3 phase in-tank fuel pump - Mercedes W212 Petrol

Ask for and share advice on using the PicoScope kit to fix vehicles here.
Post Reply
S-Prihadi
OneWave
OneWave
Posts: 10
Joined: Thu Dec 30, 2021 4:59 pm

3 phase in-tank fuel pump - Mercedes W212 Petrol

Post by S-Prihadi »

Hi All,

Thanks for the excellent post by Steve viewtopic.php?p=100521#p100521

I would like to add some information and also seek advice.
I am a complete newbie for scope use :oops:

Information on the 3 phase BLDC pump in this post :
https://automotivetechinfo.com/wp-conte ... hicles.pdf

This is a returnless fuel system. So variable speed of the pump is they key.
Mercedes specify fuel pressure to be minimum 4.5 BAR and maximum 6.5BAR.
This booklet at page 23 - 27 has good information on the low-pressure and high-pressure fuel system : https://mbworld.org/forums/attachments/ ... e-info.pdf


I tried capturing the waveform , but on a 4425A 4-channel only.
The car model is 2014, W212 E400, 3.0 V6 Twin-Turbo GDI. Fuel pump Mercedes P/N is A218 470 00 94
The pump controller is by Continental, MB P/N is A 000 900 17 04 and is called N118 in MB Xentry Diagnostic software.

Since I can't attach files (yet), I can't include the wiring diaghram for N118 or the waveform capture.

SEEKING HELP :
I tried using Fluke 289 DMM Duty Cycle feature to read one of the phase, but since this is a 3 phase pump,
I dont think it is accurate ... yes ?

Autel or MB Xentry does not show duty cycle too, but will inform variable voltage of/to the pump.
At idle 600ish RPM voltage is 7 volt and 45% Duty cycle-Fluke. When pushed to WOT it can go near 10 volts but duty cycle-Fluke is 59% max. I am assuming these variable voltages is like a duty cycle equivalent ..sort of ?

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

My actual intention is to create a database on fuel filter expected time to replace before any minor clogging can effect WOT , based on fuel pressure and the duty cycle/voltage values as preventive maintenance plan.

Fuel pressure is easy to monitor as the fuel filter has fuel pressure sensor, but the duty cycle or variable voltage to pump, I do not know yet what is the voltage* ( *Xentry /Scanner ) for a 100% actual duty cycle.

I am very familiar with year <2009 600-1400HP ( 2,300 RPM Max) marine diesel engines fuel monitoring system on yachts, the low pressure side where a lift pump is used, so suction mode it is on the filter element and not pressure mode like this W212. The Racor FG1000 series fuel filter has vacuum gauge and the lift pump is pure mechanical, so by vacuum value we can tell amount of blockage the filter element has at cruise speed, usually 2,100 RPM. It was so easy to do preventive maintenance.


Any advice is much appreaciated with thanks.


PS - I have also scoped out the Bosch 200 BAR pump Quantity Control Valve. It will be for another post.
The HP pump is this family : https://www.bosch-motorsport.com/conten ... 98763.html

S-Prihadi
OneWave
OneWave
Posts: 10
Joined: Thu Dec 30, 2021 4:59 pm

Re: 3 phase in-tank fuel pump - Mercedes W212 Petrol

Post by S-Prihadi »

EDITED , my bad..I have MS One-Drive : https://1drv.ms/u/s!Ahrq0GmYQf8dghwpWVB ... J?e=EN1shM
The waveform is not during a moving test drive, but a stationary one with some engine rev up.


The one below I can not scope it during a test drive. So no waveform.
"Autel or MB Xentry does not show duty cycle too, but will inform variable voltage of/to the pump.
At idle 600ish RPM voltage is 7 volt and 45% Duty cycle-Fluke. When pushed to WOT it can go near 10 volts but duty cycle-Fluke is 59% max. I am assuming these variable voltages is like a duty cycle equivalent ..sort of ?"

Steve Smith
Pico Staff Member
Pico Staff Member
Posts: 1581
Joined: Sun Aug 25, 2013 7:22 am

Re: 3 phase in-tank fuel pump - Mercedes W212 Petrol

Post by Steve Smith »

Hello and thank you for the post, this sounds like an interesting adventure as ultimately, we can predict and so avoid a potential breakdown

Regarding the capture of a single-phase reference to phase ground of the 12 V BLDC motor your Xentry 4425A will return either a duty cycle measurement or graph the duty cycle for you using Math channels.

We have used this principle here topic22767.html on an EV PMSM

With that said, prior to going too deep, if you could post your captures (from the Xentry 4425A) and the connection/wiring diagram that would be awesome.

If you are having issues posting, please email support@picotech.com and refer to this forum post where we can view the captures and return here to share the outcome.

There may be alternative methods to determine pump activity/performance in response to filtration using current, PWM signal from PCM to pump controller, CAN, fuel pressure and even sound via NVH (We have used this NVH technique here viewtopic.php?p=99974#p99974)

Regarding, “My actual intention is to create a database on fuel filter expected time to replace before any minor clogging can effect WOT, based on fuel pressure and the duty cycle/voltage values as preventive maintenance plan.” this sounds like a great project and one we can help you with using the variety of measurement features of our Xentry/PicoScope

I hope this helps and I look forward to your feedback

Take care…….Steve

Steve Smith
Pico Staff Member
Pico Staff Member
Posts: 1581
Joined: Sun Aug 25, 2013 7:22 am

Re: 3 phase in-tank fuel pump - Mercedes W212 Petrol

Post by Steve Smith »

Thank you for the psdata file and the connection diagram, I have taken a look at the data and focusing on a single phase (Channel B, Pin 4 Phase U) our positive duty cycle = 48.94% during the rise of DC current (Channel A Pin 1 +12 V) to a peak of 4.086 A during the free revving period. See the image below:
Image 1
Image 1
You can use this technique to compare with other areas of the waveform at idle and at WOT under full load.

You could also add into the mix, fuel pressure values (priming) and pump speed using math's in order to profile the pump performance characteristics with a known good filtration system.

It would be interesting to profile Phase current against DC current as comparing both may well give an indication of “losses” within the system.

I certainly think a pump health check is a valid initial test to ensure phase current and phase voltages are as expected, i.e., phase relationship with one another and uniformity within

One area to consider with such return-less fuel systems is “system blockage”; how does the pump/system handle a blockage?

A number of VM’s utilise a relief valve within the supply line, such that a system blockage upstream (i.e., fuel filter) will result in the opening of the relief valve and the return of fuel to the fuel tank (often handled inside the fuel pump assembly)

In such a scenario your pump speed and current draw may indicate “normal” yet fuel priming pressure is low given the pump is simply drawing fuel from the tank and retuning “back to tank” via the relief valve.

You could simulate a blockage using a stop tap inserted into the fuel line and monitor/plot the behavior of the pump/circuit in relation to a momentary blockage (“dead head”)

It’s all food for thought, I hope this helps, take care…….Steve

S-Prihadi
OneWave
OneWave
Posts: 10
Joined: Thu Dec 30, 2021 4:59 pm

Re: 3 phase in-tank fuel pump - Mercedes W212 Petrol

Post by S-Prihadi »

Thank you so much Steve.

I have to learn more on the math channel use and apply your thought process as a good advice on how I may be able to predict the fuel filter life...awesome.

For now I shall include how Xentry or Autel shows the available data perimeters coming out of N118 fuel pump controller for Low Pressure pump and the engine computer N3/10 for High Pressure pump. In video form. The only way I can "sync" the Fluke with the Autel/Xentry with ease.

There is a minor lag for Autel or Xentry data compared to the car's tachometer during a rev up or WOT pull.

VIDEO 1
Car is stationary. Idling and a rev up* ( *the rev up waveform I linked , this is the partial video )
https://youtu.be/GhjEdGU5CNo


VIDEO 2
Test Drive. No waveform. Only video. Even on a Sunday the paid highway was so packed, I can't even do proper 1st and 2nd gear pull to WOT. :shock: The best is to get proper WOT pull from 1st>>2nd>>3rd gear to see fuel flow parameters, but that will need a late nite test.
https://youtu.be/_qYrFTEhS90


The high pressure pump Total actuation angle of quantity valve is the next one for the study. Also to predict its wear and tear before any DTC will be triggered by engine computer. I will post separately soon hoping your experience can assist us.

Thanks again....

S-Prihadi
OneWave
OneWave
Posts: 10
Joined: Thu Dec 30, 2021 4:59 pm

Re: 3 phase in-tank fuel pump - Mercedes W212 Petrol

Post by S-Prihadi »

I will report back when I can scope the fuel pump controller while doing 1st and 2nd gear WOT.
I think that is the only way I can at least "correlate" the VOLTAGE OF FUEL PUMP as reported by Xentry to actual duty cycle using Pico waveform.

The newer AMG version E63S 2019, has better N118 fuel pump controller version and has duty cycle data...
https://youtu.be/cWIB4Xd1k10

S-Prihadi
OneWave
OneWave
Posts: 10
Joined: Thu Dec 30, 2021 4:59 pm

Re: 3 phase in-tank fuel pump - Mercedes W212 Petrol

Post by S-Prihadi »

I gave up trying to "correlate" the VOLTAGE OF FUEL PUMP as reported by Xentry to actual duty cycle using Pico waveform.

It is also a weird thing that Xentry shown amperage of the fuel pump , is not the same as my measurement on the fuel pump main power input wires.

I got confused or misled by "bad" information from this document :
https://automotivetechinfo.com/wp-conte ... hicles.pdf

There is no such thing as my fuel pump ( E400 and not E350 as tested by author) able to do 8,000 RPM.
Also there is no such thing as power consumption of the fuel pump at even higher than 7ish amps, let alone 23 amps... LOL.


This is the final test using pressure comparison between new and old fuel filter and I got to measure the built-in static pressure or head loss of at 1.1 BAR from the delivery pipe between the fuel filter and the HP pump low pressure input. The LP fuel pressure sensor is at the fuel filter output port (at the fuel tank) and from there to the engine is approx 4-5 meters of pipe and a few bend/elbow and elevation is approx 60cm higher at engine side.

a1.JPG
.
a2.JPG
.
a3.JPG
.
a4.JPG
.
a5.JPG
.
a6.JPG
.

The SUMMARY, 3rd gear is not included
a8.JPG
.

The static pressure produced by the fuel lines from fuel tank to engine bay
a16.JPG
.
b6.JPG
.


I plan to fool the N118 fuel controller with a signal generator by under-reporting the fuel pressure by 3BAR, hoping to see if N118 can push the fuel pump to its maximum power. The signal generator has not arrived yet, it is from China AliExpress, some screw up with the shipping company and I got a wrong package instead...Dugghh.

I will update when I have more information.

S-Prihadi
OneWave
OneWave
Posts: 10
Joined: Thu Dec 30, 2021 4:59 pm

Re: 3 phase in-tank fuel pump - Mercedes W212 Petrol

Post by S-Prihadi »

For those wanting to see a more complete description/workings of fuel filter system of W212 E400, I don't want to waste this forum bandwith by posting here.

You can read here : https://mbworld.org/forums/e-class-w212 ... hange.html

It's link to another post ( Main original post) : https://mbworld.org/forums/e-class-w212 ... ystem.html
also covered the HP fuel pump I am trying to understand better but unable to do so :oops: ...yet.

Happy troubleshooting a W212 V6 petrol engines fuel system...if and when you get one.

Steve Smith
Pico Staff Member
Pico Staff Member
Posts: 1581
Joined: Sun Aug 25, 2013 7:22 am

Re: 3 phase in-tank fuel pump - Mercedes W212 Petrol

Post by Steve Smith »

Hello and thank you for the continued update.

I can see a huge amount of time & work has been invested in this investigation/research and I admire the fact you have not given up. There comes a point where you have just come too far to stop now, nice work.

I have no doubt this will help someone somewhere and your knowledge of these systems will have grown beyond those who actually designed it as you are testing in the real world under "live" conditions

This appears to be the case with your comment surrounding fuel pump current and I guess we must take everything we read with a degree of skepticism until we prove otherwise

The next step looks intriguing too re PWM manipulation and the effects upon pump output/consumption

Thank you again for posting, take care.....Steve

Post Reply