0
- My Forums
- Tiger Rant
- LSU Recruiting
- SEC Rant
- Saints Talk
- Pelicans Talk
- More Sports Board
- Fantasy Sports
- Golf Board
- Soccer Board
- O-T Lounge
- Tech Board
- Home/Garden Board
- Outdoor Board
- Health/Fitness Board
- Movie/TV Board
- Book Board
- Music Board
- Political Talk
- Money Talk
- Fark Board
- Gaming Board
- Travel Board
- Food/Drink Board
- Ticket Exchange
- TD Help Board
Customize My Forums- View All Forums
- Show Left Links
- Topic Sort Options
- Trending Topics
- Recent Topics
- Active Topics
Started By
Message
re: A question for engineers regarding the heat generated by AI centers...
Posted on 10/21/26 at 2:45 pm to guzziguy
Posted on 10/21/26 at 2:45 pm to guzziguy
quote:
Why can't they run on a closed system with radiators and fans like a watercooled engine?
A lot of them do. The Hut 8 DC in West Fel will have a closed loop system.
Posted on 10/21/26 at 2:45 pm to guzziguy
quote:
Is it true they need a constant supply of fresh water to stay cool?
Not true. Would likely be way too expensive vs a closed cooling system.
Granted, there isn't really a spec for cooling water on data centers so technically they don't have to treat.
This post was edited on 10/21/26 at 2:46 pm
Posted on 10/21/26 at 2:46 pm to Cell of Awareness
For heat to do work, it has to have quantity AND quality. A whole lot of low quality heat is basically worthless here. There's a lot of thermal energy dumped into the cooling system of a data center but it has to be concentrated to do useful work and concentrating it isn't really feasible. Things like heating and de-icing don't require that much concentrated heat but plumbing in that water would probably not be cost effective either. Maybe a couple of hydronic loops to de-ice some of the parking lot and sidewalks but not much beyond that.
This post was edited on 10/21/26 at 2:56 pm
Posted on 10/21/26 at 2:48 pm to Cell of Awareness
this is like saying food can help fertilize the ground. It can in theory.
Posted on 10/21/26 at 2:52 pm to Clames
quote:
For heat to do work, it has to have quanity AND quality. A whole lot of low quality heat is basically worthless here.
Build grain solos next to the source.
Posted on 10/21/26 at 2:55 pm to aTmTexas Dillo
quote:
Build grain solos next to the source.
They are their own fire/explosion hazards, would be a big nope for me if I owned a data center.
Posted on 10/21/26 at 2:59 pm to Breauxsif
Last time I heard the term Carnot efficiency I was paying Jai $40 cash to help me cram for a test.
Posted on 10/21/26 at 3:06 pm to Cell of Awareness
quote:
Why can the heat being generated not be used to create at least a portion of the power they utitlize?
Quit watching Chinese disinformation.
Posted on 10/21/26 at 3:21 pm to Cell of Awareness
quote:
Why can the heat being generated not be used to create at least a portion of the power they utitlize?
I suspect they may use some sort of combination of absorptive refrigeration and heat from processes to cut the energy demands of cooling systems. I don't know this for certain but I would bet many have some sort of peak shaving power generating capacity that produces a LOT of heat and combined with residual heat from servers might be used with some sort of absorptive refrigeration. If they don't do something with it they have no doubt spent a lot of money trying to do something with it. Not to generate power but to cut down on air conditioning loads.
Posted on 10/21/26 at 3:49 pm to JL
Yes.
Everyone should have a Carnot Engine in their F150.
Everyone should have a Carnot Engine in their F150.
Posted on 10/21/26 at 3:53 pm to Breauxsif
quote:
Data center silicon runs around 85°C to 105°C, meaning your exhaust coolant is sitting at a useless 40°C to 60°C.
How cool does the input water have to be?
Posted on 10/21/26 at 4:10 pm to Bjorn Cyborg
quote:
Quit watching Chinese disinformation.
I was not. I was curious.
Posted on 10/21/26 at 4:58 pm to Cell of Awareness
This is how Ghina deals with the heat
Posted on 10/21/26 at 5:02 pm to Bamafig
quote:
How cool does the input water have to be?
To actually get enough of a temperature drop, which is delta T, to make a thermodynamic recovery cycle work with a 120 degree Fahrenheit heat source, your input coolant water would need to be chilled down to near-freezing, roughly 32-40 degrees Fahrenheit. Anything warmer than that, and the thermodynamic efficiency drops so close to zero that the system generates virtually zero usable power.
Posted on 10/21/26 at 5:04 pm to notsince98
quote:
It works great for the first stage heating to increase the water temperature and decrease the load on the process.
this.
and I assume our boy elon will come up with some ways to pre-heat water so it wont come in cold to the power plants.
Also, our people will come up with AFFORDABLE ways to capture all the methane outgassed by the oil rigs.
We've always known how to do it....just not affordably.
Also, I wrote an OP about how one day will will be able to use all nuclear waste in salt reactors...imagine that , our waste becomes energy. BUt it is just not cheap now. Good news is, nuclear waste expires so slowly that we can just let it stockpile and pile up. We can use it in 100 years once we find an affordable way.
Page 2 of 2
Popular
Back to top
Follow TigerDroppings for LSU Football News