S-Prihadi

I said sometime ago, I will scope this heater for thermostat.... and I did.


Thermocouples locations for BEFORE & AFTER




Alu foil is to keep heat in. Clear double side tape is for anti vibration and acts as heat shield too. Cable tie is insurance for vibration.
Each hose end for radiator has steel ends and is wet on the inner side. So this steel "lip" is a good place for thermocouple to sense coolant temp.








Hose end steel lip





After thermocouple passed my 35.x C finger temperature, now Testing AMBIENT and Heat Gun induced heat








Since these data set below has no excel file output, I can only video log them.




Same with the Banks Gauges, rather than I take its excel file and insert into video ( waste of time and the data to video sync is a bie-cth ),I might as well video log them too.



=========================




Timecode is zero at ENGINE START.






Will continue..........

S-Prihadi

The AFTER thermocouple starts to get hotter at 41.6C than the BEFORE at 34.8C, this is because the BEFORE ( thermostat output ) , the themostat is still closed and hose end only get heat from the plastic housing of thermostat assy.
The AFTER gets hotter coolant which is from the rear of engine block ( and turbocharger ) to the HVAC heater core and injected back to suction Tee of coolant circulation pump, and this flow of coolant has passed thru cylinder head and engine block.
Suction Tee of coolant circ. pump is only like 10-15cm away from the AFTER thermocouple.



Thermostat still not opened at 89C coolant temp.
Coolant temp sensor location is very close to the path for coolant output plastic pipe heading to HVAC heater core.



At 90C coolant temp, the heater at thermostat gets some PWM power pulses by ECM at 18% duty cycle.
I do not know the relationship of thermostat millimeters opening vs heater duty cycle, except at 100% duty cycle of heater...by logic thermostat should be fully open X millimeters.





I wrote SLOW FLOW, that is I am assuming there is some flow already at the radiator, meaning thermostat is opening a bit after its heater being pulsed at 40% for a short while and was 30% duty cycle for sometime.
25th second to 46th second, that is 21 seconds the heater been pulsed at 30% ish duty cycle between last image to the one below. 95C coolant temp by now






See, now coolant temp has gone down to 87C and the BEFORE has shoot up to 50.5C and AFTER at 58.4C. 9 minute 2 seconds, that means 37 seconds total the heater been pulsed at 30ish % ducty cycle and then goes back to 18% ish again.
So indeed there is some coolant flow at the radiator already. HOWEVER, here I am missing the duty cycle of the radiator fan, which is so important as you will see later on.
AFTER technically should be cooler than BEFORE , but it is being "contaminated" by the heater core and turbo return coolant channel you see, so its not an easy overall calculation of how much heat are removed by the radiator for below.






At heater full power by ECM, no more PWM, simple full system/battery voltage.




Here we start to see some real heat removal at photo below, the delta between BEFORE and AFTER is getting wider at like 20C spread . Again, I wished I knew the duty cycle of the radiator fan DUGGHHHH
I will probably test again tomorrow and use Xentry for radiator fan duty cycle info. Me too lazy to probe fan connector as I need to lift the car first, remove steel under panel yada yada and the fan connector on mine I had given
extra insulation protection.




So I raised RPM again to get more heat into coolant.




I then slow down engine RPM.......... and I know engine coolant temperature will climb up.




I then turned OFF the engine.




WILL CONTINUE.....................

S-Prihadi

Actual engine KILL is at 22:08.440. Engine restarted at 26:27.000, so engine OFF duration approx 4.5 minutes.
So I started the engine again I then decided to disconnect 1 wire to the thermostat heater to disable it at 31:58.960 or approx 5.5 minutes into 2nd time engine running/started.
During engine 2nd time started, heater was not activated by ECM at all, until 31:08.920 or 4.7 minutes into engine already running.

Here I realized it should best to do this HEATER DISCONNECT when coolant temp is well below 60C, hoping the wax has solidified completely.
So tomorrow will be follow up test.


Anyway..................





I then raised RPM again to heat up coolant.




Here I think I am seeing safety strategy from the ECM, it seems the radiator fan running maximum earlier and longer than if the heater is connected, based on the same 104C coolant temperature reached.




Wow, Delta of 31C between BEFORE and AFTER, this was not implemented by ECM at 1st session engine running.





See, 1st session where I hit 104C too while heater is properly connected, the Delta between BEFORE and AFTER is not as great as when I disconnected the heater. Delta here is 5.5C




Delta here still a low 19.6C




At 96C coolant temp and I raise RPM for faster coolant flow, still the delta is only 21.x C , this means the radiator fan speed is not at maximum.
Pushing more coolant to radiator but no cooling air flow = kinda useless. This is a STATIONARY TEST.



So, conclusion for today, a temporary one is :
01. Wax FULLY melting temperature at thermostat is a tiny bit higher than 95C I think, maybe 100C............. let see tomorrow. I hope its not BMW crazy at 108C.
Do note that BMW coolant pressure cap is 2 BAR, we are 1.4BAR, so I doubt MB has the ***** to choose 108C wax for thermostat.

Tomorrow we test again with Radiator Fan Speed /duty cycle data.............. YEAH !!!!

CaliBenzDriver

Surya, this a lot of good data insight about that variable Tstat temperature. The way Bavarian Motors have their own version, hints of shared design.

Why is micro-managing coolant temperature such a big deal.... ??

• PWM Tstat
• PWM 100A Fan
• serpentine pump
• 'lectric pump
• oil pressure Mgt
• (engine rad shutters)

I guess warm up the engine oil exchanger and tranny faster than was possible with unmanaged thermostats.


MB radiator shutters
I pictured that car in Vietnam circa 2016. Sub-tropical country where cold is never a factor... I thought it was for dust protection while parked.
Since then I have seen Mercs drive around with these shutters closed.

S-Prihadi

This MAP based heater assist thermostat is not for faster warm up like the radiator shutter, it is for ability to do 105C at low load and save fuel. Same as the stupid ECO-Start-Stop.
As shown on Male website on its MAP based thermostat, and this guy below break it down nicely too on the MAP/Table part:




I came across this from Scanner Danner and since this is not hot rod Bavarian engine, it is more inline with our non AMG Mercedes coolant temperature strategy.





So it is probably very similar on our thermostat, 104C is the fail-safe wax will melt 100% temperature.
Since wax is not an ON-OFF thingy like a bulb, surely its wax has temperature range from hard to soft to start to melt to very liquid and this represent how much % opening on the mechanical thermostat.
If the above data from GM stated 98C to 104C as the range, probably our MB is very similar.

Something unique I saw today, while using Xentry.
Some makes sense and some doesn't.
BTW, the current at full heater power is 823 milliamps. Inline with one youtube tester testing a Mahle.


First I will show you the ECM coolant temperature demand based on cold engine just started and then I applied load , unfortunately stationary revving only.

First time engine started from cold, ECM wanted 105C coolant temp....so it seems.



By the time coolant temp has hit 89C, Xentry changed the #1 parameter to 90C .... don't ask me why it does that without even hitting 105C to begin with




Yes.... a good laugh, I do not know why Xentry indication of ECM PWM power to coolant heater R48, why opposite value of what is actually happening.


See below, now heater at full power and Xentry said it is 0%...... LOL
Also now ECM wanted 80C coolant temp. There is now way 80C can ever be achieved in my climate and the way the MAP thermostat table is set at....
There are data from Xentry which I find doesn't make sense. This is not the first time.



So I revved up the engine to heat up the coolant to 102C ish....





When 102C ish achieved, I let engine idle and more heat will be generated from slow pump impeller speed and coolant temp will then rise to 106C ish.

ABOVE : 81% Duty cyle is the highest I seen on this test. MB claimed 90% is max but I dont know if 80% to 90% is a true linear one ??




Below, with heater properly connected, cooling down seems to lock fan speed to no higher than 31% duty cycle.




WILL CONTINUE................

S-Prihadi

Before I forget.
Since my test was stationary , I did not get CEL when I unplugged the heater wire.
I do get fault code.

I once forgotten to connect my MAP sensor and the CEL only appear like 1 KM of use. I already knew something is wrong because my engine is so low power.

Here is the DTC for the heater

S-Prihadi

So, I decided to disconnect the heater wire, one, the +12V.








It seems when there is a heater problem, ECM strategy is to increase the duration of the 80% duty cycle for radiator fan and then use 39% duty cycle much longer too.
So the DELTA between BEFORE and AFTER coolant temp at radiator is really high 97.6C to 66.0C , this wont be the case if heater is properly connected. Usually the AFTER is 70ish C at best.








So the heat removal control is via heater at thermostat, thermostat and the speed of the radiator fan.
It doesn't seems a single data/method approach the ECM is using.
In the old days our cooling fan is ON-OFF only, so not much playroom.

The bottom line is, MB is not worried of heater failure leading to overheating, since the wax melting temperature is not insane like BMW.
However, having proper 50/50 coolant to raise boling temperature and making sure the pressure cap is within 1.4BAR to further raise boiling temp.... is very important, because without those 2 important criteria, our engine can not
get the approx 132C of boiling temp of the coolant.

I do not know which one will wear-out first. The heater or the mechanical thermostat.
I seen thermostat kinda binding at its moving stem and could not open 100%.
Heater as we know, has a operational life too and probably poorer performance before actual defective level able to be sensed/detected by the ECM.
So it would be wise for me to ( must use OBD2 ) observe that I am not to see higher than 95C at engine high load, except when at the track

S-Prihadi

Good thing I keep old archive of interest.

Initially before I tested this MAP based therrmostat, I am under the impression that MB on purpose wants to heat up my engine to 105 - 108 - 110C ( 221 - 226.4 - 230 F ).
After looking closely at how the Thermostat MAP table is written, which is to open up the coolant flow more in an engine higher load condition and let engine be cooler, I believe MB MAP/TABLE of the cooling fan
operation ( duty cycle ) is bad for tropical country, at least for my engine.
This is aside from the cooling system is indeed poor when driven decently agressive , see here : https://mbworld.org/forums/e-class-w...ml#post8646265

I think our M276 thermostat is Walher and not Mahle. That is what FCP Euro is showing.
Wahler been bought out by Borg Wagner.
Mahle bought Behr the OE for my aftercooler and probably W212 HVAC sysytem too.

Anyway, both Mahle and Wahler share similar temperature goal for coolant.

From Wahler, old catalog










Same temperature information as Mahle pointed out too : https://www.mahle-aftermarket.com/me.../mo-2-1013.pdf


So Mahle has the MAP/TABLE. I will compare it to my actual data.




















I kept this video log as archive and I have 3 sessions I can show and I realized the hot coolant temp I am getting is actually from car climbing the gradient and not all about the MAP based thermostat trying to steal
opportunity when in the rain ( cooler ) or when I climb higher altitude and get the cooler temp. I thought that being in a cooler ambient temp the ECM decided to heat up my coolant in the name of fuel savings.
However, when I see the engine load up close, the climbing coolant temperature is very much likely the ECM does not want to speed up the cooling fan quickly and more often.

The video below is of 3 different sessions.
1st 2 sessions was 2020 when I only have Torque App and using bluetooth OBD2 XL Link. So data capture is limited.
3rd session was 2021, I already have the Banks Gauge and Viofo DashCam.


105C , was 93C and then raining ( cooler ) and gradient on the road = 105C





108C, was 99C or less and then climbing up the mountain, with gradient kinda "polite", nothing tough.






110C, was 100C. This video is not full length. I use the part only where I stopped for a while checking my Google Map. This is the coldest ambient temp the car ever experienced... LOL.





The video of these 3 sessions is below





3rd Session, I have complete data.

The data below is exactly the same as the on screen data on the video but it is easier to see the one below










Based on Session 3 engine load data...........
So, if the MAP table from Mahle indicated at engine load +- 60%, the target temperature would be 94C. Or let say we use 55% load, target temp is 97C....... the only component which can heat up
my coolant that high at Session 1, 2 and 3 is the cooling fan. I have experimented with the MAP based thermostat as per 1st post of this thread. At 95C eventually the heater is maximum power, aka thermostat fully open.
43 MPH is 69KKM/H and you can see my Session 3 average speed is 40KM/H only , so the fan should kick-in to assist more often.

The 108C coolant temp, that zone my engine load was higher than 55% average and has a few high peaks 75% or higher.





Well... I am quite un-happy with this temperature MAP/TABLE my engine has.
Session 3, I was driving so polite and so gentle.
IF I drive agressive for Session 3, probably the coolant will hit 120C and that is lame for ambient temp so cool at 20C.
HHmmmm why why why MB ?