Project Gallium (Liquid Metal Cooled PC)

[byalexandr]

Just now, Tmt97 said:

Not to the PC it isn't. Besides, that's never stopped anyone in the food industry. All jokes aside though, your gonna be pumping it through closed pipes, not ingesting it. As long as they are careful, and wear gloves it should be fine.

That's true. Also keep in mind though, it's extremely dense so it would definitely cause some stress on the tubing, blocks, and pumps. Not to mention how high the surface tension is.

[Tmt97]

11 minutes ago, byalexandr said:

That's true. Also keep in mind though, it's extremely dense so it would definitely cause some stress on the tubing, blocks, and pumps. Not to mention how high the surface tension is.

We, you can build a supporting structure for the tubing and the blocks. And before you say that it would be too hard, keep in mind that meeting challenges like that is just part of innovation. If you don't want to meet the unique challenges of implementing a new form of PC cooling, then you should stick with water. Not sure how the surface tension will affect it though...

[byalexandr]

Just now, Tmt97 said:

We, you can build a supporting structure for the tubing and the blocks. And before you say that it would be too hard, keep in mind that meeting challenges like that is just part of innovation. If you don't want to meet the unique challenges of implementing a new form of PC cooling, then you should stick with water. Not sure how the surface tension will affect it though...

I wasn't going to say it would be too hard.

 

The pump will struggle though, and I'm wondering about the micro fins inside the blocks as well...

[Tmt97]

11 minutes ago, byalexandr said:

I wasn't going to say it would be too hard.

 

The pump will struggle though, and I'm wondering about the micro fins inside the blocks as well...

Dunno what would happen to the micro fins really... Your absolutely right about the pump, although it will be true to some degree about the gallium, indium, tin mix too. Gonna have to find a pretty beefy pump no matter what.

[pagani123]

1 hour ago, Tmt97 said:

Well who am I to discourage someone to do something just to say you did it. I showed up to the new star wars movie in a spock costume just to say I did it. Just a suggestion(I don't know specifically what problems would come from this), but what if you used mercury? I think it has a similar problem with aluminum that gallium does, but it stays liquid at least.

i dont have access to a fume hood and i want to bring this to events there is no way they would let me in a lan party with a bunch of mercury. besides they dont really allow mercury in private hands on campus. also i calculated the cost and it would come out more expensive to use mercury anyways. trust me when i say i have thought through almost all possible options  

[vanished]

2 hours ago, Tmt97 said:

Dunno what would happen to the micro fins really... Your absolutely right about the pump, although it will be true to some degree about the gallium, indium, tin mix too. Gonna have to find a pretty beefy pump no matter what.

Talk about gallium causing issues for aluminium - I've heard mercury will also eat it, and probably way worse.

[Stefan1024]

@Maybach123

Why don't you use a heat exchanger like this and bolt a big heat sink onto it?

 

 

The tubing is fully copper and the added heat resistance is very low. If you like I can help you selecting a heat sink. I estimate the cost for the heat exchanger and the heat sink to be ~200$. It will have less points of failure than the coated aluminium radiator and probably be even cheaper.

 

As for the temperatures, I consider everything <90°C a success as this a very experimental stuff with a lot of variables.

[Blu]

12 hours ago, Tmt97 said:

Do you, by chance, know what coolant was used there? That's exactly what I've been looking for, for my build  

[z3kron]

3 minutes ago, Blu said:

Do you, by chance, know what coolant was used there? That's exactly what I've been looking for, for my build  

That's nickel plated copper tubing.

[Blu]

Just now, z3kron said:

That's nickel plated copper tubing.

OH! No wonder I found nothing...I was looking in the completely wrong direction! Haha! Do you know if there are any coolant fluids that give the same effect?

 

[z3kron]

12 minutes ago, Blu said:

OH! No wonder I found nothing...I was looking in the completely wrong direction! Haha! Do you know if there are any coolant fluids that give the same effect?

 

I doubt that any like this exsist.

Though if you want a familiar look, with good lighting Mayhems "Aurora" fluid may look very metaliky.

[Tmt97]

10 hours ago, Maybach123 said:

i dont have access to a fume hood and i want to bring this to events there is no way they would let me in a lan party with a bunch of mercury. besides they dont really allow mercury in private hands on campus. also i calculated the cost and it would come out more expensive to use mercury anyways. trust me when i say i have thought through almost all possible options  

fair point

[pagani123]

8 hours ago, Stefan1024 said:

@Maybach123

Why don't you use a heat exchanger like this and bolt a big heat sink onto it?

 

The tubing is fully copper and the added heat resistance is very low. If you like I can help you selecting a heat sink. I estimate the cost for the heat exchanger and the heat sink to be ~200$. It will have less points of failure than the coated aluminium radiator and probably be even cheaper.

 

As for the temperatures, I consider everything <90°C a success as this a very experimental stuff with a lot of variables.

we were actually thinking about using this just a bit bigger if we can find it

[iamdarkyoshi]

3 minutes ago, Maybach123 said:

we were actually thinking about using this just a bit bigger if we can find it

Try looking for a window air conditioner condenser coil

[pagani123]

9 hours ago, Ryan_Vickers said:

Talk about gallium causing issues for aluminium - I've heard mercury will also eat it, and probably way worse.

mercury is way worse at least with gallium i could keep it in there for around 6 hours before it really started causing issues. this is a 2 hour timelaps of mercury and aluminum

 

 

[vanished]

2 hours ago, Maybach123 said:

mercury is way worse at least with gallium i could keep it in there for around 6 hours before it really started causing issues. this is a 2 hour timelaps of mercury and aluminum

 

 

Yeah so the mercury idea is definitely out   I thought that would happen...  

[Dash Lambda]

Seems like the microfins on the water blocks would be ill-suited for a liquid metal. Is gallium's surface tension comparable to Mercury's?

 

@Tmt97, I disagree on your assessment of the aesthetics. Liquid metal would look perfectly reflective and stationary, even in the blocks. Imagine how cool it would look inside of clear plexi block...

[pagani123]

Just now, Dash Lambda said:

Seems like the microfins on the water blocks would be ill-suited for a liquid metal. Is gallium's surface tension comparable to Mercury's?

 

@Tmt97, I disagree on your assessment of the aesthetics. Liquid metal would look perfectly reflective and stationary, even in the blocks. Imagine how cool it would look inside of clear plexi block...

gallium is worse when it comes to surface tension it is extremely sticky but according to the company were working with it wont cause an issue with the deposition of parylene

[Tmt97]

1 minute ago, Dash Lambda said:

Seems like the microfins on the water blocks would be ill-suited for a liquid metal. Is gallium's surface tension comparable to Mercury's?

 

@Tmt97, I disagree on your assessment of the aesthetics. Liquid metal would look perfectly reflective and stationary, even in the blocks. Imagine how cool it would look inside of clear plexi block...

You basically restated my point. you wont see flow at all so it will look perfectly stationary and reflective even in the blocks. Liquid metal wont look any different than solid metal even inside the blocks. It will basically look like this, but with a layer of plexi:

Personally I don't find metal coated in plastic to be any more interesting than just solid metal, and we've already seen builds with solid metal... so yea. XP

 

And to those saying how fast mercury will eat aluminum, heres what I have to say "Duh..." Not to be rude but why the fuck would you pay all this money for a custom loop with mercury OR gallium, if you plan to have aluminum in the loop? Sure, maybe gallium will last longer but your still wasting a ton of hardware. Just pick parts that aren't made of aluminum, or if you instead decide to use that layer of a "silicone-like material" coating the inside of the loop like you said, then it won't matter even if you use mercury will it? Because if the mercury doesn't have contact with the aluminum, then what does it matter how the two materials interact?

[pagani123]

19 minutes ago, Tmt97 said:

You basically restated my point. you wont see flow at all so it will look perfectly stationary and reflective even in the blocks. Liquid metal wont look any different than solid metal even inside the blocks. It will basically look like this, but with a layer of plexi:

Personally I don't find metal coated in plastic to be any more interesting than just solid metal, and we've already seen builds with solid metal... so yea. XP

 

And to those saying how fast mercury will eat aluminum, heres what I have to say "Duh..." Not to be rude but why the fuck would you pay all this money for a custom loop with mercury OR gallium, if you plan to have aluminum in the loop? Sure, maybe gallium will last longer but your still wasting a ton of hardware. Just pick parts that aren't made of aluminum, or if you instead decide to use that layer of a "silicone-like material" coating the inside of the loop like you said, then it won't matter even if you use mercury will it? Because if the mercury doesn't have contact with the aluminum, then what does it matter how the two materials interact?

i dont plan to have aluminum in the loop i was just replying to a comment about it using a hypothetical situation. also it wont be a a complete loss as i can still sell the gallium for scrap value (to my college) after i am done with it.  

[Dash Lambda]

2 hours ago, Tmt97 said:

-snip-

You're not getting what I'm saying. Imagine this with a reflective metal instead of green...

 

It wouldn't just look like a solid metal block covered in plexi, it would look almost like layered traces.

[Tmt97]

19 minutes ago, Dash Lambda said:

You're not getting what I'm saying. Imagine this with a reflective metal instead of green...

 

It wouldn't just look like a solid metal block covered in plexi, it would look almost like layered traces.

No, I AM getting it. Look where the green is in that block. Most of it is just a couple large channel, which will look like blocks of metal, covered in... you guessed it! Plexi! The only spots that look like "thin traces" are those little lines(I think they are called micro-channels?) but just that its thin won't make it look any different than any other thin piece of metal like a wire or the fins on a heatsink. Your not gonna see the metal move as it flows through those channels if that's what you think. It won't look like the aurora coolant or anything(although Ill admit some metallic silver aurora coolant would look pretty sick). Gallium looks no different flowing than when stationary. So again, were still just looking at weirdly shaped metal block, with some fin/wire things all covered in plexi... I dunno what else you expect. XP

[Dash Lambda]

5 minutes ago, Tmt97 said:

No, I AM getting it. Look where the green is in that block. Most of it is just a couple large channel, which will look like blocks of metal, covered in... you guessed it! Plexi! The only spots that look like "thin traces" are those little lines(I think they are called micro-channels?) but just that its thin won't make it look any different than any other thin piece of metal like a wire or the fins on a heatsink. Your not gonna see the metal move as it flows through those channels if that's what you think. It won't look like the aurora coolant or anything(although Ill admit some metallic silver aurora coolant would look pretty sick). Gallium looks no different flowing than when stationary. So again, were still just looking at weirdly shaped metal block, with some fin/wire things all covered in plexi... I dunno what else you expect. XP

I guess we just have different tastes. I just like the idea of those channels looking like a too-perfect-to-be-quite-real metal surface, while still actually being a flowing liquid.

Personally, I'm not a fan of fluids you can see moving. I think it's cool to have a dynamic system look static.

[Tmt97]

32 minutes ago, Dash Lambda said:

I guess we just have different tastes. I just like the idea of those channels looking like a too-perfect-to-be-quite-real metal surface, while still actually being a flowing liquid.

Personally, I'm not a fan of fluids you can see moving. I think it's cool to have a dynamic system look static.

Yea, I think it is different tastes. Although I don't know why you have this notion that it will look "too-perfect-to-be-quite-real". Gallium looks exactly the same as any other metal, not any more perfect, not any less. That's not an opinion, that's not an exaggeration, its just a fact. Google gallium, and then silver or some other metal. Its exactly the same XD. That's fine if you like a static look. I actually do as well. That's why when I make a custom loop, I'm gonna use solid copper tubing. All I am saying is, lets not kid ourselves here: the aesthetics of what he is describing is not gonna look any different from a build with solid metal tubing like this:

Except it'll be covered with plexi and maybe have a few of those little channels(if he even buys a block with them visible) which likely won't even be seen from most angles since they are on the video card. The only difference between a metal tubing build, and this, is that the way he is planning on doing it will cost a fortune, and won't cool worth shit... I am however still holding out hope that he's gonna just choose his parts carefully, and do a "legit" gallium cooled PC. Now that will truly be a marvel of engineering!

[Dash Lambda]

3 minutes ago, Tmt97 said:

-snip-

What I mean by "too-perfect-to-be-real" is the essentially featureless reflective surface liquid metals have. You can get normal metal to look like that, sure, but it's something you have to maintain and it wouldn't apply to the channels inside the water blocks.

 

Of course, I'm assuming they're going to use flexible tubing. This is one of those situations where I think hardline would actually look worse than flexible, because then it actually would be hard to tell that it's liquid metal and not just metal piping...