[1.6 fsi] - Misfire or something else

And engine earth, from behind headlight, check clean and bright.
Also, based on another members experience, unplug ECU, clean connector, both sides, allow to dry, before re-connection.
Mac.

Cleaned the headlight point. I had done it ages ago and done it again.

Misfire on Cyl 3 gone now only cylinder 4. I'll try continuity between ECU and all the rest of it.

No DTCs at all, why is that?

Evros
 
Hi @Evripidis,

Have you made a list? It should include the jobs compleated so far and those still to tackle.


Back to basics...

I asked Google AI for an answer to the question, what is a misfire in an engine?

(I didn't know the answer)

A car engine misfire occurs when one or more cylinders in the engine don't produce enough power because they fail to ignite the air-fuel mixture

Thanks Google, but what causes those missfire?
(I knew some of these)

  • Faulty spark plugs
  • Clogged fuel injectors
  • Faulty ignition coils
  • Worn cylinders
  • Vacuum leaks
  • Faulty emissions system
  • Leaking seals or gaskets
  • Faulty catalytic converter
Are there any things that Google missed?
@PlasticMac has mentioned earthing points and electrical connectors that your getting on with.

@PlasticMac is also checking all things VCDS but are other jobs that can be done in the meantime?

Any suggestions A2OC people?
 
Cleaned the headlight point. I had done it ages ago and done it again.

Misfire on Cyl 3 gone now only cylinder 4. I'll try continuity between ECU and all the rest of it.

No DTCs at all, why is that?

Evros
An ECU misfire is when the ECU doesn't see a "a kick" from the crank sensor, that would tell it that combustion has pushed a piston down sharply, when it should have.
So, the ECU will log a misfire against the appropriate cylinder.
If the ECU has no indication from all other sensors, of a reason for the lack of combustion/downforce, then it can't log a specific DTC.
That's why electrically induced misfires are the hardest to diagnose.
Edit: An electrical fault simply means no spark, so no "bang", but everything else is rickety boo.
Mac.
 
Last edited:
No DTC from VCDS, no problem
But misfire detection on 4
@PlasticMac Is it possible to check for spark with a spark plug and coil pack removed but grounded out on the engine the same way you do on a chainsaw?
It would be my first choice of checking continuity in the wiring loom to number 4 before getting the multimeter.
I hope so, If its as simple as weak spark or no spark, then
A. New spark plug in 4 and test before a
B. New coil on 4

@Evripidis What is the voltage on the battery?
  1. Engine off
  2. Engine running
 
Do FSI spark plugs have those screw on connecting tips like some plugs do.
If they do check them and give 'em a tightening up with some pliers.
My chainsaw has fallen foul of poor running because of a loose spark plug tip.
20240223_190753.jpg
 
Last edited:
Hi to all,

Thank you all for replying and for your suggestions. I will reply collectively.

The voltages are what they are supposed to be, 12.5v bat 14 or so at the alternator with high beam on.

The Opcom diagnostic for GM cars (the VCDS counterpart) has a coil pack output test sequence in which the coils are fired sequentially maybe for a couple of seconds each. This helps you identify faults like these without the engine shaking trying to start and also disables fuel injection for you. Is there an output test for VAG that can be performed for VAG cars, I haven't been able to find info online? In any case it can still be performed traditionally.

@rotifer II When I say carry over, I mean when you transfer the coil pack/spark plug from the problematic cylinder to a good one and vice versa.

This is interesting to know, so misfires without a DTC. This makes the work of @Andrew and @PlasticMac on the label file a very helpful addition.

Evros
 
Thank you for explaining carry over.
May I ask how did you performed the carry over between number 3 and 4.

My test sequence would be to check the tip of the plug was screwed down thightly.

A. Leave the old spark plug in and swap coil 3 to postion 4
TEST RESULT =
B. Remove old spark plug from 4 and fit a new spark plug continue to use coil 3 in position 4
TEST RESULT =
C. Remove coil 3 and replace with coil 4 in position 4 with the new spark plug
TEST RESULT =
D. Put coil 4 in position 3
TEST RESULT =

Did I miss anything out?

The problem doesn't carry over so I think that these are the results from your test

A.Result= misfire
B.Result= misfire
C.Result = misfire
D.Result = No misfire

Are these the correct results?
 
Last edited:
In your first post it says,

9. I replaced all the injector and coil connectors.

Have the electrical plugs been rewired with new items?
@A2Z Indeed. I re-checked all that you have suggested. The dipstick is RTV'd in the sump and the upper connections are tight. So as the second filler cap. What I have not checked is for any holes in the tube.

I also did fit check the purge valve with a stethoscope. It does not click and I also unplugged it which cause a CEL to come up.

I am not sure if I do have a leak at the exhaust manifold (upper cat).

I replaced two out of the four coils. Nothing wrong with the old ones, they just seemed a bit weathered on the inside. I replaced them with the ones that came with the engine, that where in a better condition.

The battery is suspect. I got a brand new bosch battery yesterday for my dad so I might try it out.

Evros
This post has my interest, 2 of the 4 coils were replaced but with second hand parts.

Which cylinders have recycled coils?

How old are the old coils? You say they are weathered on the inside?
Could the insulation material inside the coil have started to deteriorate but you cannot see the damage because it is minute but big enough for the spark to ark to the engine.

Can you get hold of 4 new Bosch coils?

Did you find the time to check for holes in the filler tube?

Has the new battery been fitted?

🙂👍
 
Last edited:
Hi to all,

Thank you all for replying and for your suggestions. I will reply collectively.

The voltages are what they are supposed to be, 12.5v bat 14 or so at the alternator with high beam on.

The Opcom diagnostic for GM cars (the VCDS counterpart) has a coil pack output test sequence in which the coils are fired sequentially maybe for a couple of seconds each. This helps you identify faults like these without the engine shaking trying to start and also disables fuel injection for you. Is there an output test for VAG that can be performed for VAG cars, I haven't been able to find info online? In any case it can still be performed traditionally.

@rotifer II When I say carry over, I mean when you transfer the coil pack/spark plug from the problematic cylinder to a good one and vice versa.

This is interesting to know, so misfires without a DTC. This makes the work of @Andrew and @PlasticMac on the label file a very helpful addition.

Evros
Is the diagnosis of misfire coming from the GM OBD tool, with no misfire being logged in an ECU fault scan?
In other words, a scan of the engine controller is clear, but the GM tool is showing misfires?
Mac.
Edit: Are you certain that the FSI ECU supports the ignition output tests that your scan tool is performing?
Mac.
 
Last edited:
GM tool is for other makes, e.g., saab, vx, etc.

I was wondering if such a function exists in vcds as well.

Is there an ecu connector pinout anywhere? A quick engine bay continuity test checked out fine.

The ecu connectios remain to be tested.

If not wiring then either injector or a dried solder joint inside.
 
GM tool is for other makes, e.g., saab, vx, etc.

I was wondering if such a function exists in vcds as well.

Is there an ecu connector pinout anywhere? A quick engine bay continuity test checked out fine.

The ecu connectios remain to be tested.

If not wiring then either injector or a dried solder joint inside.
That ignition circuit test is not available in VCDS. I'd be very surprised if the FSI ECU supports those commands. Personally, I'd take no notice of the misfire it's reporting. It's unlikely that the GM tool fully supports the VAG specific data, especially the BAD Engine. Even VCDS, which has been the VAG specific tool of choice for over twenty years, didn't support the FSI, until A2OC created the BAD label file last year.
Mac.
 
Apologies Mac,

My English is getting worse year after year ever since I moved out.What I wrote had an ambiguous meaning.

I was not implying the use of a different tool for the a2. I was trying to present what an other tool's functionality and whether this exists in vcds.

EDIT: I am only using vcds for the a2 diagnostics.
 
Apologies Mac,

My English is getting worse year after year ever since I moved out.What I wrote had an ambiguous meaning.

I was not implying the use of a different tool for the a2. I was trying to present what an other tool's functionality and whether this exists in vcds.

EDIT: I am only using vcds for the a2 diagnostics.
Misunderstanding understood!
And I'm relieved.
Mac.
 
In your first post it says,

9. I replaced all the injector and coil connectors.

Have the electrical plugs been rewired with new items?

This post has my interest, 2 of the 4 coils were replaced but with second hand parts.

Which cylinders have recycled coils?

How old are the old coils? You say they are weathered on the inside?
Could the insulation material inside the coil have started to deteriorate but you cannot see the damage because it is minute but big enough for the spark to ark to the engine.

Can you get hold of 4 new Bosch coils?

Did you find the time to check for holes in the filler tube?

Has the new battery been fitted?

🙂👍
Quite expensive I should expect, getting the coils I mean. The rest have been checked indeed.

The loom is in bits at the moment trying to check the connection to the ecu.
 
Note (Post edited on 12/10/24)
Summary

  1. Make a note of the 7 sensors from page 40 of SSP 253
  2. Use FSI engine group and block listing (v1.0) to identify sensor group in VCDS
  3. Check operation of sensors with VCDS
  4. Replace broken sensors
Finish
The J220 ECU now has all 7 sensor data inputs required to generating a spark at the right time on the piston compression stroke (according to SSP 253, Pg.40
)


Hi @Evripidis,

I've spent some time today reading the self study programme (SSP) to find sensors that feed data to the J220 ECU so it knows where within the piston compression stroke to fire a coil.
According to the diagram in SSP 253 there are only seven sensors that do this.

Oh,... the self study guide is written for the VW polo engine, I hope its not wrong for the A2.

If we note down the sensor and its code from the SSP 253 diagram on page 40, then use VCDS FSI Engine Group and Block listing (v 1.0) document, that @PlasticMac and the team wrote, to find the group and block number to test the sensor with VCDS.

The aim is to confirm all seven sensors are functional and providing the J220 ECU the data it needs to create the spark at the correct piston compression set point for the different engine fuel/ air mixture modes.

It is possible that some of the seven sensors need checking with a multimeter if they are suspected as faulty or do not report to VCDS.

You maybe know all the above informatiin already? 🙂👍

I am working on finding other tests to do when a misfire DTC is flagged in VCDS.

My knowledge is limited, I'll keep asking question's, posing theories and getting it wrong to piece together the FSI.

I'll copy you into the PM I PM'd myself.Yes, that's right, I've been talking to myself.

The info in the PM is almost complete, let me know your idea's?

I hope what I wrote here and the PM invite document makes sense .

🙂👍

Untested theory - AKA speculation

If we know which of the three fuel delivery modes + the warm up/regen modes the misfires happened further reading may uncover sensors that are active or inactive in each fuel/air mixture mode.
 
Last edited:
@rotifer II
That's a good idea to try and pin down what the status of the engine is, when it misfires.
This can be done by running a VCDS log.
The misfire counter for each cylinder is in Groups 015, (cylinders 1, 2, & 3) and Group 016, (cylinder 4). Group 007 will give you Engine speed, Engine load, Coolant Temperature (G62), and Operating State, ( Knock Protection, Stratified, Homogeneous, etc).
The log gives each group a time stamp, so there's a correlation between the misfires on each cylinder, and the relevant engine parameters.
I've only suggested three Groups, because that's the limit with VCDS Lite. If you have full VCDS,I think you can select more blocks you want to log. I'm a Lite user, so not familiar with the posh version.
Mac.
 
OK, I've done quite a bit in between. I have taken into considetaration all that everyone has written on here. So I thank you all. Here it goes:

1. Checked continuity between the coil loom and the ecu. Eveything checks out.
2. Borrowed a brand new bosch s4 007 battery.

Made no difference this far.

3. I read the workshop manuals that @rotifer II has provided and more particularly the measuring block/diagnostic sections. Despite them being for the 1.4 I was able to understand many things.

4. I read most of the forum posts on here regarding the misfire detection of the ECU, etc.

5. In combination with the well-catered label file of @PlasticMac/@Andrew I was able to make some sense of the readings and the output tests, and basic settings.

6. I am still not very familiar with the terms lambda and fuel trim the way they are presented in the label file but there can indeed be a stuck injector on cylinder 4.

7. In the manuals it states that if the msifire detector (not the error code) reports a misfire then the EGR can indeed be the cause. The EGR is closer to cylinder 4. I get a lot of misfired on 4 and less than 2 on cylinder 3 (practically never in nearly all of my tests).

8. Fuel pressures are what they are supposed to be.

9. After running the fuel trim test it appears that things got better with the misfire rate number on 4 reduced but not zeroed. There are periods when there is no misfire at all. It is actually a lot better than before. I cannot exactly remember how the previous engine was 2 years ago but this is close, the idling seems the same. So I am thinking that whatever is wrong with the injectors now, must have been there back then also.

10. I have not checked the resistance of the injectors. I could have done so but forgot.

11. I can get new injectors from a shop in germany with at the cost of 150 euros each. I cannot find any one in Cyprus that refurbishes injectors and sending them anywhere is false economy. I might as well buy then new. Which is a lot of money.

I will get 2 tires for it, get the tracking done, MOT and a 3 month tax and insurance. I cannot see anything else wrong with it at the moment and I might as well use it to see if anything else pops up.

In the best of cases the injectors are shot and I have to get anothe set. In the worst of cases the engine is toast and I scrap it.

Evros
 

Edit: 23/10/24​

@Evripidis is correct in what they say SSP253 is for the 1.4FSI because it says so in SSP252 on the first page.

1.4FSI.JPG

Self-Study Programme 252: 1.4 l 77 kW Engine with Petrol Direct Injection System in the Lupo FSI

I do apologise for sending the wrong document. Must not have had my glasses on. 🤓

which one is wrong? I'll amend it :)

I work from these doc's also and if I have made a mistake it would be most helpful if you could point it out please

FSI_Ignition_System.JPG

The Self Study Programme 253 if for the polo 1.4 FSI?
I have the self study programme for the 1.6 FSI chain driven engines but not a specific belt driven one for the A2.🙂👍
 

Attachments

Last edited:
I do apologise for sending the wrong document. Must not have had my glasses on. 🤓

which one is wrong? I'll amend it :)

I work from these doc's also and if I have made a mistake it would be most helpful if you could point it out please

View attachment 129918
What is the purpose of:
"Take a test drive and identify speed/load of the hesitation
Remove N80 electrical connector and take a test drive and copy previous speed and load"

Is it to check the N80, (tank vent storage purge valve), is, or isn't opening?
I think you'd need to to run a log to identify the load/speed, and misfire count, pause the log, unplug the N80, restart the log, and repeat.
Mac
 
Last edited:
Back
Top