Exhaust gas leak. Suspect manifold

[DamirGTI 323]

If it doesn't consume any on then it'll be good .. was just a thought that perhaps the rings didn't seal up still being freshly rebuilt .. but stable oil level should indicate the running in process being done successfully .

 

 

Missed this one ! Indeed the manifold does look like it runs lean , should be dark gray to light black :

[DamirGTI 323]

5 hours ago, jackherer said:

After screwing that back in the AFR was about 14.7 at all the RPMs we tried it at. This was all stationary in neutral.

 

Tried driving it with the lambda probe in the exhaust ? if so what does the lambda do while driving ?

 

D

 

 

Edited September 13, 2019 by DamirGTI

[jackherer 543]

2 hours ago, AlexRS2782 said:

These are the results from all the MOT's i've put mine through over the years - 1991 J Reg 1.9 GTI - standard management & no chips / remapping.

May 2019 - CO 0.881 HC 628

May 2018 - CO 0.690 HC 376

May 2017 - CO 1.094 HC 288

May 2016 - CO 0.375 HC 119

April 2015 - CO 0.996 HC 129

April 2014 - CO 0.990 HC 151

April 2013 - CO 2.540 HC 150

April 2012 - CO 1.530 HC 122

April 2011 - CO 2.680 HC 160

April 2010 - CO 0.890 HC 109

April 2009 - CO 0.840 HC 127

Thanks. It's interesting that the recent ones seem to have jumped up to the sort of HC level both mine and Paul's Emerald equipped 205s have. Does yours smell of fuel with HC at 628?

[jackherer 543]

1 hour ago, Andy said:

Love to see a post of your fuelling  and ignition map if you could post them.

I have the maps from after both of Paul's mapping sessions, they're on an old laptop with no internet connection but I'll dig them out in a bit.

[jackherer 543]

15 minutes ago, DamirGTI said:

 

Tried driving it with the lambda probe in the exhaust ? if so what does the lambda do while driving ?

 

D

 

 

No, we ran out of time because Paul had to go to work. I should have driven it really, just to experience the symptoms if nothing else.

[DamirGTI 323]

Just checked mine , last year MOT was measured :

 

CO - 0.490 (out of 3.500 maxi value)

HC - 229 ppm (300ppm being maxi value for the HC)

 

No funky exhaust smell , apart from the cold start till it warms up when it emits some but that's normal .

 

D

[AlexRS2782 136 1 Cars]

1 hour ago, jackherer said:

Thanks. It's interesting that the recent ones seem to have jumped up to the sort of HC level both mine and Paul's Emerald equipped 205s have. Does yours smell of fuel with HC at 628?

Maybe a hint if you're stood near the rear of the car / exhaust itself but definitely nothing of note from the engine bay or coming into the car itself.  In honesty it certainly doesn't seem much different throughout the years i've owned it.  The only exception being when the original manifold cracked back in 2013, which did result in a noticeable fuel smell in the cabin.

 

The MOT's up to & including 2013 were done at a mates garage who works on my Focus RS & everything since has been with the place Miles uses so i guess has been driven down and put through nearly straight away so hot / up to temp.  I must admit i did notice the increase this year but the last 2 years the car has done less mileage than it did before and i have been chasing a niggly on & off running / starting issue which could be fuelling related (as all the other normal bits have been changed / sorted) so not sure if that's had an effect tbh.  The only change to the car last year was i finally dropped the standard manifold & old OMP exhaust and it's now on a Magnex manifold (modified by Torque Technique with a new merge collector as the old one had cracked up badly) and they also made a custom exhaust at the same time.

 

TBH my Mk1 Focus RS, that i recently put back on the road, runs on Autronic SM4 & due to the turbo setup / WRC spec exhaust that's far worse for fuel smell than anything the 205 has ever kicked out 

[Andy 105 1 Cars]

Morning,

 Another thought / check. Your Emerald will controlthe fuel pump via a relay that it switches to ground. With the ignition off, remove the relay and , witha short piece of wire with a spade terminal at each end, activate the pump by connecting across the high current feed and pump terminals . This will run the pump continuously but nothing else. Then you can check thoroughly for fuel leaks, starting at the tank end, checking the steel fuel lines ( are they new by the way or old ones that you have cleaned and repainted ) , then the connections at the lower bulkhead near the rack, on to the bottom of the filter, out of the filter on to the  fuel rail, then off the rail back to the return pipe . They are all probably fine but if there is absolutely no smell of fuel, you will then be certain that the smell is connected with the combustion process .

As for the setup you have, it should work well.Rhe fact that is does not suggests a problem. The ecu will not control the SAD so I assume that is removed and the pipework disabled / blocked. One of the screws that locates the oil filler fits into the side of the inlet manifold. Make sure that this screw is both in and sealed as it goes straight through the casting and would cause an air leak . Once hot , I assume that your engine idles smoothly at around 900rpm . You may have it set up so that the air bypass screw is slightly open, but on a single throttle inlet, I would expect that the tickover was set on the throttle stop so that the throttle was open by maybe a thou or two . In this static situation, with a standard engine, it should be running with about 10degrees of advance ( but see my earlier comment above) and an afr of 14 or just over ( 14.7 at tickover can be difficult to achieve ) With the laptop hooked up, it is easy to adjust both ignition and fuelling on the maps at that one cell( and the ones either side of the target cell)  to achieve a smooth tickover with the correct MOT emissions .

If it will not do that , then there is , in my view, some mechanical problem elsewhere that is preventing the correct tickover .

I know you are a long way from me , but I do have an old laptop with a Serial port and the Emerald software loaded, so would be happy to help if you happened to be in Derbyshire !!!!.

With an Emerald and wasted spark, your engine should be running better than any standard Peugeot 205gti on Jetronic and a distributor .

[Andy 105 1 Cars]

Oh. MAP sensor. Usually the tuner will want to use either throttle pot or MAP as the primary load input but not both. Throttle pot has the advantage of a faster response so is the preferred option. .

[Telf 222 2 Cars]

Andy,

After the car came out of Emerald the second time it was idling fine and starting spot on - instant rise to 1000 rpm and just sitting there with no movement whatsoever.

It did seem ok- i drove home with Kieran and the smell seemed to have gone. A couple of weeks later i went out with the wife in it and she commented on the exhaust/fuel smell whilst we were driving.

 

This lead me to re visit the issue( i probably done about 200 miles since the second emerald visit) So i changed the manifold because of the crack as detailed here.

I also removed the cam cover and sealed it with a loctite sealant as it was slightly leaking. ( the same sealant is also on the manifold gaskets). next i checked all the breather hoses and found i could tightened some clips.

 

When i tried to start it after the hoses it wouldnt idle without gassing and holding the throttle pedal. this led me to think the mapping must have been achieved with an air leak somewhere. 

So i now wonder if this leak has effected the entire map? I'm not experienced enough in this to know the answer. This coupled with the exhaust hole i found yesterday - would these factors ruin the mapping?

 

Ive also checked the fuel lines completely from the back of the metal hoses to the rail- thy are all brand new from baker BM. there are no leaks.

[Telf 222 2 Cars]

Kieran has the software to look at the Map values but i dunno if he has ever had to alter any values before - is it a viable thing to do?

 

Edited September 14, 2019 by Telf

[Telf 222 2 Cars]

1 hour ago, Andy said:

The ecu will not control the SAD so I assume that is removed and the pipework disabled / blocked.

correct the SAD is removed and the pipe work blocked- I had to tighten the blanked port under the inlet chamber post the second mapping. the screw you mention in the inlet chamber is also tight- its loctitened and sealed as ive had this issue before

[Andy 105 1 Cars]

Morning,

 Right. Fuel system fine. After your second trip to see Dave Walker , the car idled and ran fine . So something changed subsequently which you have detailed. An air leak would affect the idle settings on the map, but once the throttle is open, you have a big air leak (!) so the rest of the fuelling map will not be seriously affected . If Kieran has the laptop and software, altering map values is easy to do and is equally easy to reverse. However, it is best done with a wideband sensor installed, so that you can see the afr values change as the fuelling is altered. A lambda probe stuffed up the tailpipe will suffice although I use a Bosch 4.9 sensor installed in the exhaust pipe just downstream of the manifold and have it permanently wired to my ecu  via a driver and gauge .

Having said all this , if  your car was fine pre manifold change, then you are back to some kind of mechanical issue rather than a faulty map . Any air leak in the oil breather system will play havoc with tickover, but perversely , increase the idle speed , not reduce it . If you can meet up with Keiran , , hook up a laptop and a lambda probe, then that will tell you a great deal about fuelling without having to disturb or dismantle anything . Oh, compression check is a good idea too, but I bet that will show a good pressure on all four !

 

[Telf 222 2 Cars]

Andy, I'm going to take the car to Kieran today. The car was fine start wise after the manifold and only started with the idle issue after I went to work on the hoses

[Telf 222 2 Cars]

Andy I went out an shorted the pump to make it run . Had a good look checking the unions , no leaks apparent. The fuel is flowing freely back to the tank. I would say there is a faint smell of fuel but I don't know if that's because its being pumped but not used. 

 

I left it running for about 5 minutes. I then left the car to stand for another 5 then started it. I could definitely smell fuel at that point but I guess that might be down to the system running unused for a period of time. It would appear in don't have an injector open also as there was no bang on start

Edited September 14, 2019 by Telf

[jackherer 543]

Injection first map:

Injection second map:

Ignition first map:

Ignition second map:

[Telf 222 2 Cars]

Can anybody make sense of these maps and suggest any changes ?

[jackherer 543]

3 minutes ago, Telf said:

Can anybody make sense of these maps and suggest any changes ? 

Not without a wideband lambda and rolling road!

 

The thing is it was obviously OK at Emerald after the first time they mapped it, then something changed and we went back. But that something must be something like air leaks or exhaust leaks either being created or being repaired.

 

Then we took it back and they mapped it again and it was OK again. But then something changed again.

 

I know it immediately failed the MOT straight after the first mapping session but there is no way Emerald let it leave with the static idle fuelling so wrong it would fail the MOT. It must have changed/broken/been repaired almost immediately.

[jackherer 543]

BTW there is very little difference between those two maps that I can see, the ignition is very similar if not identical and there is a small amount of fuel that has been taken out at some points.

[Andy 105 1 Cars]

Interesting. Second mapis leaned off very slightly at tickover, but only by a tiny percentage. What is more interesting, and what I suspected, is that it has been set up with 0 advance at 1000 rpm . So , if your tickover strays close to 1000 rpm, the late timing will mean that the air fuel mixture is not completely combusted . And, because the tickover is at 1000rpm, the spark scatter idle control will not have taken control and tried to adjust the idle speed by altering the ignition timing ( by up to 10 degrees either way). 

However, all this is still academic if it ran perfectly well on these settings prior to your ‘ hose ‘ work.That has to be your starting point . 

[hoodygoodwood 70 3 Cars]

My 309 has a modified 1.9 GTI engine running on Omex with water temp , air temp , throttle potentiometer , flywheel pickup and lambda probe inputs . About 6 months after finishing it i had a terrible misfire on start up , i did the usual check and fiddle with every wiring connector and it started and ran fine , it did it again months later and i tracked it down to the 3 pin Bosch type connector to the lambda - one of the wires pushed back in a few mm . When it did it a 3rd time i carefully removed the wires one at a time and found the tiny barb was not so high on one of the tiny spade connectors , i bent it out a bit and when refitted it went in with a definite click . 18 months later its never played up again . I have had a few other problems with std GTI looms and sensors and i do spend a lot of time checking these male and female spades out .

[Thijs_Rallye 86 4 Cars]

11 hours ago, Andy said:

Your Emerald will controlthe fuel pump via a relay that it switches to ground. With the ignition off, remove the relay and , witha short piece of wire with a spade terminal at each end, activate the pump by connecting across the high current feed and pump terminals . This will run the pump continuously but nothing else. Then you can check thoroughly for fuel leaks, starting at the tank end, checking the steel fuel lines ( are they new by the way or old ones that you have cleaned and repainted ) , then the connections at the lower bulkhead near the rack, on to the bottom of the filter, out of the filter on to the  fuel rail, then off the rail back to the return pipe .

@Telf Have you already tried this what Andy said above?

 

About the obvious petrol smell, I do not think high HC tailpipe emissions are blowing back into the cabin unless your exhaust is leaking. I used to have a petrol smell as well last year when I had replaced my rear axle and rear brake pipes. The top breather hose had slid off of the fuel tank. Might be worth checking as well. They also often perish due to age. I only smelled it while driving due to the sloshing of the fuel.

[Andy 105 1 Cars]

Miles makes a point. With the laptop, worth check8ng that the water temp sensor is reading correctly. There is a lot of fuel enrichment in the warm up stage , and if the sensor is faulty, the fuelling will be all over the place.

 Also worth checking is the tank ventilation  as Thijs mentions 

 

[Telf 222 2 Cars]

Tank vents have been checked . I built them and it wasn't hard to check them. Fuel checked as Andy suggested. It's been to Kierans today with the laptop and a lamda prone attached . It initially all seemed ok then we started getting some random readings with the AFR dropping to 11.8 to mid 13s at about 2000 rpm (stable). I think Kieran realised how bad it smelt at this point as it had been behaving quite well. 

 

Kieran thinks it's an intermittent electrical interference issue that is feeding incorrect data to the ECU. The ECU also randomly disconnects from the laptop.  So the car is now with Kieran who is determined to find the fault. I personally don't think it's a mechanical issue. I spent too long bullding the engine and the car. I think he's nearer the mark with an electrical gremlin. I think I've probably connected a circuit to a vital earth or run a cable to near another cable thats having a EMI effect but I dunno.

 

At the moment I don't want anything much to do with the car it's eaten up so much of my time in the last 2 years so I'm leaving it in his capable hands for now. 

 

Out!

Edited September 14, 2019 by Telf

[DamirGTI 323]

I have no experience with aftermarket ECU's (nobody over here does aftermartek ECU stuff) , but in my free time i often read various engine tuning books .. and i remember that they say for SA ECU's one must be cautious when building wiring loom to either run wire shielding or most important route signal wires well away from the any electrical field (ignition coil especially) .

 

One of the books i have is written by Dave Walker , i can make you an copy of pages about signal "Interference" , if you want to read about that .. or i guess you're fed up with it all together .

 

 

D

Edited September 14, 2019 by DamirGTI