I was checking eval monitors while driving, I'd cleared a code that was my fault. interestingly enough, dash gauge says 90 to 95 C, but scan data says 221 F. Something don't jive. Care runs beautifully, coolant replaced 3 months ago, reading on gauge before and after is the same. handheld temp gun reads 199 F at thermostat housing. I was in OBD live data on scan tool, a ThinkTool Pro.

Any thoughts are welcome.

 

Are you going by the instrument panel gauge for the temp at 90°-95°C? If so, I wouldn't trust it. VW was notorious for the gauge reading straight up, but the actual temps fluctuating greatly. However, 221°F seems a little high from normal. My car is steady at 198°-199°F, which is exactly what you were reading at the coolant flange.

 

I had @BenzNinja code my instrument panel with the AMG menu, which is a nice feature. My instrument panel temperature gauge is showing ~92°C which equals ~198°F.

 

The 90 to 95 C is from the Instrument cluster, I've always referred it as "the Dash". My old W123 ran 80C but using meter at the sensor was actually190F. If the temp was really that high the fan would be on high, it's not.

Going back into normal scan data, it appears the ECM requested that temp of 221F. After sitting there idling, as temp got close to 199F, it requested 178F, the heat was off, so I turned it on and the requested temp increased to 198F. But actual engine temp was 184F.

It also shows fan request Run on-time and percentage of on time, as well as requested voltage to fan. On request from ECM, on request from A/C, as well as temp sensor start up heater on-time and voltage.

Quite a few more parameters than what I'm use to seeing. I am satisfied the car was not overheating but was doing what it was commanded by the ECM to do.

 

I'll have to check with Peter to see if it's available for mine as it's a 14' E350 W212. My dash...Instrument cluster is much different than yours. I have a couple other things I need too.

 

One thing I forgot. The whole time of the re-test, the instrument cluster gauge was steady at 90C while ECM temp varied from 165F to 191F assuming during thermostat operation.

Other cars I've worked on, ECM temp and instrument gauge would still move up and down when thermostat would open and close.

 

Now, If I could figure out hoe to keep it from getting over 210, I'm not comfortable with 221 at all.

I wonder if it can be coded to run cooler, like a hot weather environment strategy.

 

A couple of days ago, my coolant temp hit 212°F when it was warmer out (~60°F). Not sure what the ambient temps are where you are at.

 

Yes it was about 67F here when I was testing.

 

I used to think there was something fishy about the extreme high temps in mild weather.

The idiot gauge is a decoy for fools... I understand MB want this to look pristine for resale value. I wish there was a way to code the gauge to display true temps!!! What a concept!

Now I think there may be something buggy about temp sensings or smart Tstat remote controlling...
we have something to discover with this poor cooling control. Tstat is not allowing proper circulation regardless of PWM control. One of those hidden type faults.

 

I agree Cali. I thought of measuring different temp sensors to lower temps reported to computer, but the PCM would just compensate for the variance.

Gotta find out the conditions the ECM looks for to raise temps and remove just one of those conditions.

 

Thanks Rick, all this good thinking might lead us to another positive discovery.

Granted we know Bosch target hotter engine temperatures... for some reasons

What's involved is smart Tstat PWM heater control and temp sensor at the rear of Bank-2. With that ECU uses a lookup map to gently smooth things out with computations.

Something simple hidden in that complexity is not working well.

 

Perhaps the ECU is using some cycling algorithm, and that will not look nice on the cluster with the gauge dancing during normal driving with a good engine. So they have prepared the decoy to stay within the expected range of the algorithm so the oscillations have been SMOOTHED out.

Now, let me be cynical here for a moment. If there is such cycling, is it for the good of the engine, fuel savings or to stress out the plastics enough so they fail at a given interval?

 

You right Juan the idiot gauge Surya first described for us is only here to keep us amazed. I don't have a problem with that so long we know to expect lies...

The issue I am kind of getting concerned with with reducing heat normally like any car. I think my fan is still more hyperactive than I wish for a giant clean radiator in mild-weather

Before unplugging the solenoid everything under the hood was sizzeling hot... CPS toaster.
Coolant and leaky oil smell since Day-1 factory new. Nice!

Lubricating the pistons tremendously help remove the heat storred by the engine block itself. That is great transformation!

The other day I saw the stupid gauge doing cazy dancing like a remaining painted GND messing with my temps. I saw idiot gauge move a quarter inch... plus the restlessness of supersonic fan in mild weather without demand.... I think there a yoyo in there asking for my personal attention.

I am not in ballistic mode, just close watch observation stage.

CReader graphic is next to useless because the time scale is preset to 5 seconds. I am interested about temps over 15Mn!

I fully expect a map between the head sensor, Tstat heater and the PWM Fan.
This being a perfect example of a spaghetti system missing useful error codes.
I give it full benefit of the doubt as I don't exactly know the tipping points of this design.

I may well be that fan is actuated through computed data that are not current. Now I have less heat, I expect sleepy fan.

I have zero concern connecting this heat topic to Solenoid-Delete that greatly helps remove heat through the heat exchanger.


I am curious to learn something with my own set of "itchy diver's hands".


++++ OBSERVATIONS:
Scanned temps looked right
Fan can act a bit hyperactive
This is SAFE but a bit odd
(Unlike legacy: T° Sensor > Relay > Fan On/Off)

+++++ LIVE OBSERVATION:
As expected it's a combination of variables...
Tstat opening, block temp, fan speed + software logic.

M276 NA cooling

The ECU builds a map to deliver best control of Tstat opening + Fan PWM Ratio for engine Temp control.

 

My ML550 in 100k miles has never shown an full operating temptatutre on the dash of anything more or less than 85-90 degrees C. That includes hot weather that can reach into the low 40s (110F) here during summer, and with a car load of kids and their holiday stuff. That alone tells you the temp gauge is rigged, especially since the thermostat only opens at 212 degrees F/100degress C. The only thing I use the guage for is to know if I've reached full operating temp so I can put my foot down, or to look for a change in temp.