Zitat von »Taubenhaucher«

I think the sensor will do exactly what the thread creator expects it to do and the expectations are in sync with the results over the course of the year.

Why argue further, you will not be able to convince him, he believes in his sensor.

The fact that you can achieve the same or better results with a modern motherboard, a few temperature sensors and the fan control in the bios will not convince him.

There are simply people who always have to reinvent the wheel.

That's OK, but please don't try to impress everyone else who has been dealing with the matter for years with concepts that have long been exposed as nonsense.

For the water cooling of a PC, common sense and a little bit of physics is enough.

Translated with www.DeepL.com/Translator (free version)

Well i tried '

I have tried so many times and you can't beat faith, no matter what the subject is.

My two cents.

I found this entire thread fascinating. I have seen some quite clever Playground creations on this forum and have learned a lot from some of them, so when I saw the Thermal Regulator and the Intercooler Thermal Regulator I was intrigued. Once I got past the Superhero looking guy with the puffy raincoat about to take out (or maybe be fried by) 3 space ships, and the bad sci-fi novel intro, I read the description several times and traced through the diagrams. The first one wasn't too bad but the meaning or relevance of some of the calculated values was fuzzy to me. He lost me on the second one. Hats off to Remayz and cptninc who both seemed to understand what this thing was doing and even predict it's behavior in a smaller loop. And credit to Teknophage for thinking out of the box, and dedicating a huge amount of time to thinking about the subject in general. He also did make an earnest effort to respond to every counter statement.

Remayz's statement that Δt between coolant and ambient is the single most important metric to define rad efficiency and control fan speed is absolutely true. Ironically, coolant temp and ambient temp are the only 2 inputs to the Intercooler Regulator. I may never understand the number processing he does on these 2 values, but it would be easy to hold ambient steady and plot the output as coolant temp increases. What this does to the fan speed will depend on the curve it drives. Remayez actually did this, with less than stellar results in his case. If I understand correctly, the difference in performance was attributed to differences in the loop total thermal mass and a very flat fan curve. It surprised me that loop thermal mass apparently had such a significant effect on the behavior of the Intercooler Regulator.

Personally, I'll stick to a constant pump speed and coolant/ambient Δt for fans. If I decide to use a set-point, the PID controllers Aquasuite provides do that well. I think Taubenhaucher called it. Teknophage spent a year dinking with this and firmly believes his Intercooler Regulator is superior to simply using coolant/ambient Δt to control fan speeds. He also thinks varying pump speed is worth doing. He is pleased with how these sensors perform in his system and he will not be convinced that this an overly complex waste of time after spending a year working on it. I guess if nothing else, it demonstrates the power of AQS, misapplied or not. Certainly an interesting thread.

Zitat von »Speedy-VI«

If I understand correctly, the difference in performance was attributed to differences in the loop total thermal mass and a very flat fan curve. It surprised me that loop thermal mass apparently had such a significant effect on the behavior of the Intercooler Regulator.

Turns out it does't have any relevance at all. That was my assumption before i looked at the OP's fan curve that is almost flat. That's why his fan response was so small and mine so drastic. Having the virtual sensor vary speed between 90 and 95%PWM doesn't do much. But set on a fan curve that actually slows down the fans in idle, it had pretty explosive effects.