Inlet Manifold Air Leaks

[GLPoomobile 958]

I've read topics in the past where air leaks have been mentioned as possible causes for running problems. But I've not really got any first hand experience of the situation to relate to. I'm reasonably sure that some of my current Mi problems/behaviours are due to an air leak, but thought I'd see what you lot think.

 

Let me elaborate on my situation.

 

As I removed my manifold to re-wire the engine, I took the opportunity to purchase a 3mm PTFE gasket. Not thinking ahead, when I came to refit the manifold, I found the bolts where only just long enough with the new gasket and only just nipped up. Despite being extra cautious, I still managed to strip the thread on the end bolt hole (above the alternator). On the plus side, it'll only be the first few threads. So I've been running the car with that bolt missing for now.

 

Despite this, the car was idling fine. Sat at near as damn it 1k rpm cold or hot, and reasonably smooth. Not perfect, the odd slight miss now and again, but pretty good. The exhaust was making "thrum thrum thrum thrum" rythm though.

 

So it was fine like that anyway. Then I decided to remove the manifold again to fix my oil leak. When doing that I found that I'd not previously fixed the TB on properly. All 4 bolts were only just on, not even done up finger tight (I remember taking it off to re-route the throttle cable and then getting distracted ). So in theory, it could have been sucking in a lot of unmetered air around the edge of the TB, and yet despite this, as I said, it idled fine (and revved etc perfectly).

 

When I refitted the manifold for the 2nd time, I did the bolts up tightish, but not all the way, as I wanted to be able to whip it off quickly if I hadn't solved the oil leak. I figured I'd do the bolts up tight at the end. I solved the oil leak, so remembered I needed to nip up the manifold bolts, but of course I had the injector rail on by then, so couldn't access the bolts (hex head, so allen keys couldn't get in any sense). And I didn't want to remove the injector rail again. Why? Read on.

 

When fitting the injector rail, I managed to shear off one of the bolts that clamps it down. So I only had it held on by one top bolt and the one that goes through sideways by the pressure regulator. I figured if I removed the rail again to get to the manifold bolts, I was less likely to get such a good seal again with only 1 top bolt. Of course, chances are I potentially do still have an air leak through the injector seals anyway, which may be compounding the air leak I think I have through the manifold.

 

So, with everything fitted back, the car now idles at about 1500rpm. It still sits reasonably level at 1500rpm, by which I mean it doesn't idle up and down, but it's definately a lot rougher. The "thrum thrum thrum thrum" rythm from the exhaust is now even more pronounced, and it definately sounds like it is missing a lot more than before, although it stll revs cleanly.

 

Point to note - For some reason I decided to unplug the CTS plug to see what would happen, and the result was that the revs went up to about 2k rpm and it sounded much smoother. My theory is that if it was sucking in a lot of unmetered air, the additional fuel supplied by the ECU when suddenly seeing no CTS signal would balance out the extra air and increase the revs.

 

Anyhoo, I convinced myself that the new idle behavour was proof that I was getting an air leak around the manifold gasket and possibly the injector seals too. The next step was to buy a Tap & Die kit (got now), and some longer manifold bolts, to fix these problems. So that was the plan for this weekend, and I'll be doing it tomorrow.

 

But, something else entered the equation a last night, and I'm wondering if it is related, and therefore will be fixed tomorrow, or if it's likely to be something else for me to hunt down and ponder.

 

Went out to the car at about 7pm to check the battery voltage, as it's a new battery, hadn't been used in nearly 2 weeks, and the weather is freezing, so I thought I'd see how it was fairing. Was showing 12.2v IIRC, so then decided to start the engine and run it for a minute, for no real reason.

 

Engine started no problemo and proceeded to idle normally. Left it for a minute and then gave it some gentle revving, up to about 2k. Then when I revved it, it cut out as I let off the throttle. Turned the key, crancked but wouldn't fire. Waited a few seconds, tried again, bit hesitant but then fired up and revved OK and settled to idle OK again. Continued to gently rev it, then it cut out again. Go through same process as before. The 3rd time it happened, when I restarted it, it only lasted a second, just revved up once and died immediately, and this time would not restart at all. So I left it for the night.

 

This morning, it started first time, and it idled fine. Drove it 10 metres down the road to move parking spaces, and half way through reversing in to a space, I gently revved it and it cut out again. Try and restart and nothing. Pushed it in to the space, tried it again and it started. At lunchtime, after swapping my driveshaft, I tried it again, started first time and left it running. Left it for 5 minutes while I tidied up the tools, gave it some revs on and off, and it was fine the whole time.

 

OH, and also it's done it a couple of times when left on it's own. Will be idling, and all of a sudden just goes.

 

Now I'm hoping this is something to do with the air leak combined with the cold weather, but I'm just not sure. You see when this happens, it doesn't stutter and stumble, and then stall. It almost seems like an electrical fault, maybe ignition, as it just cuts dead and the revs drop like a stone. You rev it, the revs start to drop as normal, and within about 200rpm the needle just free falls. And when it's not restarting, it's not even trying to fire. Just the starter turning.

 

Sorry for the long post. I like to be concise! I'll have more answers of my own tomorrow hopefully, but apart from helping me, I thought this might be useful to anyone else doing a search, and it's quite similar to some of the problems a few other people have had recently.

[grandos 16]

A quick help with air leaks is to use brake cleaner over every joint (careful of fire )

 

The other and also more involved route is to undo any item that is in the intake system after the meter (not sure what meter is used with Mi16 although all cars are the same) and see if the idle/smooth running of the engine is changed.

 

By this you should see the engine run rough if you undo any item in the system after the meter, however if you undo the faulty item there will be no change in the running of the engine until you introduce a mass air leak (i.e. remove a vacuum hose/ sensor and leave a big hole in the intake system.)

 

Hope this helps

[Veero 1]

I use a little taper for work for detecting air leaks in houses. Basically it's like the thing you get with a box of fireworks to keep lighting them but it gives out more smoke. With the engine on waft one of those around all the joints (under the inlet manifold/head join might be tricky) on the inlet system and you should hopefully be able to see where it's coming in.

 

My work buys them for me so I dont know where to get them, Google Smoke Pencil, I'm sure they're only a couple quid.

[welshpug 1,657]

the problem I encountered with an air leak was very poor combustion and rough running, almost as if running rich, but the engine was getting too much air and simply couldn't burn the fuel and was chucking most of it out into the exhaust, cue foot long flames

 

ICV was raising idle speed a tad to compensate, once the manifold-head join was sealed properly the idle was smooth and exhaust note was a purr and far quieter.

 

 

A bit of WD40 or as mentioned brake or carb cleaner will soon trace the leaks.

 

I would not wave what's essentially a bare flame around an engine bay, that's just asking for trouble.

Edited November 2, 2008 by welshpug

[Miles 331 1 Cars]

I had a air leak on one of my old Mi16 405's (was new back then) and that lead to very good MPG figures and the running was not effected, All that had happened was a small piece of inlet gasket about 5mm long had got sucked into the engine at a guess.

[GLPoomobile 958]

Just to clarify, my problem with the stalling and not restarting is NOT related to the air leak. I'm getting no injection when it happens, so I'm off to test the relays etc in a mo.

 

Didn't seem to make any difference when I sparyed WD40 around the edge of the manifold, so I guess I'll have to see how the car behaves once I've replaced the bolts, done them up tight and fixed the fuel rail mounting bolt.

[Veero 1]

I would not wave what's essentially a bare flame around an engine bay, that's just asking for trouble.

 

Sorry, should have clarified, it's nowhere near a bare flame at all, hardly even a glow. Good point but it's nothing like waving a match around.