[petre]

Would a cooling pond with controlled discharge into the river maybe mitigate this problem? One could also cycle the water from the cooling pond through the plant's cooling loop. The worse that could happen is evaporate the water from the cooling pond which is gets replaced with new water (sans fish) from the river.

[topspin]

> Would a cooling pond with controlled discharge...

There are any number of methods of dealing with this. Using natural bodies of water is a lower cost (thus highly popular) method of sinking heat. Palo Verde has been operating 3 large reactors in the Sonoran Desert in Arizona since the 1970s. There is no reason in engineering or physics that a power reactor must be highly sensitive to any particular lake, river or ocean's temperature. This is strictly about money, being cheaper to rely on a big "free" heatsink, and leaving too little margin for changing conditions.

[wcoenen]

Nuclear plants can run with those hyperboloid cooling towers as the sink for waste heat. They are just a lot more expensive to build, run and maintain compared to using river water.

[slow_typist]

You still need water when cooling with towers. Instead of heating the river you consume some of its water and release it into the atmosphere.

[bequanna]

Not TOO much loss tho, right?

I thought the cooling tower design helped maximize cooling and minimize loss as steam.

[downrightmike]

making modern websites not 100's of megabytes heavy would go a long way

[delecti]

FWIW, this one was only about 5 MB.

[tjoff]

The images related to the article are about 0.5MB.

[Choco31415]

There's plenty of work happening with adapting non evaporative cooling loops for data centers. Maybe that can be applied here too, even if to mitigate some of the cooling needed?

[Retric]

Data centers can afford to spend way more per kWh of heat discharged. Nuclear needs to dump 3+kWh of heat per kWh of electricity generated.

[CobaltFire]

Nitpick here:

Nuclear discharges more like 2kWh of heat per kWh of electricity produced.

Perhaps slightly more, but its in that ballpark. Seems like a small difference but you are overstating the waste by near 50%.

Source: Degree in Nuclear Engineering, previous life was a Reactor Operator. Also, just look up thermal efficiency of currently operating reactors.

[Retric]

That 3+:1 is actually from being pedantic.

2:1 implies 33% thermal efficiency to the grid which is commonly achieved in normal operations.

Unfortunately engineering means you can’t work with such overly simplified models. Many ways of dumping heat run into issues from solar gain prevents dumping significant heat under the employe parking lot etc.

[CobaltFire]

I don't disagree but am having a hard time parsing the second half of your comment.

In my experience we saw some losses when taken as a system that prevented us from hitting 2:1, but they weren't anywhere near 50%. We also ran closer to 40% than 33% thermal efficiency in the power generating loop, so many of the system thermal costs are baked into that 33% efficiency to the grid.

Basically, each reactor install will have it's own minor issues. Overall they should be hitting around 2:1 as an entire system. If they aren't there is something going on that I haven't come into contact with (which is quite a lot, I'm sure).

[Retric]

> hard time parsing the second half of your comment.

I’ll put it as a spherical cow engineering problem. Let’s suppose the temperature is 30C and you want to dissipate 2GW of heat across a 1kmx1km flat plate, what temperature will it get? That’s a relatively straightforward calculation if you don’t need high precision.

However in the real world it wouldn’t have a single equilibrium temperature. 30C at night the heatsink would be one temperature and in the daytime it would be a different temperature due to sunlight.

Evaporative cooling largely sidesteps this issue, but it’s worth keeping in mind why that’s chosen over other seemingly cheaper options.