Technical Head Gasket Failed

[thornebt]

My 100hp Panda suddenly started misfiring. Seemed like it was running on three cylinders. I plugged my code reader in and got:

P0300 Random Cylinder Misfiring
P0302 Cylinder 2 Misfiring
P0303 Cylinder 3 Misfiring

I swapped out the ignition coils on 2 & 3 with some old used ones believed to be ok and put in a complete set of new plugs that I had.

This made no difference. I topped up the coolant before my last journey as it was about 1" below the midpoint on the reservoir. It has now dropped to that same level. The attached photo shows what happens to the coolant level with the engine running. It drops to the bottom of the reservoir on the left side and goes up on the right side.

I think there is a slight smell of hot coolant. There was also some slight steam but I couldn't tell where it was coming from. The temp gauge reads normal. When I plugged the code reader in a second time it came up with misfire messages for cylinders 1&2.

I'm thinking it must be a blown head gasket. I have a compression tester on the garage but haven't looked at it yet to see if I can use it with the deeply recessed plug holes.

Is that the way to go please? Or does it seem obvious that it's the head gasket at fault?

 

[koalar]

If the head does let go it's normally between 1 and 2

But you normally don't get any coolant loss

And you don't normally get a misfire code

Normally if you floor it with the engine cold, there is a knocking noise

To save wasting loads of suggestions on every possibility and seeing you have a compression test, it worth starting there

I suspect it will not be the head gasket, but there's no guarantee

 

[thornebt]

The coolant level has definitely dropped. All four plugs couldn't be any blacker! Compression test showed approx 13psi on cylinders 1, 3 and 4 - but zero on cylinder 2 - the needle on the gauge didn't move in the slightest.

I've got an endoscope to stick into cylinder 2 but can't get it to work at the moment! I will persevere with that but in the meantime if anyone has any suggestions I'd be grateful to hear them.

 

[thornebt]

Got the endoscope working. This is what it sees in cylinder 2 where there is zero compression. Not a great image but it's the best I could get.

 

[The Panda Nut]

You are right thats a rubbish pic!! But it looks as if its looking at a valve? If so it does not look healthy. If its running black and it is a valve and there is no compresion it may be a valve burned out. See if you can get any other / better photos to improve the undestanding. Valve does fit with the symptoms (not the water use though) DO you have access to MES? Other sonsor readings may help to narrow things down further. I had these codes on our 169 1.2 60 and it was nothing more than dirty MAP, fouled inlet manifold and dirty throttle body. All is not necessarily lost.

 

[thornebt]

I put the endoscope down the spark plug hole so it is looking at the piston crown. It can't be looking at a valve. I'll have another shot at getting a better photo.

 

[thornebt]

You are right thats a rubbish pic!! But it looks as if its looking at a valve? If so it does not look healthy. If its running black and it is a valve and there is no compresion it may be a valve burned out. See if you can get any other / better photos to improve the undestanding. Valve does fit with the symptoms (not the water use though) DO you have access to MES? Other sonsor readings may help to narrow things down further. I had these codes on our 169 1.2 60 and it was nothing more than dirty MAP, fouled inlet manifold and dirty throttle body. All is not necessarily lost.

It is very difficult to get any good images. The four above must be showing the piston crown which appears to be intact. So the complete lack of any compression must be a valve problem?

Looking back at my first photo in this thread it looked like there was some sort of catastrophic damage to the piston but moving the endoscope around it now looks ok!

I guess it's a head removal job - following the procedure in the Haynes Punto manual for the 16v engine?

 

[koalar]

Compression test again

The results you've got aren't correct

With those pressures it's not even going to fire a single cylinder, or even close

I suspect you're not getting a seal at the plug seat


Tops of the pistons look fine in post 7

 

[koalar]

Compression need to be over 75 psi for the cylinder to fire at all

Compression should be around 220 psi on a good cylinder

Anything over 200 is fine though

 

[thornebt]

Sorry. I was reading the wrong scale on the compression tester. It was approx 12.5bar which is around 180psi. That was with a cold engine as it wasn't really viable to run it to get it up to temperature.

 

[The Panda Nut]

There is a lot of carbon but its been blow or washed away which could be head related. I would be against pulling it apart unless you are sure its nothing less dramatic. Personally I would if necessary pay for a compression test first and a sniff test on the water system then you will have something you can rely on.

 

[thornebt]

There is a lot of carbon but its been blow or washed away which could be head related. I would be against pulling it apart unless you are sure its nothing less dramatic. Personally I would if necessary pay for a compression test first and a sniff test on teh water system then you will have something you can rely on.

Thank you.

I've done a compression test which shows approx 180psi for three cylinders with the engine cold and zero compression on cylinder 2. So I'm not sure what would be gained by paying a garage to do another compression test.

I'm not familiar with sniff tests but I'm guessing it might be testing the coolant for oil / petrol? Can that be done if I give them a sample of the coolant? I don't really want to drive the car as it's running so badly.

 

[thornebt]

If I were to remove the cylinder head extension (the upper half of the cylinder head) I should then be able to see the valves and if one is stuck open on cylinder 2? I'm not in a rush to do that as Haynes rate it at four spanners out of five for difficulty level and I believe them! But that might be the only way to find out what is happening with the valves?

 

[The Panda Nut]

Sniff test tests chemically, for exhaust gas in the water. ,

 

[koalar]

We have conflicting information

P0300 Random Cylinder Misfiring
P0302 Cylinder 2 Misfiring
P0303 Cylinder 3 Misfiring

And

I've done a compression test which shows approx 180psi for three cylinders with the engine cold and zero compression on cylinder 2.

And

The endoscope images

Zero compression is zero chance of cylinder 2 firing,

No firing equals no misfire code, the injector would also keep firing, spray fuel, Should be able to see a difference

A blown head gasket does not in my experience cause zero PSI or a P030? code, there's will always an exception, but it's very unlikely to be a head gasket

Even with a chunk missing from the head gasket you normally get around 75 psi, it would have to be a very big gap to cause zero

That said, we still need to find a way forward

 

[koalar]

We can check if cylinder 2 isn't contributing

Yes your engine is coil over plug the principal is the same

If the compression test is correct, there should be no difference when started with cylinder 2 disconnect, but a difference when any other is disconnected

 

[koalar]

Assuming the compression test is correct

The easiest way to confirm where the problem lies is to put the piston at top dead centre, modify a spark plug by removing the centre and connecting compressed air to it

Depending where the air comes out will tell you where to look before dismantle the engine

Inlet manifold
Exhaust manifold
Oil tube
Coolant

for example exhaust manifold could be something like this

This is a 1.4 16v with a camshaft lobe snapped off and was an easy repair (not my photo)


180 PSI isn't really enough on the other cylinders, although it will run fine at that

Should be another 40 psi cold compression test, although some gauges aren't always accurate


If they were all down the same amount maybe timing

 

[thornebt]

Thank you Koalar. My fault code reader had the codes as 'pending fault codes'.

Instead of using a modified spark plug I can use the flexible pipe for my compression tester. Just need a compressor now! I think that's going to be the way to proceed.

 

[QuboPete]

Hi Thornebt

Sorry to hear your bad news. No avoiding it if you are sticking with this engine - this is a cylinder head off job without doubt.

Assuming there are no leaks, coolant loss (often coupled with excessive pressure in the header tank) is often a syptom of failure of the head gasket to maintain a seal between one or more cylinders / combustion chambers and some part of the water jacket. This results in coolant entering the affected cylinders under conditions of low / zero combustion pressure (typically when the throttle is shut or the engine is stopped with pressure still in the cooling system). Under compression and combustion pressure the situation is reversed and fuel/air mix and combusted gasses find their way into the water jacket causing excess pressure in the cooling system. Symptoms are often a burnt smell in the coolant header, relief valves blowing, distended hoses, and gurgling noises from the cooling system (and often the car's cabin heating system).

Coolant entering one or more combustion chambers will often cause a misfire as the water / coolant vapour 'shorts' the plug electrode(s) to earth. In early stages, this usually happens more on start up, and may clear when the engine warms and the plug(s) are not wetted so much by coolant. Another potential cause of rough running maybe the effects excess water (from coolant) may have on the Lambda sensors.

Having said all that, I agree with Koalar that zero compression pressure does sound like a valve problem - either a bent valve from an engine overspeed (in these days of limiters this is almost always a too early downshift), or a burned out valve from insufficient tappet clearance or just for no particular reason.

If you tackle the head off job, take care with the cam timing particularly if its VVT.

Sorry to be another bearer of bad news. Had quite a head gasket run myself recently - Peugeot Partner 1.4, Lotus Europa TC, Qubo 1.4 VVT, and Berlingo 1.4.

Good hunting

QuboPete

 

[Pugglt Auld Jock]

Thank you Koalar. My fault code reader had the codes as 'pending fault codes'.

Instead of using a modified spark plug I can use the flexible pipe for my compression tester. Just need a compressor now! I think that's going to be the way to proceed.

I can strongly recommend leak testing. Compression testing, where you screw a pressure gauge into the plug hole and crank the engine over until the highest reading is obtained (usually between 6 and 10 times over compression and you need to have the throttle butterfly fully open which can be difficult on some modern engines) is quite a crude test and not very diagnostically accurate. Yes it'll identify a cylinder with low compression but won't tell you much else.

A leak down test, because you are introducing compressed air into the cylinder through the plug hole is much more controlled and lets you identify where the leak is - listen at the intake where a hiss means a leaking inlet valve. At the exhaust pipe and a hiss means a leaking exhaust valve. At the oil filler, where you'll always hear a hiss because the rings can never seal 100% but it shouldn't be a very loud hiss and bubbles in the radiator header tank means a blown gasket or cracked head. A proper Leakdown tester has two gauges and a pressure regulator which lets you put figures on the %age of leakage - there are plenty of Youtube videos on the subject if you are interested - but simply being able to introduce compressed air into the cylinder and listening/watching for where it's leaking is much more diagnostic than a simple compression test. Of course the cylinder you're testing must be on TDC on it's compression stroke and you'll need some way of locking the engine to stop it rotating when you inject the compressed air (Locking the flywheel works best for which you may need to remove the starter motor) The big drawback for many "driveway grease monkeys" is they don't have a source of compressed air so the good old compression tester has to do but sometimes leaves you guessing what is actually wrong.