repaired. And I never said a meat thermometer was accurate.

He's claiming his northbridge is at 21c... that in itself should raise some questions as to the accuracy of his measuring device. Positioning of the thermometer, as well as the amount of contact it has are probably the BIGGEST factors effecting his temps.

The 100c test has "good" contact, at the base of the heatsink. The 70c test has terrible contact at the top. The combination of location and contact is gonna give a HUGE difference in temps.

Oh and let me restate... 21c northbridge.. on water.

Don't get me wrong, I'm not saying his methodology is anywhere near accurate or reliable.
 
He has no reason why this should work
Maybe he doesn't, but I do. Contact is contact. Even a tiny bit of it will allow for heat transfer. Now, without thermal paste, or a lot of pressure, I don't see this contact actually alotting to much... But it's certainly not harmful. He's not blocking airflow to the NB. He's not circumventing convection. He's probably increasing contact between the NB and the NB HSF with the extra pressure. All of which are good for the chip in the end (or, at least not bad)

When compared to a NB with a fan, yes, I'm sure this method lags behind. But, it's much quieter than a goddamned 40mm fan. Even at 5v those things rate around 28dba, which is way too loud.

He's claiming his northbridge is at 21c..
No. He's not stating that at all. He's saying (And showed with a thermometer) that there's a 21f DELTA between his method and a bare NB. 100f without the block. 79f with it. That's 21f

There was never any talk about Celsius, and he was never claiming his NB temperature was lower than his ambient temperature.

The combination of location and contact is gonna give a HUGE difference in temps.

This I agree with. I think the test needs to be re-done. However, I don't think the new results will work in your favor Fujimitsu.
 
Fujimitsu, if your not going to reproduce the situation and prove him wrong yourself, then shut up. Its simple dude, just stop typing.

The [H] community encourages new and exciting things...Perhaps this community isnt right for you.
 
All I know is the guy has a Q6600 running at 4.1GHz. That's no small feat. If this works, then it works.. I'm more inclined to believe his setup is beneficial before I believe he is an idiot and doing it wrong.
 
Fujimitsu, if your not going to reproduce the situation and prove him wrong yourself, then shut up. Its simple dude, just stop typing.

The [H] community encourages new and exciting things...Perhaps this community isnt right for you.

We are all up for new and exciting things, as long as it can be reproducted and proven to work. I'm willing to do some testing to refute the claims myself but to make it worthwhile, I need the right motherboard with the same heatpipe setup (I have a P5K-E, which use a different heatpipe setup), same waterblock and a ACCURATE thermometer (IR ones are good) along with a good before/after.

annaconda didn't provide the right testing parameters and didn't stay constant doing a comparaison test. Even his inability to prove it make this invalid.

I still stand by my own conclusion that the only positive effect is the better mounting of the heatpipe assembly, not the waterblock itself. The nuts and bolts always provide more pressure on the die than plastic holders with weak springs. Do you trust installing a cpu heatsink using just plastic holders and springs ?

 
The [H] community encourages new and exciting things...Perhaps this community isnt right for you.

I don't think that's a fair thing to say. Encouraging new ideas without any true proof of their validity is simply bad science.

I think he's got the right idea, but is going about it the wrong way.

When I get my server up and running, I might as well do a simple test (no water in the waterblock, just an evaluation of increased contact).
 
I don't understand why everybody jumps up to defend this setup. There hasn't been a shred of credible evidence put forth that would indicated this actually works. Basic knowledge of thermodynamics supports the fact that this is a terrible idea. Additionally, the base of a perfectly good waterblock is now marred and would take a good amount of lapping to get it back into a usable state.

I can agree that the [H] community supports new and exciting ideas, however I do not think it promotes bad ideas. If this were presented with credible testing methods then I'm sure we'd all be supporting it. Until then...

Droc, asking someone else to buy an identical setup and go test it on their own, just to prove that common knowledge applied to this scenario is more accurate than meat thermometer readings from different areas of a heatsink, is pretty retarded.

rhagz, because his system boots into Windows at 4.1ghz doesn't mean this is a good solution. Check out the stability. I'm not trying to be a dick here, but please think for yourself rather than assume.
 
Irhagz, because his system boots into Windows at 4.1ghz doesn't mean this is a good solution. Check out the stability. I'm not trying to be a dick here, but please think for yourself rather than assume.

My apologies for not stalking him across the internet at every possible hardware forum. :rolleyes:

Still don't see why people care so much. If it works for him, great. It's not like he is trying to sell you something.
 
Droc, asking someone else to buy an identical setup and go test it on their own, just to prove that common knowledge applied to this scenario is more accurate than meat thermometer readings from different areas of a heatsink, is pretty retarded.

No one cares if it's the same motherboard: the concept is pretty generic, it can be tried on plenty of Asus boards. So your point is moot. What we want to see is performance of any wb on any heat-pipe+fins.

If you have a problem with the guy's testing go frakin do it yourself. He is NOT proclaiming it a law of thermodynamics. He was frakin testing it for fun.

Nice party pooping :rolleyes: :mad:
 
My apologies for not stalking him across the internet at every possible hardware forum. :rolleyes:

Still don't see why people care so much. If it works for him, great. It's not like he is trying to sell you something.

If you had read the whole thread you would have seen that I had posted the same link on page 3 (post 57).

No one cares if it's the same motherboard: the concept is pretty generic, it can be tried on plenty of Asus boards. So your point is moot. What we want to see is performance of any wb on any heat-pipe+fins.

If you have a problem with the guy's testing go frakin do it yourself. He is NOT proclaiming it a law of thermodynamics. He was frakin testing it for fun.

Nice party pooping :rolleyes: :mad:

I have no desire to ruin the base of my MCW30 and I doubt anybody else does. Want to ship someone a $30 block? While nobody did say "it's the law of thermodynamics", promoting a poor coolling solution is not something anybody should want to see here.
 
Fujimitsu, if your not going to reproduce the situation and prove him wrong yourself, then shut up. Its simple dude, just stop typing. .

For me to prove it wrong, there would have to be some reason for it to work at all, thus necessitating further testing. However, there has been absolutely no reason whatsoever presented. If i told you that coral supplements cured cancer, would you really have to kill someone to prove me wrong? (look at me, I'm kevin trudeau) No, you wouldn't because I gave you no reason whatsoever to even entertain the idea in the first place.

And Arcygenical you're right, he didn't claim that it was his northbridge temp, he was just showing a delta. But the testing is still totally worthless.

As far as his overclock goes, it's worthless unless its stable. And, it was achieved before this mod, so what bearing does it have on this thread at all?

And like a few people have said.. there are interesting ideas and there are totally pointless misguided ideas. I don't want to see any of the poor saps in this thread who have said "great idea" and the like, go out and ruin their $30 blocks for no reason. I'm not attacking anyone, I'm simply asking that he provide some kind of proof that this works before making claims that it's awesome.
 
Man up and spend a dollar or two and buy a 1/16 inch thick square of polished copper and place it between the waterblock and the heatpipe for protection then prove him wrong.

You'd think that with all your supreme intelligence that you could figure out how protect the surface of a waterblock for testing....heck, the fact that you seem to think that gently setting a waterblock on the heatpipe is going to destroy it just goes to show how pathetic your excuse to try it yourself have become.

Grow a pair and show us some number yourself. Until then, go stroke your e-penis somewhere else. I look forward to seeing your results.
 
Man up and spend a dollar or two and buy a 1/16 inch thick square of polished copper and place it between the waterblock and the heatpipe for protection then prove him wrong.

You'd think that with all your supreme intelligence that you could figure out how protect the surface of a waterblock for testing....heck, the fact that you seem to think that gently setting a waterblock on the heatpipe is going to destroy it just goes to show how pathetic your excuse to try it yourself have become.

Grow a pair and show us some number yourself. Until then, go stroke your e-penis somewhere else. I look forward to seeing your results.

Swiftech's blocks are beautifully lapped, and if applied with proper mounting pressure to the fins of a heatsink... they will be gored. By the way heatpipe and heatsink are two different things, not sure if you knew that or not but you use them interchangably so I wasnt sure.

Using a sheet of copper would require me to use TIM, and hurt the performance of this already poor idea even more.

Like I've said, it doesn't require testing to be proven wrong.

I'm just gonna stop responding to you know, given the fact that you have no actual information to contribute.
 
Man up and spend a dollar or two and buy a 1/16 inch thick square of polished copper and place it between the waterblock and the heatpipe for protection then prove him wrong.

That would introduce another variable and violate the scientific method.

Are you seriously expecting someone to go buy a motherboard and destroy a waterblock just to prove the laws of thermodynamics all over again? His setup is flawed and is a complete waste of time. Let's just reiterate what we all ready know. First, the waterblock makes very little contact with the northbridge heatsink and probably does a better job of insulating it than cooling it. Second, he used a meat thermometer to gather data and introduced the variable of thermometer placement. His "experiment" does not follow the scientific method and therefore it's no one's job to try and disprove his experiment. He needs to conduct a proper experiment before anyone can be expected to try and recreate his mod. Thank god most of you aren't in the science field...

How about you just "man up" and admit that you're wrong?
 
The reason everyone is attacking Annaconda is because he is spreading false info. I for one think his idea is horrible, and the way he went about testing and proving whether his experiment or not is worse still.

Bottom line: It was a bad idea, and the way he defended it was bad. There is no way anyone can defend him when his results are biased and wrong like that.
 
If you do not agree with the OP or his system that is completely fine, please by all means debate it.

However, we have rules here against thread crapping, flaming and name calling. I came in here yesterday and said, if you can't follow the rules then don't post.

I'm going to state this one more time. If you cannot follow our forum rules with your responses, then please don't post.

If I have to come in here again to babysit a bunch of childlike behavior we will start handing out vacations.
 
That would introduce another variable and violate the scientific method.

Are you seriously expecting someone to go buy a motherboard and destroy a waterblock just to prove the laws of thermodynamics all over again? His setup is flawed and is a complete waste of time. Let's just reiterate what we all ready know. First, the waterblock makes very little contact with the northbridge heatsink and probably does a better job of insulating it than cooling it. Second, he used a meat thermometer to gather data and introduced the variable of thermometer placement. His "experiment" does not follow the scientific method and therefore it's no one's job to try and disprove his experiment. He needs to conduct a proper experiment before anyone can be expected to try and recreate his mod. Thank god most of you aren't in the science field...

How about you just "man up" and admit that you're wrong?

How am I wrong? Its not my topic.

Perhaps your in the wrong thread, because I dont see annaconda claim anywhere that he is performing a scientific experiment?
What you have is a guy who tried something different and tested it with the tools he has on hand.
It may not fallow your scientific method, but this is not NASA.

And Im not telling anyone to go out and buy a motherboard. Many of us, including Fujimitsu have motherboards with heatpipes.

And how is it going to insulate it? Its a water cooled system anyways. There is little to no airflow as it is. Look at the heatsink. Its folded over at the top with a pipe down the middle. Adding a waterblock isnt going to obstruct much....considering that airflow is almost nonexistent, putting a cooler object on a hot object is going to transfer some heat.
I can see your point about insulating it is it was blocking airflow from a fan, but its not.
These heatsinks get bloody hot....stupid hot. If airflow isnt pulling the heat off, then I can see how a waterblock, even with low contact area, could.

I have a similar meat thermometer and its accurate within 4 degrees C at 500 degrees.
I thought, ah heck, I have heatpipes and a similar NB setup to his, Ill take my meat thermometer and check different spots on my NB heatsink under load. You know what? Im getting 1-2 degrees difference.
Ok, so his placement was off. I move the placement around and no matter what I do, I cannot get 30 degrees difference.


I agree that adding a copper spacer will reduce the effectiveness and introduce a new variable. But for the sake of argument, it would allow someone to place a waterblock on the heatsink without risking damage to the waterblock. If it showed any reduction at all in temperature over running it uncovered with zero airflow, then he is justified in his idea.

As for a waste of time, that depends on how people value their time. You have people in here bashing the idea over the period of 20 days, starting threads to bash this one on other forums...Yet if they wanted to physically prove him wrong, it would take what? an hour of time?
Unless you have the tools to test it yourself, then don't criticize someone elses method.
 
First, the waterblock makes very little contact with the northbridge heatsink and probably does a better job of insulating it than cooling it.

"Probably". You don't know how it performs and you are speculating. You can't prove that it performs bad. It doesn't bother me that people provide alternate views, what bothers me is the negative insulting attitude.

You have no proof that it performs bad, so don't claim that it does.

To Everyone:
If you want to be assy about him not making claims, then NO ONE in this thread is entitled to until they make the testing themselves. Until then keep your views respectful.

Theory can easily be wrong because of overlooked and miscalculated variables.
 
And how is it going to insulate it? Its a water cooled system anyways. There is little to no airflow as it is. Look at the heatsink. Its folded over at the top with a pipe down the middle. Adding a waterblock isnt going to obstruct much....considering that airflow is almost nonexistent, putting a cooler object on a hot object is going to transfer some heat.
I can see your point about insulating it is it was blocking airflow from a fan, but its not.
These heatsinks get bloody hot....stupid hot. If airflow isnt pulling the heat off, then I can see how a waterblock, even with low contact area, could.

I agree. I'm not a fan of covered heatsinks but my Asus P5W DH Deluxe has this heatsink:

image03ps4.jpg


So it seems that heat dissipation will happen sideways and the covered top won't make much of a difference. So this means that having a MCW30 on top of a heatsink should not insulate anything; heat dissipation will happen sideways.

That is unless our resident "scientists" know better than Asus? :rolleyes:
 
It's also been mentioned a couple of times now... why would the MCW30 get "ruined" by placing it on top of the heatsink? My guess is that it might get somewhat scratched but it could easily be retouched by polishing, but "ruined"? Can someone explain why this would happen? Is there any proof or examples available? If anyone can develop this I'd appreciate it.
 
It's also been mentioned a couple of times now... why would the MCW30 get "ruined" by placing it on top of the heatsink? My guess is that it might get somewhat scratched but it could easily be retouched by polishing, but "ruined"? Can someone explain why this would happen? Is there any proof or examples available? If anyone can develop this I'd appreciate it.

The fins would scratch the surface, therefore requiring it to be lapped (not polished). Swiftech blocks come machine lapped and any lapping job you can do by hand would be inferior to the state it was in when you took it out of the box. Please don't ask me to prove that metal can scratch metal...
 
The fins would scratch the surface, therefore requiring it to be lapped (not polished). Swiftech blocks come machine lapped and any lapping job you can do by hand would be inferior to the state it was in when you took it out of the box. Please don't ask me to prove that metal can scratch metal...

I donno, some people would argue that the difference between an out of the box lapping and a hand done lapping is little to none, and the overall performance difference is practically impossible to determine.
 
The fins would scratch the surface, therefore requiring it to be lapped (not polished). Swiftech blocks come machine lapped and any lapping job you can do by hand would be inferior to the state it was in when you took it out of the box. Please don't ask me to prove that metal can scratch metal...

Ok... well then, we'd need to see what kind of damage if any would occur. I was hoping for examples. From what I can see the block is by no means "ruined". I think this matter should be kept open until we have more data.
 
Regardless of whatever you guys want to think about quality of lapping, it creates extra work if you ever want to use the block for it's intended purpose. As far as examples go, I'm sure you can find some with google. I, for one, am not going to purposely damage a good block (such as my Storm sitting on my shelf) just to prove that metal can in fact scratch/damage other metal objects.
 
I have a question... why not just take the stock cooling off and using the NB water block like it's intended for?

And I don't know about you, or whether you have ever owned a Swiftech water block, but the stock lapping is pretty damn good. You'd be a fool to mess it up on purpose.
 
I have a question... why not just take the stock cooling off and using the NB water block like it's intended for?

And I don't know about you, or whether you have ever owned a Swiftech water block, but the stock lapping is pretty damn good. You'd be a fool to mess it up on purpose.

He already stated why he didn't cool the NB directly. please read his post.
 
Yeah, but then he said that it would be a good idea to cool the SB the same way... which totally contradicts his first reason.

I'm afraid I see no indication of that. what I do see him saying is that he is using the nature of how the heatpipe works in conjunction with the WC
 
I have a question... why not just take the stock cooling off and using the NB water block like it's intended for?

Because he'd have to cool all 3 chips with waterblocks. I think the idea is to avoid the hassle and $$$ and have 1 wb cool everything through heat pipes.

I do agree that adding a second wb on top of the sb heatsink would defeat the purpose. I think the challenge is to successfully have 1 wb cool all 3 chips.
 
Zip tie a 80mm-120mm fan over the NB heatsink to cool all three. Costs $10 and takes 5 minutes. /end thread.
 
Alright guys, i will conduct another test tommorrow.

This time it will be same contact point, and also will compare with air (means small fan). I will also post some pics of the bottom of the water block, and take temperature readings from all three parts (NB + SB + power module heatsink) with or without waterblock.

Any other thing please let me know i will try to see if i can do it. Oh by the way, i am not going to buy any other Thermometer. That Meat thermometer will do it :D
 
Alright guys, i will conduct another test tommorrow.

This time it will be same contact point, and also will compare with air (means small fan). I will also post some pics of the bottom of the water block, and take temperature readings from all three parts (NB + SB + power module heatsink) with or without waterblock.

Nice :)
 
Alright guys, i will conduct another test tommorrow.

This time it will be same contact point, and also will compare with air (means small fan). I will also post some pics of the bottom of the water block, and take temperature readings from all three parts (NB + SB + power module heatsink) with or without waterblock.

Any other thing please let me know i will try to see if i can do it. Oh by the way, i am not going to buy any other Thermometer. That Meat thermometer will do it :D

Just to warn you, you're not gonna change anyone's minds with a $5 meat thermometer.

No one's making you change your setup, it's your choice if you want to spend the $ on a thermometer to really test this. If you're happy with it, leave it and let the thread die. But just know that you're probably just gonna be bashed more if you post meat thermometer pics again.

Glad to see you attempting to prove this works, despite the criticism. It's at least respectable.
 
Back
Top