Quote:
Originally Posted by KillerBMotorsport
This has been debated to death. The problem with wanting to know 'worst case scenario' is that it's a massive rabbit hole. Where do you take those temps at to get the maximum temp? No one knows. Outlet of the pump, before the mains, at the mains, after and before it splits to the left or right head, intake or exhaust cams, or any of the drain-back passages? I'm sure I'm missing some. My bet is on the exhaust cam journals being the hottest.
It's just not practical unless you have an engine on an engine dyno on a test stand. It's likely be done, but that data isn't available, which brings up another point. How do you know if temps at those locations are normal or not? There's no normal operational data out there to go off. What would be the coarse of action if you did see a temperature you felt was too high, but the OEM sensor was within normal? Assume that's OK, or there's an issue?
I'm all for more data, but it's a tricky rabbit hole. The sump is a standard location for measuring oil temps (what going into the engine) in motorsports. Not that measuring in the pump is bad; it's certainly better than nothing.
As another note, there is speculation that Subaru skews this temp through the ECU to project higher temps on purpose. Some have speculated to be on the conservative side of safety? Maybe it's a cheap sensor? Ours are lab grade and get calibrated at each use. Whatever the reason, we'll continue to monitor this over time. Maybe with higher ambient temps that delta decreases? IDK, we're just here to share information  |
I would certainly understand if OEM sensors project values on the safer side, but from an ECU calibration standpoint it probably wouldn't make much sense as all aspects for that calibration related to that sensor would similarly be skewed. You would have a coolant temperature showing at 90C when it is actually at 80C, with the tables based on 80C which it is actually 90C. Probably would make warranty claims based on freeze-frame data a lot more difficult as well if everything else was similarly skewed... but I see what you mean.
As a person who entirely relies on OEM sensors - thinking through I would not expect the sensor to be accurate but rather be consistent, at the very least there is a point of reference to work with.