Mercedes-Benz Forum banner

Troubleshooting Ingition Module/Coil

14K views 38 replies 4 participants last post by  forrera 
#1 ·
My W168 is missing on cylinder #2. I'm looking for information on how to troubleshoot the ignition module/coil before I purchase a new one.

I recently replaced the starter using the top replacement technique. After reassembly, cylinder #4 was missing as per ECU code P0304 (I believe). I reattached the module connector to ensure a good connection, cleared the code and then it switched to missing on cylinder #2 with ECU code P0302. I again ensured there was a good connection to the module a couple of more times and each time I start the car is comes up as missing on #2 (P0302).

I replaced all four spark plugs but still code P0302 is coming up.

Can anyone assist with troubleshooting procedure or point me to a resource online with information?
 
#2 ·
Hello,

Did you check the ignition coil package? If the cable and connections between the ECU and the ignition coil package are good enough then you can think about the ignition coil package, or ECU.

In your case I think you have bad connection somewhere. Either in the jack from the right side of the ignition coil pack (front side view) or in some of the spar plugs. Also the wires from the ECU to the ignition coil could cause sporadic problems if they have stripped insulation or cut somewhere on the path.

Try to dismount the ignition coil pack and assure yourself that each spark plug is tightly fitting inside the coil slots.
 
#4 ·
Thank you both for replying!

I will check those suggestions.

I looked on WIS for information about resistance values I could verify on the coil pack or from ECU but I don't see any. While going through the suggestions above, are there any values I should confirm?
 
#6 ·
can you tell me if your ecu is later smaller one without air mass meter?
I'm not certain what "later smaller" means, perhaps later model, smaller...

As far as I know it's factory original on this 2002 W168. It bolts onto the intake and the air mass meter slides on with a large rubber boot (See image).
 

Attachments

#7 ·
I tried to upload pictures here but couldn't.
However, I managed to upload the pictures onto my facebook page with details of the fault, how to check and fix it. forrera a class.
 
#10 ·
I used my metre and verified all three contacts from the coil pack to the ECU are good. As well, I was able to verify all five wires going to the four fuel injectors to the ECU are good.

I located a used coil pack at a junk-yard. I installed it and still cylinder #2 is missing.

I started the car with cylinder 2 injector plug removed and verified there is 12v+ going to it but not a pulsed 12v-, like cylinder 1 has. Instead, the negative terminal was a very weak 12v+ that never flickered. While the car is running, when I plug cylinder 2 injector connector in, it runs a little smoother so it is sensed. I put my tester between the two connectors and it does nothing.

So now I have new spark plugs, different coil pack and good wiring to ECU, but the injector isn't receiving a pulse.

I guess I could swap injectors and see if the missing follows the injector or stay on cylinder 2, but I don't have the eight $13 O-rings needed once all four are pulled. I've read once they're pulled, you should not reuse them.

Is this my last option?
 
#12 ·
Hello,

I see you are suspecting the injectors, but have you had some DTC error related to the injectors when you connected the Star diagnosis? I mean if you had faulty injector the diagnostic tool should catch it as one of the following :

P0264 Cylinder 2 Injector Circuit Low
P0265 Cylinder 2 Injector Circuit High
P0266 Cylinder 2 Contribution/Balance Fault

You are saying that you have stable 12V to the positive terminal of the injector, but how you are checking the ground pulses on the negative terminal?
With regular multimeter switched on DC gauge, I think you can not catch it. When engine is idling at ~1000rmp the frequency of the ground pulses is at least 500 times per second, which I think is too high frequency for multimeter at DC gauge (it is almost AC, since we have electrical transition processes between the pulses). I think the only tool useful to see these pulses is oscilloscope.

I will try to look for the procedure of manual testing the injector circuitry, provided by MB, and post it here. Also will try to find the firing diagram for this engine. May be they will be helpful during the debug.
Also I think that the spark plugs are working in pairs. Simultaneously two plugs are firing during different phases of the engine running, but may be I am wrong - will check.
 
#13 ·
I see you are suspecting the injectors
No, not the injector, it's the signal to the injector.

You are saying that you have stable 12V to the positive terminal of the injector, but how you are checking the ground pulses on the negative terminal?
I used a LED test light for the injector. I can see the pulses on cylinder 1 injector plug but only a solid very dimly lite LED on cylinder 2.

I will try to look for the procedure of manual testing the injector circuitry, provided by MB, and post it here.
That would be great, thanks!

Also will try to find the firing diagram for this engine. May be they will be helpful during the debug.
Also I think that the spark plugs are working in pairs. Simultaneously two plugs are firing during different phases of the engine running, but may be I am wrong - will check.
I can say cylinder 1 & 4 and then cylinder 2 & 3 work as pairs. I don't know if both in the pair are fired at the same time or not.

Thanks for you assistance!
 
#14 ·
I gave you the answer a few days ago as I've had this problem every time I replaced a starter on a car 2002 onwards.
the switching wire to injector number 2 is a little shorter than it should be and eventually breaks internally.
I can't upload any more pictures here.
I put the picture on facebook, aclass forrera, the picture with 2 plugs.
the small plug is for injector number 2 and the bigger one is the ECU lower plug.
the meter is connected to the relevant pins to check continuity between injector and ecu.
 
#18 ·
meter is connected to the relevant pins to check continuity between injector and ecu.
I was able to verify continuity to all five wires. Perhaps the wire's break occurs when everything is in normal position and moving it re-established the continuity. I'll dig a little deeper on this and perhaps replace the wire even if it continues to show continuity.

Is your "Check engine" light illuminated on the instrument cluster?
It is and the code comes up as P0302.

Check the resistance between the following pins. You should measure 14 to 17 ohms in each case.
Will do before replacing the wire and yes, put new spark plugs in and tried a second coil pack.

Thanks for all the input! We'll see how I go...
 
#17 ·
I've just had a look through my WIS for your engine 166 problem. First I would do a resistance check on the 4 injectors.

1. Remove connector 2 from MSM (N3/5).
2. Check the resistance between the following pins. You should measure 14 to 17 ohms in each case.

a) pin 38 and pin 1 (Y62y1 is for cylinder 1)
b) pin 38 and pin 2 (Y62y2 is for cylinder 2)
c) pin 38 and pin 3 (Y62y3 is for cylinder 3)
d) pin 38 and pin 4 (Y62y4 is for cylinder 4)

If they are all OK, the problem is probably with the MSM - if you have already eliminated plugs and Ignition coil module.
 
#20 ·
Wowww... brave turn!

I don't think you will see any obvious signs of damage there. Usually the burned out part is the transistor which operates the pin. It is either CMOS or bipolar type (In our case I bet on CMOS).
You can try to trace the path from this pin to the components connected to it and find this transistor, but I am not sure if it is possible. Is the board multi-layer or is just two side cooper. With the multi-layer it will be almost impossible.

But before disassemble the ECU did you check the actual injector, I mean the mechanical part of the injector. To see if it is spraying correctly. May be it is clogged or damaged.
 
#22 ·
Is the board multi-layer or is just two side cooper. With the multi-layer it will be almost impossible.

But before disassemble the ECU did you check the actual injector, I mean the mechanical part of the injector. To see if it is spraying correctly. May be it is clogged or damaged.
I was able to open & access the circuit board but I could not remove it without damaging it so I can't see the back of it. I just put it back together after taking several photos.

As for the injector actually working, I have not removed any of them. In my mind, everything points to the ECU not outputting pulses. However, I can swap with another cylinder to see if the problem follows the injector or not.

This is how the injector works.

I doubt that you can see the brief pulses properly with anything other than an oscilloscope.
WOW, what software is that? Must be a new version of WIS? The WIS I have is an absolute nightmare to use...

Anyway, I have an oscilloscope so I will try it this weekend. However, I can see the pulses for cylinder #1 . They're very bright with full on/off blinks at idle, whereas #2 has none.

Thanks again for all the input!
 
#23 ·
Burnt Resistors?

Upon second inspection, it looks four resisters may be burnt (Outlined with blue boxes in attached image)?

The colour bands are not well defined and some are blurred a dark brown.

I'm going to check there resistance to see if any remain. I'm also going to put a heat-gun to the board in-case there are any cold solider joins.

Then I'll put everything back together and if still missing, check the output with an oscilloscope.
 

Attachments

#24 ·
Interesting observartions. These components could be diods as well. Check it for correct ventil operation.
But i don't think only resisyors could cause eventual problem with the ECU. Usually the burned out are transistors or entire ICs.
The osciloscope approach will show the situation as it is.
 
#26 · (Edited)
Here is small information for the ignition processes in this ECU. I promised to find such a document.
Unfortunately the images are too small to read. Do you have a larger version?

Interesting observartions... These components could be diods as well. The osciloscope approach will show the situation as it is.
I checked them with meter and they all still have resistance.

I put the oscilloscope on cylinder 1 & 2 . There is a noticeable difference as cylinder 2 isn't getting the pulses.

However, I can see the pulses without an oscilloscope. I created a short video (Convert to .gif file and attached), of cylinder 1 to demonstrate.

Interestingly, while I had injector 1 disconnected, to start recording the pulses, they stopped. I was worried I may have damaged the ECU output to cylinder 1! I plugged the injector in again and 1 did not start firing again.

I turned the car off, checked the codes and there were three this time:
  • P0302 - Misfire 2 (This always comes up);
  • P0202 - Injector 2 Circuit (Haven't seen this one);
  • P0301 - Misfire 1 (This is new but expected).

Interestingly, I do not get P0301 (Cylinder 2 Misfire.) I cleared the codes and started the car again, 1 was firing normally... Whew! I quickly disconnected and created the video before it stopped again.

Given 1 stopped the way it did, I was hoping 2 may have stopped because ECU doesn't see its injector. So I swapped 1 & 2 injectors but the problem did not follow the injector, 2 continues to misfire.

With everything I have gone through, no matter what I test/swap/replace, the issue points to ECU 2 output.

During the week I called about repair/replace ECU and the best price I found in Australia was $150 to inspect, with shipping. If repair is needed, $1100 with shipping.

Looks like I'm left with worst-case-scenario...
 

Attachments

#32 ·
I sent my ECU to Australian ECU Repair and was helped by Justin. Initially he tried to restore my ECU but couldn't ensure the restore would work as he cannot verify cylinder 2 without the car there. He located a second-hand unit, programmed it and sent to me for $500.

I installed it and the problem continues. I rang him and he suggested I solider a wire from the injector to ECU, bypassing the factory wire, just as forrera suggested. He explained, even if there is continuity, it may be due to a strand or two still unbroken within the wire, however this isn't enough to fire the injector.

I went back and changed the way I am using the test-light. Instead of keeping it under the hood and running a hot to it, I moved it to the battery and ran a wire from the injector connector. I cranked the car with the tester on the battery positive and the ECU is sending pulses for about 2 seconds. I never saw the two seconds of blinking before as they stop before I can get under the hood to check it. So, it looks like I just wasted $500.

Justin also suggested this may all be due to a compression issue but I don't know how to ensure this. I have used an OBD II app via Bluetooth and the intake pressure is ~60kPa while running and 102kPa with the ignition on.

Final note, one code continues to be P0302 but there is a second code now, P0202 (Ignition Circuit - Cylinder 2). The only other time I have seen this one is when I injector 2 & 1 disconnected.

Unless someone can chime in with another suggestion, I believe I'm now to the point of taking the car to a MB mechanic as this could have been sorted by now and for potentially less money.
 
#33 ·
Well, I got it running... and it was in-fact the GRY/BLU wire going from injector 2 to the ECU.

After speaking with Justin again, he suggest soldering a wire between the two, just as forrera suggested three weeks ago. It sputtered initially but then ran smooth. I took it for a spin and it seems like it is running better than it used to.

Thank you everyone who chimed in. If you need assistance with your ECU in Australia, give Australian ECU Repair a go. Justin was very helpful!
 
#36 ·
After driving the car for a few weeks now, the missing on cylinder two returned last week. Quite frustrating!

I removed the harness from the ECU & injector again, tested continuity (All good) and put it back together again. It's no longer missing but I imagine the problem will return as it may be the connection at the pins with the wire at either end.

I think I am going to just replace the car as I have to have more dependable transportation.
 
#37 ·
The car still hasn't started missing again but the check engine light is coming on quite a bit. The code is P0202 (Cylinder 2 Ignition Coil Circuit) previously P0302 (Cylinder 2 Miss Firing) was the code.

I can't replace anything closer to the ECU, should I try replacing the injector connector?
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top