Linus' home server cooling setup will have condensation issues in Winter without a heat exchanger bypass.

[Zierlyn]

==EDIT==

Just watched the follow-up video when Linus mentioned the pool being maintained at a minimum temperature. While not a perfect solution, it's reasonable, and if the pool is ever emptied in the winter, Linus can just manually switch all the devices to internal cooling. As such I would delete this comment, but that doesn't appear to be an option!

 

(Original comment below)

 

Just watched the recent pool cooling update video, and with Winter approaching, I feel compelled to warn Linus that without a way to bypass the heat exchanger, if/when the pool (glycol) loop nears 0°C in the winter, there will be no way to stop the indoor loop from similarly reaching the same temperature, creating a condensation risk.

 

At a minimum there needs to be a bypass valve/line that opens if the indoor loop temperature drops below the dew point.

 

The optimal solution would be a fan coil on the bypass line (located in the server room), so that the server would still have cooling if the pool loop shuts down for any reason (or for maintenance).

 

Extra bonus awesomness solution would be a modulating valve and temperature sensors with pump status monitored and controlled by an Arduino/Raspberry Pi, and connected to the internet to be monitored from anywhere in the world and can send out automatic alert emails/texts if things go wrong.

 

In any case, bare minimum there needs to be a bypass installed before Winter. Knowledgeable advice from a former HVAC Automation Controls technician. 

Edited October 27, 2023 by Zierlyn
Made this post before watching the follow-up video where the issue was reasonably addressed.

[emosun]

who cares , linus lives literally a mile a minute and even if it worked 100% perfectly it would get changed out for some other setup in 2 months anyway. not the kind of person that considers longevity in the slightest

[Needfuldoer]

If the pipes are buried below the frost line, that may be less of a concern.

 

He's using a water-to-water heat exchanger now anyway, so the PC side loop temperature should be at least somewhat warmer than the pool loop side.

[Kisai]

4 hours ago, Zierlyn said:

 

Just watched the recent pool cooling update video, and with Winter approaching, I feel compelled to warn Linus that without a way to bypass the heat exchanger, if/when the pool (glycol) loop nears 0°C in the winter, there will be no way to stop the indoor loop from similarly reaching the same temperature, creating a condensation risk.

ls technician. 

It never gets cold enough in Metro Vancouver to freeze a pool. 

Last year was a bit unusual, but a typical weather pattern for Metro Vancouver is no snow, ever. Sometimes it gets cold enough to snow for a few days, but it never sticks around because it needs to start snowing at night for it to stick around more than a day. 

https://weatherspark.com/h/y/1005/2022/Historical-Weather-during-2022-in-Surrey-Canada

 

Basically for MOST of the rest of the country, still water will freeze overnight in the winter. Metro Vancouver (Vancouver, Burnaby, Richmond, New Westminster, Coquitlam, Port Coquitlam, Surrey, Langley, Delta,  Tsawwassen) share the same temperate rainforest climate with Seattle.

 

You can occasionally see frozen-over shallow mud puddles in the mornings between October and March, but they're usually gone by 11am. Anything bigger than about half a meter usually won't even form ice.

 

Like even pipes don't tend to freeze in the winter unless they were leaking.

[nbrowser]

On 10/27/2023 at 12:05 AM, Kisai said:

It never gets cold enough in Metro Vancouver to freeze a pool. 

Eh it does get a touch chilly here once in a while...but nothing extended like say the BS that happens in my home province of Manitoba. The lower mainland here has fairly mild winters and that's not a bad thing.

[HenrySalayne]

Condensation doesn't only happen at temperatures near 0°C. The dew point for 25°C with relative humidity of 50% is 14°C.

However, Linus complained about high temperatures in the room and the loop, so it seems this isn't really a problem, because there isn't as much of a delta T.