Alphacool CORE 1 CPU block - bulging with danger of splitting?

[A Computer Guy]

I redid the measurement with larger rad and temperature monitoring (measuring water temperature in the rad output feeding the block), this time with 200W heat load. ( I do not want to degrade my CPU with 253W anymore... )

I thought 253W was still within Intel spec?

 

[reggie_fils_aime]

Who will win: an entire water cooling engineering department OR one guy with some ink in his apartment????

 

[BoggledBeagle]

I thought 253W was still within Intel spec?

With the larger rad and lower power draw, the temps were 15°C lower than before, so the CPU surely suffered less during the testing. Almost half an hour of 100% load and 200W of heat flowing from that small chip is not a negligible stress either. I wish I had a heatplate instead...

Who will win: an entire water cooling engineering department OR one guy with some ink in his apartment????

I won, I could reach the same result without deforming the block.

 

[Noci]

This thread is a beautiful demonstration of the Dunning-Kruger effect in action.

I couldn't agree more, to be honest with a major dose of tunnel vision as well.

 

[FoulOnWhite]

Buy block, fit it on CPU, run PC.......end

 

[A Computer Guy]

Buy block, fit it on CPU, run PC.......end

You forgot the paste!

 

[Carillon]

You forgot the paste!

And the loop!

 

[Steevo]

With the larger rad and lower power draw, the temps were 15°C lower than before, so the CPU surely suffered less during the testing. Almost half an hour of 100% load and 200W of heat flowing from that small chip is not a negligible stress either. I wish I had a heatplate instead...


I won, I could reach the same result without deforming the block.

I think your willingness to try is great. Its where a lot of us old farts started, don't take the posts as mean spirited, its just how some of us are that have been around for awhile, we all tried things too, lapped CPUs, coolers, multiple remounts and shimming to get better contact, swapping out paste, fans, coolant, to get better results. Sometimes we dug too deep, or occasionally not deep enough.

 

[Vya Domus]

I won, I could reach the same result without deforming the block.

Another way of thinking about it is that you did a bunch of extra work that resulted in no net positive.

 

[BoggledBeagle]

Another way of thinking about it is that you did a bunch of extra work that resulted in no net positive.

What do you mean? I de-pressurised the block and removed the danger of splitting (if it ever existed - now it does not exist for sure).

 

[Vya Domus]

What do you mean? I de-pressurised the block and removed the danger of splitting (if it ever existed - now it does not exist for sure).

If you think this solved your irrational fear of the block bursting OK I guess.

 

[tabascosauz]

I won, I could reach the same result without deforming the block.

Did you just answer a rhetorical meme question.....?

I guess that's one way of looking at it, but damn impressive mental gymnastics required to reach that conclusion. Is it this difficult to consider that your premise and base assumptions could have been flawed? All I see are margin of error results that can be wholly and easily attributed to paste application and block mounting.

Maybe the whole ordeal was a testament to how resistant the block design is against being tampered with for no good reason.

 

[BoggledBeagle]

If you think this solved your irrational fear of the block bursting OK I guess.

When you look at the second photo in the first post, these creases are in the real life very apparent, in the photo it may not look as dramatic as in real life.



Have you noticed that machining boo-boo - circled? What does it tell you?

Frankly I have no idea where you guys get all your confidence...

 

[tabascosauz]

Have you noticed that machining boo-boo - circled? What does it tell you?

That the Alphacool block is an inexpensive block with correspondingly inexpensive manufacturing?

Frankly I have no idea where you guys get all your confidence...

 

[BoggledBeagle]

That the Alphacool block is an inexpensive block with correspondingly inexpensive manufacturing?

Inexpensive manufacturing you say? And you would with such manufacturing confidently make such very deep cuts in the base, not fearing a bit that some of these cuts may end up a little bit TOO DEEP by accident? Similarly as the milling bit got a little lower than it was supposed to?

 

[Vya Domus]

not fearing a bit that some of these cuts may end up a little bit TOO DEEP by accident

These aren't done by hand my man, they're CNC'ed, there's not gonna be an accident.

 

[Steevo]

We get it from years of building gaming PC's, workstations, dumb boxes, servers. Fucking things up like when I filled my loop and didn't get the plug back in and it killed a SSD. Like when I was hot, tired, and pissed about old IBM towers and jammed RAM in backwards and killed a whole old tower, when a member here left their double peltier stack on and their water cooler off, when sparks fly out of a PSU in a server on a Monday, getting a PSU repair by the OEM when its been heavily modified and disassembled at least twice, cracking bare dies, overvolting hardware to death.

And yet we are still here and mostly our data is safe.

 

[A Computer Guy]

When you look at the second photo in the first post, these creases are in the real life very apparent, in the photo it may not look as dramatic as in real life.

View attachment 345906

Have you noticed that machining boo-boo - circled? What does it tell you?

That looks like a machining error on the top, not the cold plate. Is this a trick question?

Frankly I have no idea where you guys get all your confidence...

Perhaps the only way to dispel your concern for the block breaking is to try to break the block and see how tough it really is. You know, for science!

 

[FoulOnWhite]

What do you mean? I de-pressurised the block and removed the danger of splitting (if it ever existed - now it does not exist for sure).

There is no way that block would split. As i have said, i had the same block, with a gen 1 black rubber insert. I used it with a contact frame, with no problems and without it splitting.

Take a 2mm thick piece of copper, and bend it, does it split? no it would not, it would bend. There is not enough water pressure or pressure from the rubber sheet pressing on the inside to make it split. How much metal do you think is left where the grooves are machined? fractions of a millimeter, i don't think so. Even if it was .5mm i still doubt very much it would split.

If i had the spare cash, i would buy another and destroy it just to proove my point.

No offense but you prooved nothing, and just had a bit of an adventure with your new block.

 

[thesmokingman]

Hey Guys, look at me! /thread

 

[A Computer Guy]

With all due respect to the OP and contributors I'm out and good luck.

 

[Assimilator]

Perhaps OP should try hitting the block against a harder surface to see if it will indeed split. I recommend OP's forehead.

 

[reggie_fils_aime]

When you look at the second photo in the first post, these creases are in the real life very apparent, in the photo it may not look as dramatic as in real life.

View attachment 345906

Have you noticed that machining boo-boo - circled? What does it tell you?

Frankly I have no idea where you guys get all your confidence...

I have no idea why you'd think several millimeters of metal is going to split like an old tire with a grand total of half a cup of water sloshing around inside of it?

 

[BoggledBeagle]

I have no idea why you'd think several millimeters of metal is going to split ...

I am sorry I was not able to measure the thickness of the materiel left under the water channel cut, I bet it is less than half millimeter thick, and it is compromised metal, you are OPENING A CUT, you are creating a desired shape of the base by DEFORMING and STRESSING the metal.

I would like to measure the thickness of the remaining metal somehow...

 

[FoulOnWhite]

I am sorry I was not able to measure the thickness of the materiel left under the water channel cut, I bet it is less than half millimeter thick, and it is compromised metal, you are OPENING A CUT, you are creating a desired shape of the base by DEFORMING and STRESSING the metal.

I would like to measure the thickness of the remaining metal somehow...

Machining a groove into a piece of metal is not deforming it. Bending to a shape is as it upsets the structure imo. Imo I don't think there is any weakness caused by machining/CNC.