When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
ECO charging Algo vs Valeo built in charge controller
Yes... I am still not happy with MB ECO charging algo
Tested as follows :
01. Remove connector to B95 Battery Sensor, Hyundai Mobis. Power and LIN, in that connector. PLAN CANCELLED. I lost amperage data.... too precious to loose.
Hoping that the charging algo resides there at the Hyundai Mobis. Because it has Software Version number ..... LOL
So, B95 Battery Sensor is highly likely used by MB only for its shunt, amperage measurement.
Another mystery here, so....where is the UB : VOLTAGE information coming from ?.... now that VOLTAGE is still available even after B95 Battery sensor mini connector got un-plugged.
Perhaps power reading/sensing point is at Front SAM ...or Rear SAM or F32 Prefuse ? Who knows... if by analog means it is difficult to track.
One thing for sure, the VOLTAGE information sensing point is somewhere very upstream where voltage drop downstream does not effect it that much.
02. NEXT ALTERNATIVE. REMOVE THE LIN CONNECTOR TO ALTERNATOR. DONE
Check engine light does not show, but Trouble Code can be read at Front SAM, or Rear SAM, I forgot which SAM.
03. I hook up a direct battery post, a multi-meter. This is to compared the VOLTAGE reading from instrument cluster vs direct at battery post. DONE
04. ECU voltage and OBD2 Bus Voltage or pin #16 of OBD2 connector voltage which is supposedly provided by rear SAM is read by Bank iDash 1.8 data logging gauge called Data Monster. https://www.bankspower.com/series-2-...0engine%20temp.
Logging is at 1 Hz or 1 second per data point.
05. Since the Instrument Cluster voltage and amperage data is critical to monitor for this test, no other way but to record the instrument cluster and do a video logging.
I added the ECU voltage and OBD2 Bus Voltage from Banks gauge from its CSV file using Race Render video editor and is showned as a Moving Graph.
The graph Left To Right (X Axis) is worth 45 seconds.
Accuracy between instrument cluster video frame to the data overlay from CSV file is within 0.5 to 0.7 second.
I have attached the 2 CSV files if anyone is a data junkie.
Since the Banks gauge is set to auto logging and sense RPM as activation to power up and then do record/logging, after I refueled, it produced a 2nd CSV file ( Log 43 ) and the data overlay is 10 seconds late after refueling's engine start.
Approx 10 seconds is the device boot time to finally create the CSV file.
This is a hardwired gauge and not via any wireless means from OBD2 connector to unit.
The first CSV file ( Log 42 ), I manually turned on the Banks gauge ahead of engine start. So, that file has data a few seconds before engine started.
BELOW
Engine started for 1.134 seconds to 07.407 seconds. Time Code up to refueling stop is engine running minutes & seconds accurate to the dot.
Battery was fully charged the day before and lost some to the video preparation and easy it is still at 90 ish % charged.
Looks like any normal charging: 50 ish amps at very first second or two and then reduced.
By minute 1:40 , charging is down to 15.1 amps. This is inline with Varta Silver Dynamic AGM 15% charge rate of battery capacity as good & gentle.
Later, a very short burst of discharge to the battery, from sudden power requirement of the EPS. Only 0.3 seconds worth of discharge 7.3 to 3.8 amps , good thing it is on video.
.
Alternator being gentle............
.
The Valeo charging profile is a nice slow slope down. Damn....2.9A stable, such a civiized setting from Valeo
.
Yes, 1.4A constantly decreasing from the 2.9A. This is Top Up Territory to nearly a Float Charge. Awesome.
After fuel stop.
Key in ignition but at level 1 / ACC1, stll an insane 10.4 amps dicharge the ENTIRE 3 minute 11 seconds refuelling. ...DUH...
Because the battery was basically fully charged so to speak, before the fuel stop, the Valeo reduced the charging rate so very fast to 5.2 amps within 2.5 minutes after engine start at fuel station. Nice.
.
Below is a slow voltage dip and not up to the point of discharge to the battery as I took a 90 degrees left turn.
Car has forward momentum and tire does not fight the road friction like when a car is stationary while I execute a fast steering wheel angle change.
See 2 photos below .... 2/3 and 3/3 ... how a horn can effect battery voltage. It is an inductive electromagnet device afterall. So it has a good inrush current.
.
For photo 2/3 and 3/3, there was no steering angle change. Its all purely the horn. 1 second burst 3 times with 03. to 0.5 second in between.
CONCLUSION :
I will keep monitoring the charging profile for sometime to come. Plan to keep LIN un-plugged to the alternator for a thousand or so kilometers.
Let see how it goes. The short 50 minutes test does look promising...for me
End of Test.
Last edited by S-Prihadi; 06-21-2021 at 07:12 AM.
Reason: typo
Son of a Bi-etch........
The ECU algo will do operating temperature idling at 730-750 RPM if LIN is disconnected from the alternator and D2 crawl at +-737 RPM.
With LIN connected, it will be 591 RPM idling and D2 crawl at 528-535RPM.
730 RPM is not nice at creeping Jakarta traffic. I like the E400 for being one of the best low speed crawler car I ever had
Let me read more on why the ECU decided so, it can't be because of low RPM and alternator can't pump out enough power.
it is very interesting to get some test results. Do you feel a nice difference with how your engine runs stronger whem supplied with decent voltage??
I didn't know the Hyundai sensor was not the source of voltage display, only Amps are from it.. live and learn ✌️
It remains a mystery to me why the alternator output is commanded off-line by the ECU. This seems to be a dangerous gap in the logic and a missing fail-safe. Searching for solutions...
----- EDIT---- Troubleshooting BATT SENSOR :
Once you're able to reproduce the alternator output going below 12.5V... unplug Hyundai (use the ODB2 to read the voltage situation) - I really wonder if a bad batch of software in these sensor is not the source of all this madness.
The problem is we don't know enough about this sensor to test the software logic. We know it is more than just a data logging probe... what kind of smart does it try to provide?
It is a knock-off version of the similar Bosch sensor.
It could well go in a failure mode when it measures 0Amp for too long post charge cycle and call a discharge logic from untested code.
After having it reset by unplugging, it caused my alternator to give me best charging efforts up to 1Amp under 14.9V (extremely charged up) then to 12.6v float then couple days later back to rollercoaster yoyo madness. Do you see similar behavior dealing with your Hyundai?
Last edited by CaliBenzDriver; 06-21-2021 at 06:07 PM.
Reason: Troubleshooting condition
Cali wrote :
01. Once you're able to reproduce the alternator output going below 12.5V... unplug Hyundai 02. Do you feel a nice difference with how your engine runs stronger whem supplied with decent voltage??
03. After having it reset by unplugging, it caused my alternator to give me best charging efforts up to 1Amp under 14.9V (extremely charged up) then to 12.6v float then couple days later back to rollercoaster yoyo madness. Do you see similar behavior dealing with your Hyundai?
01. In normal condition with LIN plugged to alternator, this <12.5V event/s are short burst and car must be in gear. So can't plug-off the hyundai sensor is not do-able.
02. I was driving so very gentle during that test, I need to see how alternator charging output behave under low RPM. Later I will try Italian Tune driving style and get back to you.
03. I was cleaning the Hyundai sensor few days back too. It does not change its profile though, still whacky YoYo voltage. Battery was fully charged using Ctek, as usual.
Cali wrote : It remains a mystery to me why the alternator output is commanded off-line by the ECU. This seems to be a dangerous gap in the logic and a missing fail-safe. Searching for solutions... I was kinda worried too about un-plugging the LIN may cause dangerous overcharging, but it seems not.
Now, without a LIN and without dedicated B+ from ignition ( only single LIN wire at connector ) it still charges well. I worry it may be parasitic draw when engine OFF, I have to check that.
Alternator may also not show any alert if it is mailfunctioning, since there is no LIN wire to communicate with ECU/ECM.
The idle speed difference is called Dynamic Idling under : Engine on energy management, function - GF5410p1060FL
I guess ECO algo pushed RPM down, to reduce C02 and fuel consumption and since it has the command to alternator, it also has idling RPM authority.
But its funny, why default non-ECO idle RPM so high without Dynamic Idling ? This is not a <2.0 liter engine needing idle that high. This engine has very good low torque.
.
06-21-2021, 06:10 AM
