223 replies to this topic

[DaveyC]

From your first post "Coil pack - it's had two different coil packs"

 

Can you describe which 2 you used, and where they came from.

(to rule out the knows issue with the china knockoff versions of these that can cause problems)

 

 

The one that was on the car (I assume was original) and a used one from a eBay.

[CHILL Gone DUTCH]

So maybe a fuel learn would be a good idea

[CHILL Gone DUTCH]

That was for under heavy load though, wasn't it?   We went off what the wideband was reading and inputted values into OBD tuner until the wideband read what we were after so 11.4 in the table could well have corresponded to 11.8 on the wideband.   The stuttering all happens on quite low MAP values where target afr is 14.7. As soon as you get into the load parts of the table where you want richer AFRs the stuttering goes away.

Think this could be your issue

[CHILL Gone DUTCH]

Some Wideband do need calibrating as they can be far out

[smiley]

Can you post some high res pictures of your engine bay. (especially arround the coilpack area)

 

I know it sounds silly, but you never know.

 

[DaveyC]

Some Wideband do need calibrating as they can be far out

 

Just to clarify the work process I did:

• Fuel learn was done for entire fuel table
• Wideband was installed and calibrated in free air (not connected to ECU at all, just to a gauge)
• Wideband was reading 14.7 when cruising around (same target AFRs as in the fuel table)
• The car went on the rolling road and we adjusted only the cells where the engine is under heavy load i.e. not the place the engine stutters. All the cells where the engine stutters are still the original values from the fuel learn. I'm inclined to rely on the wideband for the load areas of the fuel table because this is where the ECU is in open loop anyway and the narrowband can't be relied upon for figuring out these values, but we were happy to leave all the closed loop stuff up to the standard narrowband and the original fuel learn.

I can go out tonight and do another fuel learn for only the low load and low rev area so it doesn't mess with the high load areas.

[DaveyC]

Can you post some high res pictures of your engine bay. (especially arround the coilpack area)

 

I know it sounds silly, but you never know.

 

 

Can do. I'll grab the camera tonight.

[CHILL Gone DUTCH]

I rich mixture may be choking the car up but only showing its face under low loads It's is up to you But I would use the first or second test file I sent you and do a fuel learn session You should have to run that rich to get your wideband to read correctly

[DaveyC]

I rich mixture may be choking the car up but only showing its face under low loads It's is up to you But I would use the first or second test file I sent you and do a fuel learn session You should have to run that rich to get your wideband to read correctly

 

I'll give it a go tonight and see how I get on.

 

Surely aroud 11.4-11.8 the narrowband is soooooo far out of its accuracy zone that the wideband is much more likely to be right? What's even the point of a wideband if the standard narrowband can read reliably down to 11.4?

[CHILL Gone DUTCH]

Whilst using obd tuner The wideband is primary used for WOT fuel learn and ignition tuning above 250 hp If you have a dial in the car it's used as a check or worry meter 😜

[CHILL Gone DUTCH]

Obd tuber was developed around Oem narrowband lambda So we need to be using the info as much as possible from the narrowband One other thing I did wonder but I'm not a 100% of the answer is that you are running a CAT so should the cat check be switched on ( I'm not sure as I don't run a CAT )

[smiley]

I have a cat, but in odb still "cat check" disabled. (as leftover from last years straight pipe adventure)

(sailed through mot with that just fine, and ran it like that all this year)

Edited by smiley, 12 August 2015 - 09:10 AM.

[DaveyC]

Considering I have a cat, my 2nd O2 reports rather weird values.

[CHILL Gone DUTCH]

Considering I have a cat, my 2nd O2 reports rather weird values.

I don't think there weird because you have a CAT If you didn't I would of guessed there is something wrong with the Second lambda But it could be something else to try as the second lambda is not reading the same as the first So worth trying to switch CAT check on

Edited by CHILL Gone DUTCH, 12 August 2015 - 09:21 AM.

[oakmere]

I think you are wasting your time trying to map this out. It did it before the obd tuner and now with it so unlikely a map issue. There is either and electrical or physical fault with the engine. Fuel pumps / pipes on SC conversions have caused many issues as have wiring problems. Have you checked fuel pressure?

[DaveyC]

I think you are wasting your time trying to map this out. It did it before the obd tuner and now with it so unlikely a map issue. There is either and electrical or physical fault with the engine. Fuel pumps / pipes on SC conversions have caused many issues as have wiring problems. Have you checked fuel pressure?

 

Fuel pressure stays consistent during the misfiring. Recently replaced the plug to the fuel pump before checking the pressure too.

 

Also I've cleaned all the injector plugs.

 

Odd that it only happens at a particular rev range, electrical noise perhaps?

[oakmere]

It is very strange. Did you replace the map sensor?

[DaveyC]

It is very strange. Did you replace the map sensor?

 

Yes, it was on a MAP sensor on the old ECU and I switched to TMAP for the Dutch ECU.

[Exmantaa]

What injectors are you using? Same as before OBD?

 

And to summarize your tuning process and how the car is now:

• Fuel learn was done for entire fuel table
• Wideband was installed and calibrated in free air (not connected to ECU at all, just to a gauge)
• Wideband was reading 14.7 when cruising around (same target AFRs as in the fuel table)
• The car went on the rolling road and we adjusted only the cells where the engine is under heavy load i.e. not the place the engine stutters. All the cells where the engine stutters are still the original values from the fuel learn. I'm inclined to rely on the wideband for the load areas of the fuel table because this is where the ECU is in open loop anyway and the narrowband can't be relied upon for figuring out these values, but we were happy to leave all the closed loop stuff up to the standard narrowband and the original fuel learn.

This process all sounds spot on and no much to improve upon. On low loads you're running closed loop and the fuel trims will always steer it back to 14.7AFR. (assuming your narrowband is OK and no exhaust leaks, but the 14.7 Wideband reading confirms everything is fine fuelling/mapping wise)

[DaveyC]

What injectors are you using? Same as before OBD?

 

 

Before OBD tuner I was using VXT injectors (which also had the problem)

When I changed to OBD tuner I put in some of the Siemens ones. I forget the specific spec, 650cc???? They're configured correctly in the parameters tab where the serial numbers on the injectors correspond with the setting.