I got my ifixit kit from my wife as an early father's day present but there were no explosions when I opened it. I need my own Phil 😭😭😭😂

CPU and mobo be like: WTF is this psycho gonna do to us today 😂

Your setup looks like the early stages of a moonshine stil...

Never give up no matter how much water is leaking out... that’s what I’ve learnt 😂😂

The amount of tries he managed to finish after he said "we're on borrowed time" is actually amazing.

The water flow on the outside of the copper coil is relatively important, since as the heat transfers out of the loop and into the bucket pool, what would happen is it would make 'hot pockets' around the copper heat pipe, in turn, over time the loop would warm up faster. Since you dropped dry ice just by itself into the bucket, that creates flow due to its energy loss and co2 emission. The only issue with that style is your dry ice usage is potentially inefficient and uncontrolled. When the dry ice was in an inner bucket, the surface area was limited, limited to only the walls the dry ice touches, and the air that slowly transfers energy to the dry ice; this allows the for the dry ice to continue to cool the loop, but not over cool, whilst dropping dry ice into a liquid causes the surface area be of the full dry ice block, and the dry ice will quickly be wasted, with the efficiency dropping, as the amount of energy that needs to come to the bucket exponentially increases, which makes other sources, like the room, your hands, etc, quickly transfer heat and waste the volume that could be used on the loop. Basically, there's a stopping point for when you're wasting more dry ice then actually converting into usability for the loop.

As an 'alcohol enthusiast', in my comment on the last video, when I said ethanol, I wasn't referring to vodka. I was referring to 90%+ denatured or isopropyl, such as can be found at the pharmacy (pandemic aside) or in the soldering section of Micro Center. And as for throwing away anything it touches, your blocks and pumps might need new o-rings, but otherwise they should be fine. Moving to a liquid that doesn't freeze is the only way to get into those super low temps. Also, ethanol has a higher thermal transmissivity than glycol, so not only will it go colder, but it will be more efficient in getting there.

If you did want to try ethanol as a coolant liquid, it's freezing point is *-114C*, so you're not going to need to worry about freezing. 200 Proof (100%) Lab grade alcohol can be bought in bottles online relatively inexpensively. If you want cheaper, or non-food grade alcohol, you can get Methanol freezes at -97C, Propanol is -126C, Isopropyl is -89C. Any of those would be perfectly fine, won't culture any bacteria or algae, and I don't think they'll react with the metal, but google that before you commit to a flavor of coolant. EDIT: if you can't get 100%, then slowly lower the temp of your water containing liquid, and screen the ice crystals from it. I think that will work, but I haven't tried it.

I love seeing Jay trying to actually do it correctly 😂😂😂

I’m no guy with a phd but if you were to create a vacuum around the water block the condensation would not build up as the cold can not reach the warm air

The two loops are a good idea (disclaimer, my experience comes from brewing beer professionally). Using a cooling coil is a standard approach in small batch brewing or homebrewing but the problem (in this scenario in brewing) that people often make is that they are not moving both liquids (cooling liquid and hot beer). You end up cooling at a much slower rate as the temperature localizes around the coil and then is less efficient in transferring through the rest of the liquid in the vessel. By whirlpooling the liquid (running it from top to bottom causing it to spin in the vessel) like you were doing in the bucket, you significantly increase cooling efficiency.

You guys are tech comedians. Tech wizards of goof. Thank you Bois. And the ifixit commercials are hilarious to wait around for!

I discovered this channel because I've grown more and more interested in tech (lead to me now working in a tech repair shop chain) and yet fate reminds me of my first job working at Harkins theatres by showing me it's "awards" card being used to apply thermal compound...thank you

Really enjoyed watching your experiments with cooling the CPU. Many have suggested using alcohol. Vodka is no use as it contains water. You can try pure ethanol as it is not poisonous. Its boiling point is 78.37C or 173F and its freezing point is -114.1C or -173.38F. Need to ensure that the alcohol does not reach its boiling point. And it should not freeze and block the cooling pipes. Waiting for your next video.

I'm not a science license holder person, but I am a homebrew guy. I recommend two things if you ever play with this stuff again. 1) counterflow chiller you can get these at most brew supply stores, or rig you own with a garden hose and copper tubing, should let you use a regular garden water pump to circulate your ice side cooling loop. 2) An ethanol and water solution will get you a better freeze point to viscosity balance, and I think ethanol also conducts heat better than any of the glycols.

By adding salt you could go -4c -5c with this setup. (needs a lot! ~200 grams+)

Linus: watercooling is impractical

Oh jesus, that copper coil threw me back into making beer with my brother in our backyard. We had a copper coil probably twice that length for the 12 gal pot we used.

My favorite bourbon. Awesome story and distillery tour.

Glad you were able to get a better result on your scores Jay. (B.S. in mechanical engineering) Not sure if anyone else has brought this up, but the fin density of the block is not really designed for more viscous materials such as glycol. There is a flow characteristic called the laminar flow boundary layer that will limit thermal transfer because of a thin layer of material that basically sticks to the surfaces in your loop and that heats up without being replaced by new coolant (think a drop of water sticking to glass, it is all surface tension). You get over this in a similar way to the extra cooling you get by dipping the radiator or coil in and out of the bucket because you are disrupting that flow boundary replacing the "hot" fluid contacting the fins with cooler fluid allowing better heat transfer until that fluid warms. Fins in the block do the same thing, but the distance between the fins has to be designed to optimize the laminar flow disruption for the intended coolant (making the fluid more turbulent, thus increasing heat transfer). This new block was likely better, but what may be happening is that a little bit of your coolant may be going into the fins, but not actually be replaced because the fluid sticking between the fins is causing too much resistance to flow between them and just pushing the material around the outside of the fins instead. You did hit on a better alternative to the glycol though. High percentage Alcohol, such as rubbing alcohol at 90% is closer in viscosity to water than glycol and doesn't freeze until -89C. If you decided to go this route, you do need to be careful because you are dealing with a flammable material running through a pump that also can eat away at your O-rings that are already going to be stressed due to the cold temps. Beyond that, you are talking needing to do custom part designs such as der8auer has done to meet your needs for specific circumstances.