Maybe GRM got an answer to a different question than what they asked. It doesn't make sense for GM to go to the extra complication of computing something that can be so easily measured. In the grand scheme of things a sensor, the extra wiring and a bung in the spot where they measure the temp cost them a couple of dollars per car. Compare that to the incorrect warranty issues that show up due to a screw up in their algorithm.

Bill

 

Yes and your instrument panel displays the output. Now, who woulda thunk that?

 

Most automotive engineers are capable of doing a fair estimate of the oil temps when the engine is in the computer design stage. It is a straightforward analysis that most bachelors candidates can do. If the test mule measurements agree within about 10% of the model most engineers declare victory and the models are ignored at that point.

It makes perfect sense to compute the values before committing to a design. It's a great deal easier to change a computer model than it is to re-design an engine after all of the castings are machined and the internal parts are purchased and assembled.

 

Bill
The following is email I sent to GRM and the reply .Or should I say the reply then the questions I asked considering they talked to GM on the subject:
"Hi Gary…

Yay! Someone actually learned something form one of out project cars. That’s super exciting.

As for differences in temps between the analog sensor and the modeling, they were pretty close within the normal operating range. So up to about the low 200s the modeling was a dead ringer for reality. It’s when we started using ca outside normal operating ranges the the discrepancies started. One of GM’s theories was that part of the reason we were so far off was that the car was pretty much only being used on track. There were no street miles for the software to calibrate against, and the only use it saw was basically hard use. The theory was that this fooled the software into guessing high since it had no real baseline to compare against.

Now, as for the “Why”? Good question. When I asked the GM engineers I got a lot of rather plausible explanations regarding packing, cross-platform applications, etc. I have to assume that one some spreadsheet somewhere this was the best—or a least a good—solution. And, probably, for 99.9% of the drivers in 99.9% of situations it works fine. But, yeah, it does seem like an overly complicated solution to a rather simple problem. Maybe when you’re dealing with numbers and volumes like GM must be, though, economies of scale shift you toward solutions that may not see simple to us regular folks. Although, it’s somewhat telling that Chevy had the modified drain plugs with the thermocouple just sitting around for their testing. It’s clearly a parameter that they value measuring accurately, at lest in the development phase.

Actually, a lot of the development guys I talked to sort of silently agreed it’s a less-than-ideal solution, but, again, it’s hard to build cars for the 1% when you’re a massive company. I think the takeaway message is that we’re lucky they’re building such cool stuff to begin with, and are willing to support us when we find the weak points in the system. That’s how I comfort myself, anyway

Glad it worked out for you, and glad we could help. "


On Jan 22, 2016, at 2:10 PM, wrote:

Dear JG,

I have been a subscriber of Grassroots Motorsports for many years and I enjoyed your October 2015 article Factory Backed Part 2.
The timing of the article was right on in relationship to a project that I was doing on my 2005 Corvette track car that I use for HPDEs .
I just installed an AiM MXL2 dash that collect data from the OBDII port that includes along with 20 other channels the engine oil temperature . I discovered that the reading on the OEM dash digital display did not agree with the display on the AiM dash. It was found to be sometimes 30 deg F higher than the OEM reading . I discovered with the help of the AiM engineers that the AiM dash was reading the signal before the computer modeling took place and we performed tests that showed that the OEM dash display would never agree with the real oil temperature.
AiM sent me a new sensor to install in a separate location from the OEM sensor that would be used as my engine oil temperature displayed in the MXL2 dash as well as being logged . This signal will come into the MXL2 in a separate channel independent of the OBDII inputs. I now have the new sensor located in the outlet of my ARE dry sump scavenging pump. All of the other OBDII inputs such as oil pressure and engine coolant temperature seem to be accurate.
My questions to you are : 1. How does the new drain plug sensor temperature display that you installed in your Camaro compare to the OEM display? 2.What reason if any did GM give you for computer modeling of the OEM display?
This modeling of oil temperature is crazy and makes no practical sense. AiM had no idea that GM was doing this modeling so from now on they will recommend to their customers to use an independent sensor
Respectfully
Gary Bowler

 

Gary,

Very nice job of doing real research rather than merely speculating!

Now, spoiler alert - SARCASIM ahead. GM doesn't really care if the driver knows what the real oil temperature is - they only care that the ECU knows. They are trying to meet ever unrealistic EPA requirements and provide a 100K mile warranty on a motor approaching NASCAR power-to-displacement ratios. The ECU will take all of the performance inputs, including oil temperature at some pre-determined point in the oil flow stream, and "fix" any problem the operator creates by pulling timing, altering fuel flow mapping, going into limp mode etc. Throwing a big oil temp number on the DIC and "suggesting" the driver run a cool down lap at some point is a nice feel-good gesture, but the ECU will soon fix the problem anyway. GM is just protecting their investment (and warranty repair dollars) from the masses, many of whom are just in over their head with this level of performance car.

Pappy

 

Hate to break it to you ace, but I am an engineer. I wasn't speculating; I was speaking from experience.

 

Kind of thin skin? - I wasn't talking, or even thinking about you, or anybody in particular. I was simply praising Gary's diligence and hands-on approach to solving a problem as he saw it. Lots of people offer an opinion on this forum, but very few have any real facts or documented research to back it up. I am also an engineer, and fairly knowledgeable in the use of HPCs in developing, verifying, validating, testing, and implementing advanced modeling and simulation programs in the defense aerospace industry - but that does not make me an expert in the Z06 ECU. I was interested in Gary's follow-up with GRC because I am experiencing the same oil temperature anomalies with my 1LE SS that GRM saw, and it was reassuring to see that Gary, using AIM data pick-up, validated what the GRM editors were saying.....that the gage presentation of oil temperature does not always match the analog reading taken at the pan.

Have a nice day.

Pappy

 

Thanks for the complement . I spent today adding two more channels of analog inputs to the AiM MXL2. The MXL2 has the 21 OBDII channels and I have now completed installation of sensors for crankcase pressure, ARE dry sump scavenging pump discharge oil pressure to complement sensors I installed a few months ago for brake pressure , fuel pressure & o2.
By the way my day job is providing control and monitoring systems for ship's propulsion G.R. Bowler,Inc.