Exhaust Header issues - Follow up of my emission issue thread

grayvixen93

Member
Oct 15, 2020
20
10
13
Altea
I still have issues passing the emission test. (Previous thread)

I had a leak in the header of the passenger side. I replaced it with an aftermarket no-name one (was a gift) but this one did not really fit as I could not get it sealed at the ball joint to the h-pipe (original). I swapped the header back to the original one with new gaskets and tested the exhaust for leaks while the engine was running. The pressure did build up and I only heard some slight hissing from somewhere around the mufflers.

I also installed new O2 sensors, a new fuel filter, new NGK spark plugs with .54 gap, and a new MAP sensor - just in case...
Air filter, spark plug wires, fuel pump, throttle body + spacer (70mm), air intake manifold gasket, water pump, thermostat (195°), and water hoses had been replaced about 1-2 months ago - only went about 50 miles so far with that setup.

Error codes: only 11.
TPS: .96V

So I went back to get the emissions tested.

Current values @ idle (timing: 6 degrees):
CO213.8 %
O22.3 %
HC12 ppm
Lambda1.116
COno values as they do not test CO

The engineer said it can only be a leak in the exhaust system or intake system.

Before I went to the test I smoke-tested and pressure-tested the intake system (Throttle body was sealed with a rubber glove while doing the tests).
Smoke: nothing visible.
Pressure (about 20psi): hissing noise from the dipstick. I sprayed soapy water on it and bubbles did build-up at the handle of the dipstick (I assume that is ok?).

This morning I tested the exhaust system again with compressed air:
  • Tiny bubbles around the joints of my mufflers
    Video
  • Huge bubbles again in the center of the exhaust header collector (where all pipes get together)
    Video
See the results in the attached video.

So I guess I will have to spend the extra dollars on a new set of headers...

I have a stock engine so far but I might want to modify a bit in the future - not going too crazy.
I was reading that unequal length headers are the way to go if the mods are not that much.
I also would like to keep my shorty setup :)

The engine itself is optically "clean" so a set of shiny headers would be a nice upgrade, but I would also be ok with the stock headers.

What would be your recommendations?
 
  • Sponsors (?)


Ceramic coated. They do make a difference in heat reduction. There are videos all over YouTube showing what a difference they make.

Another benefit in theory is that keep the exhaust heat in the pipes, which travel down to your cats and will make them more efficient. That's assuming they aren't already at 100% anyway. Cats need heat to be efficient.

I'd go shorty unequal. Equal lenghts will offer slightly more power, but they can sometimes be a PITA to install and have clearance issues.
 
  • Like
Reactions: 1 user
Lambdometer
lol yeah I would guess with a tail pipe sniffer. My point was how do you measure Lambda when you don't know the Lambse that's being commanded by the EEC? So the test equipment must be basing Lambda off of Stoich which varies based on the fuel used. Stoich for non-ethanol pump fuel is 14.64 and for e10 is 14.08. Who knows what it is for the fuel OP is running in Spain. That old ass fox A9L calibration thinks that he is running 87-93 octane non-ethanol pump gas and is commanding fueling based on that.

In other words if getting the exhaust issues fixed doesn't make it pass, a tune may be in OP's future.

OP pull your computer, open it up and check the capacitors on the board to ensure they are not leaking. They are at the age that they are probably failing, and when they fail they can effect the readings from the O2 sensors.
 
  • Useful
  • Like
Reactions: 1 users
lambda for any fuel burned is 1.00 dont matter if its gasoline or ethonal... lambda is lambda.. the EEC is using a narrow band sensor that helps with switching around lambda it can only read rich or lean in a very small window being a nerst cell sensor. regardless of the ethonal content the ecu in closed loop should be able to target this switching value with corrections and will do some learning for fuel trims over time.

A reading of 1.11 lambda is around 16.2AFR in the gasoline scale [stoic at 14.68] the engine would not be running well if this was a real reading and would experience lean misfires. This is why the engineer suspects that there must be a leak.

A leak in the header flange or the ball flange would cause the sensor to read lean and the engine would compensate by adding more fuel causing a rich condition. I would inspect the entire exhaust system for a leak if they are doing a tailpipe sniff test, you may have a leak downstream of the cat pulling in fresh air and diluting the reading
 
  • Useful
  • Like
Reactions: 1 users
a19what I understand what you are saying about Lambda. I guess I'm thinking about it wrong, or trying to explain my thoughts incorrectly.

If the exhaust leak is causing the O2 sensors to read lean, then wouldn't the computer adding fuel make the lamdba reading be less than 1 indicating a rich condition when measured from the tailpipe?
 
a19what I understand what you are saying about Lambda. I guess I'm thinking about it wrong, or trying to explain my thoughts incorrectly.

If the exhaust leak is causing the O2 sensors to read lean, then wouldn't the computer adding fuel make the lamdba reading be less than 1 indicating a rich condition when measured from the tailpipe?
Yes, but not if the leak is pulling in enough air to dilute the reading. You would also get an O2 code if it were affecting the O2 negatively
 
Thank you all for your replies!!!

I will go for the ceramic coated headers. I will just have to find a shop that ships them to Spain for a reasonable price :D

A leak in the header flange or the ball flange would cause the sensor to read lean and the engine would compensate by adding more fuel causing a rich condition. I would inspect the entire exhaust system for a leak if they are doing a tailpipe sniff test, you may have a leak downstream of the cat pulling in fresh air and diluting the reading
I have this leak in both header flanges and also close to my mufflers.
They are doing a tailpipe sniff test.

I am using 95 octane E5 fuel. You cannot get fuel < 95 octanes in Spain.

Does the timing change the emission values as well?

As I went there the first time I had my timing changed to 14 degrees BTDC, then changed it back the last time to 6 degrees BTDC and my HC went down... Not sure if this was caused by the timing setting.
 
I finally installed a set of new stock headers. The shipping for BBK or similar brands was just ridiculously expensive so I ordered a set via Amazon (Dorman).

I also did seal all connections with Copper RTV 700° F.

The exhaust system has no leaks anymore.

Yesterday I went back to get my emissions checked: failed again.

Lambda was 1.24. The rest of the data they did not tell me this time and the technician said he has no idea why it still fails.

I noticed a strong gas smell:
  • inside the cabin
  • when I am standing next to the car (driver side) with the hood closed
  • next to the trunk while the car is idling
I was checking the engine bay, crawled under the car, and checked the fuel lines but could not see any leaks.

Parts for fuel distribution replaced so far:

Lambda > 1.015 indicates the engine is running lean. Could this also be caused by a faulty fuel pressure regulator? If that might be the case: could this also be a reason for the high lambda reading?
 
The FPR could be a issue. Check your fuel pressure with the vacuum line connected and disconnected. You should see about 40psi with it disconnected and in the low 30s with it connected.

Check your evap system for leaks. Start at the tank and follow the hose up the passenger side of the car to the charcoal canister checking for pin holes. Check the canister for cracks and the solenoid and hose that goes to the intake.

I had a tiny pin hole in mine where they melted it when installing my subframe connectors and you could smell a hint of fuel 6’ from the car when it was running.
 
I would also recommend you go through the surging idle check list. I know you haven’t said it has a surging idle but it walks you through a lot of things like checking sensors that will ensure everything is functioning properly.

 
I just noticed in the first thread here, it's common for me to go back through a thread refreshing, that the timing is at 6* ? I also don't like ngk plugs but that may be just me, a good copper plug is all that is needed but more importantly the timing, I would set the timing at least 10*, 12* is what mine likes.
The fuel smell, like mentioned inspect the vapor lines and also the fill tube grommet at the tank, look for signs of fuel coming from the top of the tank.
 
The FPR could be a issue. Check your fuel pressure with the vacuum line connected and disconnected. You should see about 40psi with it disconnected and in the low 30s with it connected.
I just ordered a fuel pressure tester as mine is leaking :D Should arrive on Monday.

I pulled off the vacuum line between the intake manifold and the FPR. The line did smell like gas. Not sure if this is already an indicator for a bad FPR...

Check your evap system for leaks. Start at the tank and follow the hose up the passenger side of the car to the charcoal canister checking for pin holes. Check the canister for cracks and the solenoid and hose that goes to the intake.

I checked the line from the tank to the charcoal canister. I could not see anything but I will smoke-test the hose tomorrow to make sure the hose is ok. As I cleaned the engine bay about 5 months ago I also "cleaned" the charcoal canister. There was some liquid inside. I assume it was water as the bigger hose was "melted" away. I installed a new hose and used hose clamps on both ends. The canister itself was not cracked. I also checked for gas smell next to the canister and I couldn't smell anything.
The hose that goes to the intake is also new.

IMG_3773.jpg

I would also recommend you go through the surging idle check list. I know you haven’t said it has a surging idle but it walks you through a lot of things like checking sensors that will ensure everything is functioning properly.
Thanks! I did start following this list. I will post updates if I find anything and/or have questions ;)

I would set the timing at least 10*, 12* is what mine likes.
In the meanwhile, I went back to 10*.

The fuel smell, like mentioned inspect the vapor lines and also the fill tube grommet at the tank, look for signs of fuel coming from the top of the tank.
I replaced the grommet at the tank during restoration (must be about 4 months ago).
What I did not replace as I had the tank off the car was the fuel tank vent valve (https://www.rockauto.com/en/moreinfo.php?pk=6278379).
I am thinking of getting this one with my next order.