I bought my NTK sensor for a French company known as MA for short. Like the oil cooler on my M3, it was defective, but natuerally, everyone loves that company as it's branded names for cheap, albeit with potential quality issues (I wonder if they sell rejects TBH!).
Might have been Parts in Motion. It was a sealed genuine Bosch item with hologram etc.
I changed that because I had a feeling it was running rich prior to it, with the old one.
PIM - One of the few ebay sellers I trust, I have bought quite a few things from them and it always comes in an original box in a sealed bag. Really need Fuel trims to get much further.
Having said that I think your rear O2 is dead, 450mV is likely bias voltage - check the ground - which may be dedicated back to the ECU, but that is not causing your emissions problem. Squirt a bit of brake cleaner in at idle (assuming the car is warm) and watch to see if it goes high.
There isn't any scope for the oxygen sensors to move anywhere. One goes in the manifold, one after the cat.
The second sensor plugs into the exhaust as standard so I meant does the aftermarket exhaust have the hole for the sensor to plug into or has the sensor been moved out of the way and not functioning but as you’ve done a full rebuild I think you’d already know if that might be the problem any way.
Was the turbo manifold replaced as part of the rebuild? I’m sure you know they’re prone to cracking so potentially has a non visible hair line crack that is letting gases escape. Just thinking along lines of above suggestion about leak before first sensor.
Edited to add:
“evilution” said:
It is a common fault on the old 600 turbo for cracks to appear in the manifold. This leads to loss of pressure and escaping waste gas, if this occurs the lambda sensors get confused by the readings and the engine will pulse under normal driving.
The cracks aren't very easy to spot but some can be very obvious as seen below.
It's something I have checked previously and regularly, and up until recently I could see nothing to suggest there were any cracks or leaks around the exhaust manifold. I'll take another look whilst the exhaust is off.
I have ordered a fuel pressure regulator, as there is an OE Bosch one that is a slight upgrade on the standard item. I have read that a dodgy FPR can cause similar emissions issues.
As a side note, I have a boost gauge that is tee'd off the dump valve, which is also plumbed into the FPR vacuum line. It's the standard way of doing things, but I was curious as to whether it would have a detrimental effect with an extra few meters of vac hose involved. I can't really think of a logic reason as to why, however.
PIM - One of the few ebay sellers I trust, I have bought quite a few things from them and it always comes in an original box in a sealed bag. Really need Fuel trims to get much further.
Having said that I think your rear O2 is dead, 450mV is likely bias voltage - check the ground - which may be dedicated back to the ECU, but that is not causing your emissions problem. Squirt a bit of brake cleaner in at idle (assuming the car is warm) and watch to see if it goes high.
Found the photos from when I bought it (mid January)
As can be seen, the old one was rather 'sooty'. I expect the new one looks similar now. Probably done no more than 500 miles since then.
Thats the upstream one yes? The output oscillation at 1/s seems ok to me. I think your downstream O2 is dead and you are seeing bias voltage on it. A squirt of your chosen flammable aerosol whilst watching the output on Torque will confirm.
An exhaust/manifold leak before the upstream O2 will allow air IN, this drives the sensor low (as seen from your graph at <0.1V so the ECU richens the mix, driving up your CO and HC - the cat will probably be too cool to be working properly, but it still may be pulling some of the HC and CO out. However that same extra air then throws the emissions calculation off to show that the engine is running lean (lambda very high)
High fuel trims will back up (but not prove) the theory...
EDIT - also is that a white deposit on the sensor - could be coolant, now or in the past..
The rest of the post past me commenting has got a bit TL;DR but I am aware that the post cat should be a stable figure. I would however expect to see a SMALL variation on the figure. Not a total flat line. Going to have a quick skim and see how you’ve got on.
I checked the cat again after cleaning the exhaust out - looks a lot cleaner and there is a small sign of maybe on or two of the cells being broken, but obviously not enough to cause any of this - main thing is that it doesn't seem blocked or broken. Did have photos but they're on my diag laptop outside.
Anyway, I cleaned off the old Lambda and swapped it into the downstream location. The white wasn't coolant by the way (as asked earlier), I think it was just concrete dust from where I'd place it on the floor at the time. It's a bit old, but it did still work without an EML so I hope it'll do the job for now. They aren't distinct PNs upstream and downstream, so I shouldn't have any issues using this old upstream one.
Took the 'new' upstream one out and it was a bit sooty, so gave it a quick wipe and popped it back in.
I also checked around the entire manifold - there are no signs of leaks, cracks etc as per the previous check. I'm not going to take it off to check as it's a bigger job that I had time for. Sleeping dogs, too, and all that.
I refitted the exhaust and checked it was all tight and sealed correctly. I've ordered some new jubilee clips for the compressor side hoses as they need them.
It doesn't do many miles, but prior to checking it over recently when I actually solid it for about 10 mins I had a check/top-up of fluids. The filler cap was changed recently as it started leaking (it sits about the chain, so when they leak even slightly, it's oil all over the shop). Anyway, my drive is on a bit of a slope so I re-checked again and suspect it might have been slightly over the max, so drained maybe 1/5 litre out possibly. I don't know if that would have aggravated the emissions?
Popped a new Fuel Pressure Regulator in this evening, but didn't have any time for anything else really.
There was a significant amount of dirt in and around there, so carefully plugged it up before cleaning it out. I would imagine that it's never been changed, however...
...the old one didn't have a gauze, which I thought was weird.
I've got some braided hose of varying diameter on the way with the intent on changing out some of the older/perished vacuum and breather hoses. There are a number of them (5 or so on the wastegate actuator itself) so I thought I'd do a bit of a revamp on them all, where possible, to avoid inspecting perished hoses for leaks.
Once I've done that I'll hopefully get some time to take some more readings.
A tip I saw recently reminded me of this thread. If you have a shop vac that you can set to blow or a HVLP paint sprayer, with the engine cold hook it up to the exhaust, with duct tape, spray some soapy water on your manifold and watch for bubbles.
A tip I saw recently reminded me of this thread. If you have a shop vac that you can set to blow or a HVLP paint sprayer, with the engine cold hook it up to the exhaust, with duct tape, spray some soapy water on your manifold and watch for bubbles.
I let it warm up today and covered the exhausts with a sheet of cardboard and I'm pretty convinced there are no exhaust leaks now.
Since the MOT, I've done the following:
- New Fuel Regulator - New IAT sensor - Cataclean in the tank - New upper crankcase/PCV breather (genuine Merc jobby) as the old one was 2013 on it - Sucked a little oil out so it's between half and max - Rinsed the exhaust with thinners - Swapped the downstream Lambda with the old upstream one (PN is the same) as I am sure the old one was duff (didn't even look right) - Replaced the FPR vacuum hose - Replaced and checked the hose clamps on the turbo compressor in/out - Replaced the hose clamp on the throttle body - Checked and refitted the exhaust so it's sealed well
(on top of the fact it has had genuine Merc oil and filter recently, 6 x new NGK plugs, upstream Bosch O2 etc)
I used Torque to measure the upstream when it was warmed up and in closed loop. The first 40 seconds or so were at idle - the jump above 0.8 was me raising the throttle.
Also recorded the fuel trims pretty much at idle and at 2500rpm.
I'd ignore the O2 voltages - I don't think torque is picking up the sensor data very well in this view and it makes it look sluggish.
There are no DTCs either, by the way. I think I'll just have to pop it in again and see how it goes - I've mostly otherwise run out of ideas! Nothing is particularly jumping out at me
Well that all looks good to me, fuel trims are OK, sensors look about right - though I would expect a leaner voltage on the downstream, but nothing indicating big problems. Looks like something you did fixed it!
If the sensor readings are all good, it may just be that they purely haven't got the catalyst warm enough. No engine rpm and no oil temp reeks of just trying it and see if it will quickly go through. I've had several things (without sports cats) be a bit of a curse word to get warm enough.
But you can normally see the readings start to tumble down to minimal numbers with the engine above the fast idle speed for a few minutes and then bring the rpm down to the required specs
If the sensor readings are all good, it may just be that they purely haven't got the catalyst warm enough. No engine rpm and no oil temp reeks of just trying it and see if it will quickly go through. I've had several things (without sports cats) be a bit of a curse word to get warm enough.
But you can normally see the readings start to tumble down to minimal numbers with the engine above the fast idle speed for a few minutes and then bring the rpm down to the required specs
I drove it over there to be tested and it was up to temps.
I've ordered a new standard exhaust for it as a last dash thing - we'll see if that makes any odds.
People who talk about how one could pass emissions without a cat must be talking out of their starfish, though!
I do not understand the massive lambda difference between first fast idle and second - it is still saying that it's running hugely lean - I am assuming you would have mentioned if it was only running on 3 cylinders as that could be a missfire. - But nothing in the fuel trims.
According to the 2nd fast you are waaay lean so you would expect to see LTFT pegged at 25%... unless air is coming in after the O2 sensor which brings me back to exhaust leak.
Another thought do these have an EGR valve? Something changed between fast and 2nd fast idle. I am thinking stuck and leaking air in. At that lambda value you would have such a lean state that you would only have a partial combustion which would account for the excess HC
I do not understand the massive lambda difference between first fast idle and second - it is still saying that it's running hugely lean - I am assuming you would have mentioned if it was only running on 3 cylinders as that could be a missfire. - But nothing in the fuel trims.
According to the 2nd fast you are waaay lean so you would expect to see LTFT pegged at 25%... unless air is coming in after the O2 sensor which brings me back to exhaust leak.
Another thought do these have an EGR valve? Something changed between fast and 2nd fast idle. I am thinking stuck and leaking air in. At that lambda value you would have such a lean state that you would only have a partial combustion which would account for the excess HC
No EGR on the 600. That's only on the 700.
There's also no exhaust leaks after the O2. It's a minimal system, with the only join being the exhaust to manifold part and it 100% isn't blowing.
As noted, the high CO suggests it is rich (as does the HC) but the high Lambda suggests it is running lean.
I've reluctantly ordered a new cat, but I expect I'll have to take the manifold off which has its own inherent risks.
Is it definitely firing on all 6 coil packs? It may all be due to a lean misfire if 1 cylinder isn't getting a complete burn
As for the driving it there straight in to the emissions test, sometimes that doesn't get the catalyst itself warm enough
If just one reading was off then you could point towards these sorts of things.
I think the contradictory lambda vs CO suggests that there's got to be a small amount of air being drawn in around the manifold. I've checked it all thoroughly, but the more I read, the more it's pointing towards something like that. I'm going to have to pull the manifold off, I reckon.
If just one of the readings was off then I think we could say that it's running rich because of X,Y,Z, but it looks like the fuelling is confused.
I'm sure the next disaster will probably be a stud snapping flush with the head
Well that was my theory (manifold leak or at least between the cylinder and the O2 sensor) but the fuel trims you show say not.
Though now I am wondering how hot the engine was when you took those readings - coolant certainly was not at full temperature.
Did the car cool down before the emissions were done? That would explain the 1st/2nd Fast idle difference.
So I am revising my theory to an exhaust leak between cylinder and O2 sensor that only opens up when the exhaust gets very hot! Rather than pull the manifold off can you try the blower test I mentioned - also take a good look around the O2 sensor boss as well and make sure the engine (neighbours not withstanding) is at full, full temp.