Euro 550 OBD plug . Yes it is there and I have found it.

Wow, thanks so much for all your time and suggestions! Will try to do this as soon as possible today. (work day has just started, not sure when I get spare time during the day)

 

I had some time in between, to take another look at the back of the engine, this time with a glasfibre remote camera.
I found a connector at the beginning of the valley, underneath the 2 round (injector?) connectors, but as space is really limited, I was not able to look at it entering from the front side of the engine to see what it connects to (valley is really full of hoses and pipes blocking a path even for the fibre camera)
The connection has 2 rows of wires it seems, with 6 on top of 6. I was able to make these pictures, not sure if it is of any help:
Image Unavailable, Please Login
Image Unavailable, Please Login
Next will be measuring the resistance on the Right ECU connector to measure the resistance with the Right MAF connector (correct? Not the left ECU where I get the strange values from?)

 

Ok, just came back from measuring. I connected the multimeter to the Left ECU plug and the Right MAF connector.
All suggested pins worked for all 4 pins mentioned, and all came back with 00.4 or 00.5 , with the meter on the lowest ohm setting (200). Looks to be in order.

Then I thought it to be a good idea to measure the other side as well to see why the checks from yesterday yielded different results for left and right, so I went back to the PC to look a the electrical diagram. What a puzzle! The WSM I have has shattered the diagrams over like 10 pages...
As far as I could see and count, all connections were 1 pin off compared to the other side, so 44 instead of 45 enz.

Went back to the garage, and it did not work: no resistance on any pin. Then I used the pin numbers of the other side you gave me, and those worked!
Did I miscount or is it something else ? Remember , that was the Right ECU side that gives better values from the Left MAF.

I noticed that the "bad" ECU connector and ECU looked a bit dirty, so I cleaned them with contact cleaner and after that Deoxit and again contact cleaner.

A little bit too late now to start the car and test further, but if this does not work, it can only be a miscalculating ECU or a major engine problem like a timing belt jump. But wouldn't the compression test not have shown that?
Any other options that will not directly be multi K of euro / USD?

 

Ian, thanks again for all the time and effort you put into this. I would have been lost by now otherwise.
About the WSM, thanks for the links, I will print them out. Good to know the 355 WSM has the same issue (thankfully my 355 has no issues presently)

About the values my meter shows, when I connect the probes to each other, I get 00.4 , (no idea why it shows 2 zeros before the dot, but one setting larger and the value is 0.00 ( one zero before the dot) so I opted for the lowest possible setting "200" that still shows a number). So the direct measurements I did where not showing hightened resistance on the connected pins, the numbers are in milliohms I guess.

I had missed to understand that for checking for shorting I had to check the other MAF connector pins to the same EU connector pin as well, Will do that today.

I have not swapped the ECU's as I was afraid that it would get me in more trouble, but I can try to do that as well when the wiring has passed all resistance tests.
Does this mean that , when it turns out one ECU is bad, that it actually does not matter if you use a left or right ECU? Makes finding a spare one a bit easier.

I have checked for bent pins on the ECU itself (all good)
I think I have to assume that the connector normally won't have bent female pins, as the smaller receiving pins are not really visible, but I will try.

I had also tried to test the MAF with engine off, ingnition on, scanner attached, by blowing compressed air through it, but that shows no value at all. Too bad.

About air leaks and fuel trims, It looks like the LTFT has returned to normal after a lot of idling. I forgot the value, but below 10%.
But that can also be deceiving I understood, as with an ultra low MAF value which I have, LTFT can't be normal, or another problem is incidentally correcting the value.
The upstream O2 sensor shows a normal pendeling signal between 0.1V and 0.85V in the graph, but the downstream O2 does not show such a pattern. What pattern should I expect for a downstream O2 sensor? Or no pattern at all, just near 0.1V or 0.5V?

I have not checked the exhaust side well for leaks, only visibly, but that is in fact not really valid. can I simply try to do a smoke test starting at the exhaust pipe? Or does a smoke test on the inlet side via the valves work for the exhaust as well?

As for air leaks, the external idle valve system and the fuel vapor system can leak internally to the elbow, correct? Maybe they do not show on a smoke test? Maybe I better redo the smoke test with the Elbows removed, not just with the MAF's removed, see what happens.

 

I downloaded the diagram files, very good, thanks. I can see that it must have taken many, many hours. Great that you share it with the world.
Printing indeed is not really necessary.

Today I did a testdrive after cleaning the ecu contacts with Deoxit, but the test drive was the same as the last few weeks: hesitating, most of the time no power, but sometimes surging forwards while keeping a steady gaspedal etc.
So that clearly did not help, but one must try all simple things first...

Back to testing then, and I tested the resistance the other way round for wires touching, what I forgot yesterday.
First I started with the supposedly good bank: left MAF + right ECU connector. All connections were ok, no 2 MAF pins giving a result on the same ECU pin.
Only on the ECU pins that belonged to MAF pin 3 or 1, I saw a flicker of a value first (like 150, lowest setting again on the meter), before it showed infinity after half a second or so. I guess that is ok? But 1 and 3 behaved different from the other pins, that is for sure, even if they ended with the correct infinity.

Then the supposedly bad bank: Right MAF connector + Left ECU connector.
Here it was clear that the strange values I got yesterday, have a cause somewhere along the harness.
As ECU connector pin 6 gave a low value on 1 (00.4) but 37 on pin 3!
(Same for ECU pins 28 and 34 )
ECU pin 45 only gave resistance on maf pin 2 , so that is ok.
ECU pin 54 gave a value on 3 and 1 , this time pin1 at 37 (milliohm?) and pin 3 the correct low value of 00.4 , so the other way around (pin 1 giving the unexpected high value where it should have given infinity)
ECU pin 17 only reacted to 4 , so that was ok.

So it seems there is something not correct in the harness, but why pin 1 and 3? do they have a switched connection somewhere? or is this the needle in the haystack to find where they touch? On the good banks these 2 pins also started with a value , but that value turned to infinity after half a second, like there was a switch or a diode or something. Is the harness relayed somewhere, with a bad diode?

I knew I should have studied Electronic Engineering instead of stupid business administration and economics, now I can only ask questions and questions ...

 

It was single wire testing: To clarify, on the right bank MAF and left footwell ECU connector, red probe (paperclip actually) I inserted in pin 6 of the ECU connector, the black probe on pin 1 of the MAF connector. That gave a low value as it is supposed to do. But when I , instead of pin 1 of the MAF connector, placed the black probe to pin 3 of the MAF, I got the high value where there is supposed to be no value at all. That high value does not disappear from the meter screen.
Same situation for the other pins 28 and 34, they also gave a correct low value on pin1 of the MAF connector, but a strange high value on pin 3 of the MAF connector where there is supposed to be no value at all.
And on pin 54 of the ECU, which should only give a low value on pin 3 of the MAF connector, this time pin 1 reacted with a high value. So the other way around in values.
So there is a short in the harness that is not supposed to be existing, where pin 3 and 1 connect, correct?
I did not test the resistance between pin 1 and 3 of the MAF, will do that, and I expect there will be a value.

Now I am puzzling with the diagram to see if there is a relay somewhere which might cause this, or another electrical component like the CAT thermo ECU which also uses the same family of earth and power.
Next I will swap the thermo ECU's left to right, and try to check some values there.
As far as I understand from the diagram only the Thermo ECU and a dashboard connector share the same earth and power, or am I missing something?

 

I will test some more with the thermo ECU, but is that the left hand side or the right hand side? If I understand the diagram correctly, the right MAF is via via connected to the right thermo ECU, and only through that special/mysterious plug and cable to the left ECU connector.
So I will need to disconnect the right thermo ECU, correct?
I hope it does not trigger a solid SDL immediately, as that shuts of that bank , and measuring values then is difficult. Or does it not matter for the MAF as it only measures air input, which will be the same for a shut down bank as well as a working bank?
Maybe I better switch the thermo ecu's, to see if the underperforming MAF values follow.
Busy tomorrow, so I will do this on monday.

 

You are completely right of course, no driving needed for testing the resistance! I was not thinking.
I had found the engine earth, it is a wire connecting to the valve cover near the bulkhead, on both sides. Removed and cleaned that earlier, but no effect.

Ok, back from some testing. I disconnected both thermo ECU's , and measured restistance between left hand motronic ECU connector pin nr 6, and right hand MAF connector pin 1 and 3.
pin 1 gave the correct low value of 00.4 , pin 3 gave the strange not-correct 36.5 , so again 70 times higher than just wire resistance, just as yesterday. And this happening on a earth (ECU 6 ) to power (MAF 3), so should be no resistance at all (I think)
Conclusion, the thermo ECU's are not the reason of the strange values.
But I can try swapping them, maybe the resistance is not the problem at all.
Frankly I am getting out of ideas. The strange resistance I can't explain.

Last I can try, if it is not dangerous for the ECU's , is connecting the left MAF connector by means of paperclips+ wire stuck in the back of the connector, to the then disconnected right MAF, to give both ECU's the higher values of the left MAF, and go for a test drive.
That way, it should perform better. I am not sure what it will prove otherwise than that the engine itself is ok, but at least it means not giving up.

If I connect the 2 power wires that way (paperclip in back of connected connector to corresponding pin disconnected connector), is there danger for the ECU's?
I do not know if I overpower the lines that way with 10V instead of 5V, any idea?

By the way, the LTFT values looked ok after the last test drive, at around 4.5, so there is not really a lean condition anymore, that had started with the right fuelpump problems.
I do not understand that when the MAF gives a low g/s, and the ECU gives corresponding less fuel than is necessary for the actual amount of air, that the LTFT is ok...another mystery...
And I still get this pending code: P0102 Mass or Volume Air Flow Circ Low Input

The compression tests I did were ok, with all values between 147 and 152 psi after 4 rotations on a cold engine, so I do not expect real engine problems , which would prevent air being taken in.
So I still look in the direction of something electrical interfering with the MAF values, but it is beginning to look like we are not going to find it without an SD.

 

No, I haven't checked the supplied voltage.
I assume I can check this by disconnecting it, setting ignition "on" and measure between pin 1 and 3. Should be 5V I think. Will try this later today.

I just bought another Autel 519 scanner , hopefully it will arrive this week, so I can see what the actual values are on both banks under the exact same conditions.
Also bought another fuel pressure meter kit, hopefully this one will not leak, so I can do a testdrive with the meter connected and taped to the windscreen, to rule out occasional low fuel pressure once and for all.

About the strange resistance between the earth and the power wire, somewhere something must cause this. Am I thinking correctly that the power wire is the path to follow, not the earth? As the earth is much larger incorporating the chassis and the engine itself.
The thermo ecu was the component that used the same power and earth cable, but there are more components using that power wire through connector 179 as you say.
Any idea where 179 is located? Footwell?

 

Just tried to check the voltages, ECU´s connected, maf disconnected, ignition on, immobilizer disarmed, engine off.
Nothing....but on the good bank, nothing as well...
So I will have to repeat the test with the engine running I guess.

 

Thought some more about it, watched some more Schrödingers Box on Youtube about MAF's (over an hour...) ,and everywhere it was clear that there should be a voltage on it with ign on, engine off, at least 5 V.
I then remembered that the earth of the ECU harness was connected to the valve covers with some small M6 bolts, and that I had cleaned them, but that there still was more resistance than just about the lowest possible when measured between pin 1 maf against the top of bolt. But as the eye of the wire is clamped between the underside of the top and the valve cover, I did not pay much attention to the top of the bolt before.
So now I replaced the bolts with new ones, and they gave the lowest restistance possible on the top as it is supposed to.
And strangely enough I now get 12.16 V between disconnected Maf connector (ign on) pins 1 and 3, and on the other (suspect) bank 12.08 V ...
Don't know why, but something has changed for the better it looks.
As it is raining, 4 degrees Celcius, a test drive is out of the question, but at least the saga continues.

 

Ok, so I could try to make bypass-connection between the OBD plugs with these AMP Superseal connectors, but for now I think the one in the car is working well.
Or at least it is not connected when the scanner is, so as far as I can see it, not the culprit of the low MAF values.

Today I blocked the metal tube going into the elbow after the MAF, to check if it was leaking inside ( Engine vapour being succed in when the EGR valve (or similar in this car) is stuck open.). No difference, still a MAF value of around 8-9 g/s at 2400 rpm, where it should be twice that. For good measure I recleaned and replaced it with the good spare MAF again, no difference

LTFT still around 4.5 , so that is confusing with such a faulty low MAF value

Only thing left to do is swap the ECU's, to see if the problems transfers to the other side. Still a bit hesitant , afraid of destroying a good ECU...

And rechecking the fuel pumps, this time for capacity and not pressure. I noticed that when I drained the tanks earlier by connecting a hose to the Schrader valve, that the fuel took around 12 minutes to fill up a 5 litre jerrycan. Which I thought was very slow, but contributed that to the schrader connection pinching the flow.
But maybe the flow is just to low? Physics laws demand that pressure is related to flow, so I think that it is a dead end to investigate when pressure is good.

Another thing that I noticed today, on the OBD scanner, the throttle position sensor (left ECU, so right MAF and left TPS) was 0% up to around 2000 rpm.
You could feel a sort of barrier in the gas pedal around 1800-2000 when going from idle to 2500 rpm (standing still). When the rpm rose easier after this barrier, the tps percentage rose from 0 to say 4%.
At first I thought that this could be related (low MAF then actually true because the throttle was not opening), but the low MAF values are from the right bank where the TPS is from the left bank.
And it does not explain why the car is lacking power half of the time.
Another dead end, or not?
Any thoughts are very welcome as I spent like 40-50 hours already on it, but am about to give up because there is not much to investigate left....