Fun with ethanol cooling.
Fun with ethanol cooling.
I recently crafted an inexpensive custom loop and chose ethanol as my coolant.
(Reference)
It’s important to remember, although it offers some benefits, you need to exercise more caution than I did when handling something this risky. Take heed of my experience, everyone!
First, let's examine the impact on the acrylic reservoir:
Ethanol is not compatible with plastics.
Although the tubes and metal parts remained intact, the coolant led to pump failure by slightly expanding their interiors, causing the rotor to catch the pump body and halt operation. The plastic also weakened and warped under pressure (especially around hose clamps).
The drain plug was another casualty. It didn’t leak while running, but during draining it contracted and stuck to the screw lid, resulting in an unexpected burst when I released it.
Additionally, after draining, I didn’t flush the loop; instead, I refilled it with water (now containing only a few percent ethanol). This created a murky mixture—clear water mixed with clear ethanol and colorant formed a cloudy solution. I suspect some debris from the reservoir is suspended in the water now, though it looks intriguing until it causes another failure.
In summary, if you intend to use ethanol as a coolant, ensure no components react adversely. Opt for a glass or metal reservoir and a metal pump body (or at least one with sufficient space between parts to avoid swelling issues). Avoid acrylic blocks and flow indicators.
Yes, alcohols are excellent solvents. In a strong ethanol solution, acrylic would dissolve completely. You might also consider using plastics such as HDPE or PET (refer to the link). Ethanol would likely perform much worse than water overall—its thermal properties, like specific heat and conductivity, are about half those of water. I’m not sure what benefits others claim, but from a scientific standpoint there’s no solid evidence to back it up. An interesting experiment!
Boiler1990:
Alcohols indeed act as excellent solvents. In a strong ethanol solution, acrylic would dissolve completely. Other plastics such as HDPE or PET might also work, as noted in the referenced material. I also anticipate ethanol will underperform water significantly—its thermal properties (specific heat, conductivity, etc.) are roughly half those of water. It’s unclear what benefits others claim, but scientifically there’s no solid support for it. Interesting experiment!
It was enjoyable. The resource linked only indicates minimal interaction at best for ethanol; it can warp acrylic and cause cracking in certain cases, though it likely won’t fully dissolve it.
Its thermal performance is notably inferior to water. I’m observing roughly a 6-degree variance on my GPU during mining, similar results in stress tests. That might be fixable with a more powerful pump.