Hyundai Forums banner

2005 LC obvious multiple misfire

2K views 13 replies 3 participants last post by  Darzet Accent 
#1 ·
Hello all, just looking at a 2005 UK model Accent LC 1.3 SOHC 12V.

Initial symptoms:
Initial very occasional single miss / brief hesitation, not enough to light the check engine light. Possibly intermittent loss of power. Developed into a very obvious heavy multiple misfire, with fault codes for multiple misfire and cyls 1, 2, 3 misfire. (P0300, P0301, P0302, P0303). Car runs (just) but undriveable. No other fault codes. Possible small amount of coolant loss.

Actions taken and further symptoms:
Initial inspection (not much more than poking around to look for anything obvious), checked plug leads for condition and arcing and cleaned plugs. Engine then ran cleanly for around 15 miles then reverted to heavy misfire. Car then left overnight.
Plugs were in poor condition so replaced with NGK BKR5E-11 (per NGK parts lookup). No change, started with heavy multiple misfire.
Testing appeared to show No 2 & 3 coil might be faulty so replaced the coil pack. Car left for a week before fitting coil pack (holiday!). Coil pack fitted, engine ran cleanly for around 5 minutes and restarted OK three or four times, then reverted to a heavy misfire after having been driven round the block. Car left overnight.
Plugs removed and checked; no obvious fouling, damage, or differences in colour / residues. Regapped to 0.035" and tried again, no difference.
After 3 hours or so started car again, started cleanly and ran smoothly for around 5 minutes then reverted to heavy misfire which sounded like a single cylinder misfire, but I couldn't narrow it down to one particular cylinder.
When the car does misfire I can't narrow it down by either plug lead removal or injector lead removal, though removing the lead from No. 4 appeared to make the misfire worse.
Injectors all appear to be working electrically though of course that doesn't mean they're not clogged or dirty.
Though there is a suggestion of coolant loss, there is no obvious sign of head gasket failure.
There's no sign of vacuum building in the fuel tank.

Thoughts:
Assuming the misfire detection is accurate it's interesting there's apparently no misfire on No. 4.
That makes me think the problem is less likely to be something that you'd expect to affect all four cylinders equally such as air flow sensor, fuel pump, fuel filter, knock sensor or position sensors, though I'm not ruling these out.
I'm inclined to think that a head gasket failure significant enough to cause a multiple cylinder misfire would be showing some obvious signs, though I haven't ruled it out and will test for it.
I'm also inclined to think that head gasket or inlet manifold gasket failure would tend to get worse rather than being intermittent.
It's interesting that on several occasions the misfire has disappeared when the car has been left, only to reappear after several minutes running. That makes me think either a part is getting hot and breaking down, or the misfire is masked by the operation of the cold start loop, or there is some other thermal or time effect playing a part.
I'm struggling to find test data or technical information, which doesn't help - and I've got a stinking cold which isn't making systematic testing any easier!

Going to keep plugging on, but if anyone has any thoughts I'd be glad to hear them.
 
See less See more
#2 ·
Have you done a compression test? When your doing a compression test to diagnose a misfire, it's not the actual cylinder compressions that are important but the difference in compression between cylinders. A difference of more than 1.5bar between cylinders is enough for the PCM to detect a misfire. You probably wouldn't even notice a misfire if that's all the difference was but the PCM might. On some engine management systems after a specific number of misfires have been detected the PCM will shut off the fuel injector on the effected cylinder to protect the catalytic convertor. What you get in that situation is an engine that seems to run OK initially after it is started but then develops a noticeable misfire once the PCM switches off the injectors.
 
#3 ·
No, I haven't yet, nor a cylinder leakdown test, but I will do - I have to go and pick up some of my kit before I can do some tests.

Having said that, it doesn't feel as though the injectors are being shut off. The clean start isn't consistent; sometimes it starts with a misfire. Assuming the misfire detection is reliable, I'm getting misfire fault codes for three cylinders, but the engine is still running (albeit badly) and clearly on more than one cylinder, so I assume those cylinders are still firing at least intermittently; all the injectors can be heard ticking evenly irrespective of whether there's a misfire or not, so I assume the misfire isn't affecting or caused by injector operation.
 
#4 ·
Have had another look but not sure I've pinned anything down. I wanted to do the compression test and try and establish definitely whether it's sparking consistently or not.

Went and got my test gear, started the engine and it started and ran smoothly as before. Ran it for a few minutes to get some heat into it until it went into an obvious misfire. Checked sparks using a timing light and it appeared to be dropping occasional sparks on 1, 2 and 3. However I then got no spark detection at all on 4, so I'm not sure the timing light pick up was picking up the sparks correctly.

Switched off, removed plugs to find 1 and 2 wet and smelling of petrol, 3 and 4 clean and dry. Checked compressions at which point I found my compression tester isn't working properly, so the test was fairly meaningless. However there was a suggestion of 2 behaving differently to the other three.

One thing worthy of note; looking into the cylinders, 3 and 4 look clean with black combustion build up on the piston crown as you'd expect. 1 and 2, particularly 2, have some rusty brown discolouration on the piston crown and the edge of the crown on 2 is quite clean in places as though the combustion deposits are breaking away. This makes me lean towards head gasket failure again. Can't unfortunately find my test kit.

Left it for a couple of hours and decided to re-check the spark. Refitted the plugs and it started and ran very roughly, then smoothed out over 30 seconds or so until it was running cleanly, which it did for several minutes before lapsing back into misfire. Checked with the timing light and found 1 dropping the occasional spark, 2 and 3 dropping multiple sparks at random, 4 apparently OK - but initially when I tried 4 I had no spark, so again I'm not sure whether the timing light pick-up was working correctly.

Switched off while I did something else, then returned to it a short while later only to find the car would not fire cleanly enough to start and run.

One other thing I've noticed is that if it's running cleanly it tends to stay running cleanly until revved lightly, at which point it lapses into misfire.

I'd say this might be consistent with head gasket failure and your PCM injector shut down, if it wasn't that I can still hear the injectors clicking and I can't explain the apparent random nature of some of the behaviour. Maybe I'm over-thinking it.
 
#5 ·
I've been faffing about on this thing and haven't really felt enthusiastic enough to do much. I have done some more testing, however, so I'm thinking out loud here. If anyone has any more thoughts I'd appreciate them.

The typical behaviour is that it will start smooth then lapse into misfire as the engine warms up.

However, sometimes it starts rough, smooths out, then may drop back into misfire, often as the engine is revved or if attempting to drive the car.

The car was being driven when the misfire first occurred, and I'm told it did get better and worse.

Dropping into misfire is typically accompanied by flashing engine check light, which I assume is because of the severity of the problem and the possibility of engine damage.

Fault codes indicate multiple random misfire and misfire on 1, 2, 3 but not 4.

Freeze frame data shows short term fuel trim at 0% and long term at -0.8%, nothing else untoward.

There are no other fault codes.

I don't have live data (which is a lesson - I'd prefer in future to have a reader that does live data as it would tell me a lot more)

Pulling the plugs after misfire shows 1 and 2 wet and smelling of fuel. Often, but not always, there is a smell of fuel from the exhaust during misfire.

Checking the spark directly shows a spark on all four cylinders, though the spark on 2 seems a bit weak.

All four injectors can be heard operating irrespective of whether there is a misfire or not; however, obviously I can't tell how much they're fuelling.

I've been hampered by test kit playing up - it looks as though my timing light pick-up may not be reliable, the compression tester certainly isn't, I can't do a leak down test and I can't find my head gasket chemical tester.

Because the fault doesn't occur across all four cylinders I assume there isn't a systematic, sensor, fuel or air fault. I'm not sure if that's a legitimate assumption.

New plugs have been fitted but have not made a difference (they needed doing anyway, so it didn't hurt).

There was some indication of a coil fault so I replaced the coil, which again hasn't hurt but clearly wasn't the issue.

I've replaced plug leads with an NGK racing set I have to test, misted the existing leads and looked for arcing, and swapped individual leads, but to no effect so I assume the plug leads are not faulty.

I don't intend to start throwing parts at it without obvious faults!

I'm slightly concerned that it appears to have lost water, and on topping up appears to have lost water again.

Once on pulling No. 1 plug I had slightly oil / fuel smelling white fumes from the cylinder, but I don't think they were petrol or oil fumes - steam perhaps?

A sizzle test shows the presence of water in the oil. I'm not sure whether this could be from condensation or is conclusive of a large water leak into the oil.

The tops of No. 1 and 2 pistons are very clean and getting cleaner. I'm concluding either wash down from excess fuel or steam cleaning from a water leak.

Although there are no conclusive signs of head gasket failure I'm coming to the conclusion that it's in the early stages of a head gasket blow on 1 and 2, and to an extent that I'm trying to avoid that conclusion because I'm not confident in my fault finding and not keen on doing the job!

That might tie in with Scott's comment about the PCM detecting a light misfire (from the head gasket failure) and shutting down fuelling. I don't understand why it might sometimes smooth out though?

Next task seems like EITHER find my test set or buy a cheap 'one shot' one and test for combustion fume in the cooling system, OR just give up and whip the head off.

I don't see new head bolts on sale for these so I assume they are re-usable?

Does anyone have any links for head removal / replacement or bolt torques / tightening procedure?

Thanks again for any help.
 
#6 ·
So having gone on holiday, come back and stared at this thing for a month I raised the enthusiasm to do something about it, starting with replacing all my missing or iffy test kit.

Chemical coolant test was inconclusive, and the cooling system didn't seem to be overpressuring. It was however definitely losing water.

I struggled to get in warm enough for a decent compression test due to the worsening misfire, but the test with a new compression gauge showed 145 - 50 - 175 - 175, which confirmed the suspected issue with No. 1 and particularly No. 2 (and indeed the other two are on the low limit)

Further, while warming it up I was getting copious and increasing amounts of oil coming from the area of the head and burning off on the exhaust manifold.

Given the compressions, there seemed little alternative but to take the head off and have a look, which I did.

The head gasket fitted, which I assume to be original given low mileage and full history, is two sheets of metal with no apparent composite or sealant material. However, adhering fairly evenly to all surfaces there is a black carbonised material which I suppose might be the remains of a sealant coating or might be carbonised combustion by-products from a gasket leak.

There isn't a clear area of damaged gasket or gasket blow. There is oil and some water across the block, head and gasket faces but that might be spillage from removing the head. There is a scaly grey-brown residue between the waterways and combustion chambers of No 1 and 2, particularly No. 2, which ties in with the possible failure on that cylinder.

Additionally, the inlet manifold tracts for No 2 and No 4 are quite oily, whereas those for 1 and 3 are fairly dry.

Couple of interesting points; some of the inlet manifold bolts were not all that tight, while the mounting bolt for the fuel rail adjacent to No 2 was only finger tight at best.

Bores all seem OK with no ridge, no signs of piston ring damage and original honing marks still clearly visible. Some rock on the pistons at TDC but no more than I would expect.

Head measures in spec for flatness, so it looks like clean up, new gasket, reassemble and see what happens!
 
#7 ·
Hmmm... looks like back to square one.

I've been pottering on with this job as time allows (not helped by Christmas and delays getting a head gasket). Finally got it back together yesterday, starting it and ... misfire. Heart sank. Left it to run a little anyway and the misfire smoothed out and the check engine light went out. Hooray. Left it to warm up, all looked good. Got it hot, touched the throttle, immediate misfire and check engine light. Heart sank again.

Decided to check compressions while it was hot, got 175 psi on No 1, went to No 2 and the (new) compression tester started playing up, so I couldn't get a meaningful reading. I'm now wondering whether the bad compression on No 2 was genuine or whether it was the compression tester playing up, though I did check it twice at the time with the same result.

At a bit of a loss as to what to do now (apart from have a go at the people who sold me the compression tester) - wondering if it is a sensor or engine management fault after all.
 
#9 ·
Did I have it crack tested? No. Having said that, I'm used to looking for cracks; the head was clean and I inspected it carefully, particularly No. 2, for cracking and other damage. It would have to be a fairly major crack as well, bearing in mind the compression also appeared low on 1 as well (and was only just in spec on 3 and 4).

Going slightly against my gut feel, I persuaded myself it was a head gasket failure (or other engine damage) because of the low compressions, a suggestion of water loss and because the rocker cover seemed to be pressurising (increasing leak from the gasket and traces of oil in the inlet). I could persuade myself of a cracked head the same way. It's always possible I missed something, and my paranoia says I have, but rationally I think it's less likely, particularly since the validity of the compression test is doubtful due to the subsequent failure of the gauge.

The random nature of the problem also niggled me; the fact that sometimes it will run smooth or appear to cure itself, and sometimes run rough. That doesn't seem consistent with a head gasket failure or cracked head, though I suppose you could convince yourself of changes related to the engine warming up or the engine management cutting cylinders as it detected misfires. It still seems more consistent with an electrical, sensor or management fault.

I will get another compression tester, and probably also an OBD2 reader that does live data (it'll be a useful tool to have anyway), and retest to see what I have now, but if there are any other suggestions as to where to look next I'd be glad to hear them. Thanks.
 
#11 ·
I think this is a sticking valve or a defective valve lifter.

Got another compression tester and tried again. Still got 60 psi on No 2, the tester seems to be working OK and the others are 175 psi, so it seems there's definitely an internal engine fault.

However, for some reason I tried spinning it again after I'd got the reading on No. 2. To my surprise it jumped up to 100 psi on the first stroke, then nothing. OK, try again - jumped up to 140 psi, then nothing. Every so often I can get more compression out of it, just on a single stroke.

I'm not seeing this behaviour on the other cylinders, so I don't think it's the gauge.

The only way I can explain variable behaviour like that is if a valve is hanging open by varying amounts, unless I'm missing something - so it seems to me the valve is tending to stick open.

That'd certainly explain the intermittent behaviour and the fact it sometimes cleans up.

Just have to hope the seat isn't too burnt after all this messing about...
 
#14 ·
So a month after replacing the valve spring (without taking the head off) all is still working well.

Think there are a number of lessons here. On the plus side, I didn't change anything without evidence of an issue. On the minus side, I allowed myself to be mislead by the water loss into thinking the loss of compression was head gasket failure even though I wasn't gut feel happy with that diagnosis; and I missed the valve spring failure as I was looking for an intermittent fault, but still went for the headgasket failure even though I didn't think that would cause an intermittent fault either. Moral of the story is inspect thoroughly and follow the logic.
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top