Inferred vs Actual Oil Temperature - Results

[D/\rK•650]

Thank you for posting this and all the information you've provided. It crazy because I was wondering this just the other day. Im driving and thinking how accurate is this oil temp reading? Great job

Sponsored

 

[D/\rK•650]

So I wondered are the rest of the gauge readings direct? Only oil temp is computer inferred?

 

[D/\rK•650]

krisk > how difficult was this to install the oil temp gauge and run the line ect?

 

[MAT1955]

We had to calibrate every oil temperature sensor we purchased. For our marine racing engines that was usually autometer or gaffrig. A variance of 5 degrees from actual was common - either way 9BTW just like engine hp from mass produced engines varies 5%) anyway, even once calibrated temperature gauges relying on sensors, wiring and gauges were and I suspect still are prone to loosing their reliability. I'd be willing to bet a lot that an inferred computed modelled temperature was more accurate.

 

[krisk]

krisk > how difficult was this to install the oil temp gauge and run the line ect?

If I had done the job start to finish at one time knowing exactly how it was going to end up, I probably would have spent 6 hours. That said, I don’t get in a hurry when I’m doing these kind of jobs. In other words, I’m slow. Someone who works fast could probably get it done in half the time. I have bus bars installed above my fuse box for adding electrical components, and I was doing some other wiring at the same time. The vent gauge requires removing the display on the dash, which isn’t nearly as bad as trying to put it back in. Getting the screws back in without dropping them down the dash takes some work.

The thing I would caution someone wanting to do this on a street driven car is make sure you use an appropriate oil filter. I did a fair amount of research on oil filters trying to see what the options were, and there are few that have the bypass valve pressure setting of the FL-500S, which as best as I can tell is 22psi. It’s tough because a lot of the time specs are not published and you find conflicting information online, much of which is conjecture. I’m running the Moroso race filter which is a high flow, low restriction 27 micron filter. I have no problem with this because I am changing the oil and filter every 6-10 hours. I would not run this filter on a daily driver going 5k miles or more between oil changes. Going to a shorter filter means giving up filtering surface area, so I would be careful just slapping on any filter that “fits”.

 

[krisk]

We had to calibrate every oil temperature sensor we purchased. For our marine racing engines that was usually autometer or gaffrig. A variance of 5 degrees from actual was common - either way 9BTW just like engine hp from mass produced engines varies 5%) anyway, even once calibrated temperature gauges relying on sensors, wiring and gauges were and I suspect still are prone to loosing their reliability. I'd be willing to bet a lot that an inferred computed modelled temperature was more accurate.

I’m using the AEM X-Series temperature gauge. I have no idea the accuracy of the gauge, but there was no reference to being able to calibrate it so I’m not sure that is even an option. I expect it is within a few percentage points of actual.

Correct me if I’m wrong, but wouldn’t calibration just adjust the readout up or down by X%? In comparing the gauge reading to the car inferred value, there was no linear correlation to the temperatures. So while the AEM gauge may not be 100% accurate, even once calibrated, the rate of change is going to be exactly the same plus or minus the calibration percentage change. The relation of the temperature dropping more rapidly on the gauge versus inferred when cruising and being 10 degrees lower after being higher when at 4400 RPM would not change with a calibration. Am I thinking about this correctly?

Sponsored