348 Persistent Misfire

Hi guys,

Firstly I'd like to thank everyone for all the great advice on this site that has helped me to keep my 348 on the road for the last 8 years

My car has been sat for the last 2 years, so I decided to give it the major service before putting it back on the road. Timing Belt, Cam shaft seals, Spark plugs, rebuilt the fuel pods with new Bosch fuel pumps, and connected new fuel filters too.

On completing the work and starting the car, I noticed a misfire that comes & goes in severity, but is always present. To discover which bank, I pulled the fuel pump relays. Pulling the LH fuel pump relay makes the engine run smooth (so the RH bank is running ok). Replace this and pull the RH fuel pump relay, the engine struggles to continue running and will stall if the relay isn't replaced immediately, so the misfire is on the LH bank. Here are the things I've done to try and fix it so far...

I've removed the spark plugs (all look wet & pretty black on the LH bank, but the car hasn't run - it's just stood and misfired, so I kind of expect them to look like that). I've tried cleaning them up and replacing them, but it makes no difference.
Swapped the LH spark plugs for the old ones that never misfired before, but the problem persisted.
Removed Spark plug leads, and tested the resistance. One was dead, one was 12Kohms, the rest were around 6Kohms.
I bought some new 7mm HT cable and re-constructed all 8 leads. Now all the leads are 6Kohm apart from the 12Kohm one and that is now on the right bank 1-4 which is running great.
Swapped the coil packs from LH to RH, but the fault stayed with the LH bank.
Swapped the electronic control module from LH to RH but the fault stayed with the LH bank.
I've pulled the LH bank HT leads with the engine running, and thought the spark seemed a little weak, but I got a damn good belt off it too!
MAF sensors read the correct resistance within 3 ohms between pins 1 & 6, but I haven't swapped them over.
Cleaned all electrical connectors with contact cleaner and WD40.
Jacked the car up to inspect the crank angle sensors. The 5-8 bank sensor plug crumbled into dust when I touched it, so I replaced it with a new sensor. I also pulled the sleeve back to inspect the wiring which all looked clean & in excellent condition. I disconnected the 5-8 crank angle sensor and attempted to start the car, but it wouldn't start (which told me it was working ok), as it will start when connected, it just runs very rough, and I even tried swapping them over, but the fault stayed on the LH bank.
Checked the flow rate of the LH fuel pump - it fills a 1ltr coke bottle in less than 10 seconds with a continuous stream of fuel, so I'm thinking that's ok...
Pulled the LH injector manifold, removed all the injectors and cleaned them. Also powered them up outside of the car to see if they operated ok. As far as I can tell, they all seem to operate correctly outside of the engine, but I don't have any equipment to see how well they are performing. I can tell you that they open and close as I can see them spray the little amount of fuel / cleaner that was still inside them during testing.
I've also checked the pressure regulators on both sides - no fuel in the vacuum pipes, and a vacuum is present on both when the engine is running, but I haven't tried swapped them over.

I've found a link on the forum for making a switch to get the codes off the ECU and that's my next job, although I suspect it will just tell me I've got a misfire. I've also found the primary & secondary coil resistance values for the coil packs which I will also check this weekend and replace if necessary.

I've read everything I can find relating to 348 misfires on this forum, and I believe I've tried everything so far.

If you can think of something I haven't tried, or can see a mistake in what I've tried to date, please enlighten me....

One final note (probably not related): My headlights have also recently developed a fault. They come on, but they don't pop up anymore (never had this problem before). Shorting out the relays under the hood makes them both blink, so I know the motors & mechanisms are ok. The light lifters relay in the foot well doesnt operate the lights (even when the relay is removed and the contacts are manually shorted out). I'm guessing there's a bad ground somewhere, so I'll be checking all the ground points this weekend too.

Any suggestions/help will be greatly appreciated...

Thank you

 

Holy crap! I commend you on your detailed first post. Someone here will help very soon. Call all stooges! Good luck.

 

On the headlights, this is pretty common on cars that have sat for awhile or not had their doors operated. The motor contacts get carbon on them. There is a thread on this here about cleaning the contacts from under the car, I think. In my case, the problem went away with continued operation of the headlight doors.

For the engine sputtering, I didn't get the impression that your left hand side is getting enough energy. If you pull a spark plug from the left and crank the engine (make sure it doesn't start) do you see a nice blue spark?

 

I have had 2 occasions of misfire. first i put down to slight wear on 2 cam lobes (which was rectified in the engine out service).

Second time was due to failing alternator and was intermittent until it totally failed when misfire got bad and car would not re-start. Check your charging system is putting out 14+ volts whilst engine is running

 

A couple of suggestions:

Swap the MAF-s (also inspect the MAF connector pins for any corrosion)

Try swapping the ECU-s or unplug and plug-in again the ECU for the bad side (there could be some loss of contact at the pins due to long sitting)

 

Do the plugs still look wet & black, even after all that you've done? If so, you might check the coolant temp sensor for that bank, along with its wiring. It's possible that it that a problem there is causing the ECU to think that the engine is very, very cold, and it is therefore enriching the mixture to compensate.

 

I had a very bad mis fire, i changed and checked the usual items for weeks. Then i discoverd the problem.........the Lamba/ oxygen sensors. Seriously, one of them was sending bad infomation to the ECU. A bad sensor gives very bad misfires at both tickover and under throttle load.
Change them

 

Another vote for MAFs.

Swap them and see if anything changes. I chased down an intermittent issue like this, with everything from ECUs, Oxygen sensors, thermostats, fuel injectors- you name it. Finally swapped the MAFs and low and behold, the problem moved.

In my case, the measured resistance was still "correct," so I would not rely on that as a measure of proper function.

Good luck. Although I was frustrated during the process, I corrected a lot of minor issues and the final result was worth it all.

 

Did you make sure you used shielded wiring for the leads? You need shielded wire to help block RF interference.

 

Just a small clarification: The MAF potentiometer (between MAF Pins 1 & 6) has nothing to do with the MAF, except that it is physically there. It is actually connected to the ECU, i.e. part of the ECU circuitry.

 

There's definitely more to it.

I just wanted to emphasize that the 383 Ohms reading, which is mentioned frequently in threads about potential problems, tests only one aspect of a "good" MAF, but in itself, does not indicate that the MAF is functioning.

For the reference of any future readers, the general Bosch Motronic Fuel Injection manuals have a pretty thorough overview on the theory, testing, and adjustment process of the entire mixture system. Lots of variables, and even more when you factor in the age of our vehicles, leaks, poor connections, etc.

 

To check the MAF for proper operation (in the mixture system), there is only one function to verify: the signal (varying voltage) coming out of MAF's Pin 3. Of course, other pins should have good contacts. The function of the pins is as follows:

1 - Ground
2 - Ground
3 - Signal to the ECU - represents the mass of air flowing through the MAF
4 - Input (current from the ECU) for self-cleaning of the MAF after engine switch-off
5 - +12V ignition switched power supply to the MAF's circuitry
6 - Potentiometer output to the ECU - The Potentiometer is connected between the ECU Pin 43 and the ground; it actually belongs to the ECU but it is placed inside the MAF for ease of access, otherwise it could be placed anywhere.

 

Thanks for everyone's input...

ghardt - There's not much that's 'holy' about this problem! ;-)

vvassallo - I've not played much with the lights yet as I'm still trying to fix the engine. However, two relays under the bonnet when removed and the contacts shorted out will lift and lower both lights (without stopping until disconnected). I've operated the lights several times like this, but they still don't operate from the light stalk. I'll go around the contacts once I've got the engine running.

Christof - The voltage at the terminals when running is 14.58V so I think the alternator is charging OK. As for the lobes, the engine has less than 40,000KM and I had a good look at them when they were out for the cam seal change. They looked in very good if not excellent condition to me.

m.stojanovic - I swapped the MAF's over and the problem stayed on the LH bank. There is no significant corrosion on the connectors that I can see.

///Mike - I haven't had the chance to try this yet, but I will investigate this as soon as I can.

Jackie and bill - I forgot to mention in my last post that I also swapped over the 4 connectors under the airbox for the O2 sensors, but the problem remained on the LH bank.

ernie - The HT cable I purchased is a solid copper core, 7mm OD cable, and is exactly the same construction as the old cable that I removed (original from manufacture). In the dark, there is no visible tracking that I can see, and no interference on the radio which there would be if I had any EMC issues with the leads.

Now for an update:
I measured the coil packs against the Bosch data sheet and found one (the LH bank) to be out of spec and actually open circuit on the HT side, despite it still operating correctly (when on the RH bank). I replaced both coil packs, put my old spark plugs in the LH bank to block the holes, put the new spark plugs in the ends of the leads, removed the LH fuel pump and started the car. What a show! All 4 sparks with a nice blue spark Put them back in the car and guess what! No! the problem wasn't fixed. The engine still sounded like it was misfiring & struggling to run.

Fowled plugs and a ECU power down and reset I think. Plugs all clean and replaced, Power back on ECU's and I found the following:

Misfire gone - but still struggling to run on the LH bank.

Revving the engine sounds great, no misfire at all. I have even tried revving the engine with both LH and RH fuel relays removed (one at a time), no problems during revving, but LH bank really struggles to idle. So, I tried removing one HT lead at a time to find the faulty plug. Each lead stalled the engine immediately, so they are all sparking but the engine still runs very rough on the LH bank (but it does run... just).

The plugs I have in the car at present are NGK DPR8EIX-9. Has anyone else had issues with these plugs...? They do work perfectly on the RH bank...

The best way I can describe the engine is a fast hunt when running on the LH bank only. Vrm Vrm Vrm Vrm about twice a second about as low as the engine can idle without stalling. The RH bank just purrs gently & smoothly.

I opened the idle adjuster screw on the LH bank to see if the engine was struggling for air, the RPM picked up a little but the uneven running continued. This was done with the RH fuel pump removed so only the LH bank was running.

I tried disconnecting the throttle position sensor on the LH bank and the RPM changed and nearly stalled until it was replaced. Just for good measure I swapped the two idle control valves over as well but the problem persisted on the LH bank.

I'm going to look at the temp sensors tomorrow (if it's not snowing). Apart from that, has anyone got any ideas for this continuing problem...?

 

How are your grounds? The grounds connected to the valve covers and the ground connected to the subframe.

I would also have a look at the barrel connectors for the injectors and coolant temp sensors. They are located up under the rear of the intake and unscrew counter clockwise. Check the condition of the pins and wiring.

 

Have you check the injectors firing properly with the correct amount of fuel? Years ago I work on a 348 that had 2 bad injectors causing the car the sound like an american V8. You will need a bosch injector pulser to see if the spray pattern is the same on all 4 for the LH bank. Plus you will need a beaker cup to ensure the same amount of fuel is exiting each injector.

Otherwise you may have an intake leak causing the LH bank to run lean and increase in rpm.

 

Glad you reported back, Mike. I really like your diagnostic procedure so far, as well as the thoroughness of your posts. Sure makes trying to follow along and offer useful advice much easier.

It sounds as though that side plugs are still indicating a rich condition, but not as bad with the new coils(?) If that's correct then chasing potential causes of rich running on that bank seems logical. The CTS is the first place I'd start, maybe even comparing the resistance between the two right at the engine ECU connectors. There's a pin-out for those around somewhere and it might well be the easiest way of at least getting a sense of whether or not there's a problem (I'd still eventually want to clean their connectors with DeOxit and then apply some DeOxit Gold to help conductivity over the long run).

Also, I don't recall you mentioning having checked the fuel pressure itself, although I know you did do some checking of the regulators. If you don't find anything wrong with the CTS that would be my next suggestion, for no other reason than to just rule fuel pressure out.

Hopefully it'll be something simple like a bad CTS or bad connection in the CTS lead. Good luck and let us know what you find.

PS: if you have converters and also happen to have access to an IR temp gun, compare the temperatures of the converters after the engine has been idling for a while. As I'm sure you know, a higher converter temp on the troublesome bank would indicate either a rich condition or retarded ignition timing on that bank, since both result in excessive unburned fuel in the exhaust. Good luck!

 

maybe there is a banana in the tailpipe?

 

Or a collapsed/blocked catalytic convertor. Also, I didn't read all the posts but you did say you just did a major. Cam timing correct?

 

Cam timing out by one tooth on that bank (if the engine has just been re-belted)

 

Hello again... and again, thanks for all the suggestions.

ernie - I checked the grounds by connecting an ohms meter to the battery negative connector point in the rear of the engine bay (battery is in the front). I then used a long lead to allow me to reach various points in the vehicle, but before I started I tested the continuity back to the same point. The Lead resistance is 0.6r and therefore I expected to see resistances close to that value around the car.

Sub frame connector - open circuit. I took this off, cleaned up the boots and re-attached it.

Subframe 0.7r
Gearbox 0.7r
RH valve cover 0.7r (x2)
LH valve cover 0.7r (x2)

At this point I checked the engine again and it's still struggling to run on the LH bank.

I also delved into the temperature sensors.

LH Temp sensor resistance is 4.48r when cold.
RH Temp sensor resistance is 4.48r when cold.

No idea what the resistance value should be, but as they are the same and one bank works, but the other does not, I'm guessing it's not the temp sensors.

I cleaned the barrel connectors, and temperature sensor connectors but they looked in reasonable condition anyway.

Whilst I was there, I also went around all the vacuum connections and replaced the tubes where the connection felt a little loose' just for good measure.

zstyle - I don't have the equipment to check the correct operation of the fuel injectors (its a DIY garage). However, I might swap the fuel injector rails and pressure regulators to prove/disprove this point. I did strip the LH fuel injector rail down and tested the injectors for basic operation & coil resistance checks. They all seemed OK.

fatbillybob - haha - wish it was that simple.

johnk... - She's on test pipes at the moment, so no cats are involved, plus judging by the amount of 'steam' coming from the exhausts when started - there's no blockage there! Looks and feels even too (which in itself is surprising considering the trouble the LH bank is having).

dapper - I'm beginning to lean towards this point myself as none of the usual fixes are working, and I'm not really finding anything that has any real impact on the running of the car.

I wasn't confident about changing the timing belt myself, so I employed a Ferrari specialist engineer to fit the belt for me to ensure this part of the job was done correctly. He did have a little trouble aligning the belt, but he was confident it was correct when it was finished. This was the only part of the job I didn't do myself and I paid a hefty sum for the pleasure!

How can I tell if the belt is one tooth out on the LH bank without having to strip down the entire car again. Even if I did do this, I doubt I would be able to tell if the Ferrari Specialist couldn't. BTW - I don't have a ramp and the last timing belt change was done with the car jacked up 2 feet off the floor and the fuel tank removed, so this is a difficult task for me at best. Is there a simple way to check if the belt is one tooth out without stripping the car down...?

Could a dyno tune up specialist prove this point for me (so I can hand the car back to the Ferrari specialist with proof the job needed doing again - preferably correctly this time)...?

Thanks again for all the input...

 

Hi Mike, thanks for the follow up. Sorry it's been such an elusive problem to resolve. Gotta admit that cam timing seems like a possibility, but in my experience on other engines I'd think it'd have to be off by more than one tooth for it to have such a large effect on the way the engine runs. Then again, I've never heard an F119 engine with its cam timing mis-set, so maybe I'm wrong. But if it is indeed cam timing (or worse), and if it's causing such noticable symptoms, I'd think that you would be able to observe its effect on compression test results. So if you want to pursue that avenue you might consider performing a compression test on all cylinders, going to extreme measures to ensure that all cylinders are tested under the same conditions.

One thing I'm still a little unclear on is the condition of the plugs on the bad bank. You seemed to say initially that the plugs on that side were wet/fouled, which tends to indicate that something is causing that bank to run rich. Are the plugs still appearing wet/fouled after the coil change? If so, I'd still tend to focus on anything that could cause that bank to run rich. And if indeed it does seem as though that bank is running rich I'd still suggest comparing fuel pressure readings from side to side.

BTW, did you happen to compare the CTS resistance at the engine ECUs themselves? It's a long shot at this point, but IF the plugs seem wet on that bank it would be worth the few minutes it would take to ensure that the engine ECUs are seeing the same resistance from both CTSs, since there is always a chance that there is some resistance in the wiring on the offending bank (higher resistance = richer mixture when it comes to the CTS circuit).

Good luck! Still hoping it's something simple...

 

If you mean 4.48 Ohm, the measurement is not correct. My two new coolant temp sensors read 2 K.Ohm at about 25C.

 

To check the timing you will have to remove the valve cover. You will need to find top dead center on cylinder 1, of its compression stroke. A piston stop will make that job easier and faster. Once you have found tdc, the "quick" check is to compare the timing marks on the cams versus the timing marks on the cam caps. They should all line up to the marks. So that some on here don't get an aneurysm , the proper way to verify timing is with the use of a degree wheel and a dial gage indicator. But as I said, the quick check is to compare the marks once the engine is at tdc. If the belt jumped a tooth it should pretty obvious.

 

You may want to consider repinning the T55 connectors, which connects to the Motronic ECUs. You can remove the connector from the ecu and see if you have the old double ear pins. If so they are known to lose their ability to properly maintain proper clamping force, which can lead to signal lose. It cost me about $130 for all supplies needed, plus a full weekend, to repin both connectors using tin plated junior power timer pins. Here is the link to my DIY http://www.ferrarichat.com/forum/technical-q-sponsored-algar-ferrari/444155-348-motronic-2-7-connector-harness-repin.html

The other thing I did was to put in a fresh set of EPROMs, because the old stock EPROM material starts to degrade and loses it's ability to properly hold/process the info needed to run the engine. Here is my DIY for that. http://www.ferrarichat.com/forum/technical-q-sponsored-algar-ferrari/351819-348-ecu-removal-chip-removal.html That job only cost me the time to do it.

The reason I am suggesting both of these is because it will help eliminate them as a possible issue. You have already done everything else I can think of, and they are the only other variables that come to mind which may be the cause of your misfire problem. That and it is inexpensive to do.

 

Hi guys,

Here's the update so far...

I've taken the car back to the Ferrari guys who plugged their computer into the car. It revealed one dead oxygen sensor, one dead thermocouple and one potentially dead thermocouple ECU. All my other preparation work and fault finding has proven useful and at least I fixed those problems before giving it to the professionals The CAM timing is correct as far as everyone believes so far, but I'm still waiting to hear the roar of the V8 in full swing before I know for sure...

Having read about several guys on here removing the thermocouple ECU's as they have fitted test pipes in place of the cats, my question is this...

I already have test pipes fitted, so if I remove the thermocouple ECU's as well, does anyone know if the Motronic 2.7 uses a default map to compensate for the thermocouple ECU loss, or does it still 'learn' as it drives...?

Does anyone have any documented information on this to prove that the engine management system still tunes itself without the thermocouple ECU's?

Thanks for any advice...