Engine: kaput?

Death by natural causes?

The engine

1600 K-Series SuperSport. Seven years old, 62,419 miles of fairly hard road usage, with about 10 track days over the years.

~18 months ago

The head gasket went big-style, and was fixed by Caterham, including skimming the head.

For quite a while...

The engine sometimes doesn't run right when started, particularly from cold. Often it sounds like it's running on three cylinders, then it sounds like the 4th is trying to kick in, and the interval between the 4th cylinder attempts gets shorter until eventually it's running normally after the engine has warmed a little. This whole process doesn't take very long, maybe 30-60 seconds.

~A month ago

I noticed a very slow leak in my water rail, which took several attempts to fix (culminating in a new rail). There were no obvious repercussions of this at the time.

~A week ago

I had low oil pressure when turning right (low enough to light the warning light at least once). I checked the level with the engine running and it was low. I put about 2 litres in to bring it up to the mark while the engine was running.


When I restarted the car after stopping at a garage, there was a big plume of white smoke from the exhaust.

This morning

There's no visible water in the oil (I checked for emulsion inside the oil filler cap), and there's no visible oil in the water in the expansion tank. I turned over the engine, and it started with its fairly common misfire. The exhaust was initially clean, but then white smoke started bellowing out. I checked the oil whilst the engine was running, and I think it's gone down a bit.


Did a compression test. First problem: number two spark plug wouldn't come out. Not even a hint of budging, even after I squirted it with Plusgas. For all the other cylinders the pressure went up gradually on cranking from about 8 bar to:

Cylinder Pressure before oil (bar) Pressure after teaspoon of oil (bar)
1 12.5 16
2 n/a n/a
3 11.25 13.75
4 11.2 12.5

So I don't know whether 3 and 4 are low, or 1 is high :-(

According to the guide that came with my Gunson's compression tester, that probably means that my bores, piston rings, and possibly the head gasket are all kaput. Anyone got any other ideas? If I'd put too much oil down would that have really skewed the results? (I don't think I did, but I didn't actually measure it.)

Given my expensive experience last time round I'm tempted to junk the engine. Anyone know the answers to these questions?

Thanks for any help / advice. I'm not too happy right now.

Monday 24th August 2003: Phil came round with his two-foot torque wrench to try and get the spark plug out. It moved (just), but it wasn't happy so we didn't force it. Phil's theory was that a crack in the head may have leaked coolant and caused the plug to corrode in place. I'm debating whether to force the plug out (possibly knackering the head in the process) or to take the head off and try PlusGas from the inside (and probably still knacker the head removing the plug).

Tuesday 25th August 2003: I managed to talk to Dave Andrews (author of the fantastically informative K-series engine page) today. He said I should remove the plug before removing the head because then I can do a compression test on that cylinder which might help to diagnose the problem. He didn't seem too worried by the pressures - they are all reasonable, and the difference between them isn't huge. He also expected some difference with the oil down the bore, and said it's only really relevant if the pressure is particularly low. Dave reckoned that I could get a reasonable-nick 1800 K for about £650, but he wouldn't do that - he'd get a bit of a dog and refurbish it. Well, if that's the preferred route (I guess because then you know exactly what you've got) then I may as well do the same to my engine. Dave also advised doing a leakdown test while the engine is still in the car, because you can get more useful info this way than once the engine is removed. He also independently came up with the same theory as Phil for why the sparkplug won't come out.

Tuesday 25th August 2003: I am seriously considering having a go at this myself. After all, removing the head is only 4/5 spanners in the Haynes manual ;-) Although I haven't driven the car since I realised I had serious problems, I am contemplating taking it somewhere that can do a leakdown test. It's going to be at least Thursday before I can do this, so I've got some time to contemplate. I don't think I need many extra tools, just an engine crane (hopefully I'd be able to borrow one), a work bench for the garage, and possibly an engine stand (Machine Mart have several for less than £50). I'll also need to find competent local firms to do any engineering work (like crack-test or skim the head).

Wednesday 27th August 2003: Went to visit my parents, and Dad just happened to have a spare dial gauge (and stand) that I can borrow! That's one to cross off my "tools required" list.

Thursday 28th August 2003: Good news: I bought a breaker bar from Halfords today and managed to get the no. 2 spark plug out, although it was heavy going all the way. You know that bit of insulating ceramic around the central electrode of a spark plug? Half of it was missing, which could well explain the poor running. Bad news: unfortunately the only place it could have gone is inside the engine. Also, why did the plug crack? My current pessimistic guess is that coolant leaked (either because of a crack in the head, or "just" the head gasket) onto the hot plug. I'm getting frustrated that I can't spend more time working on this (I'm on holiday and spending all my time doing family and house things). Tomorrow I want to try putting an old plug back in the hole to see if there's any thread left, and then try starting the engine so that I can get it warm to do a compression test on the no. 2 cylinder (and check the other readings too).

Friday 29th August 2003: I did another compression test now that all the spark plugs are out:

Cylinder Pressure before oil (bar) Pressure after teaspoon of oil (bar)
1 11.75 15
2 11.75 17.5
3 11.5 12.5
4 10.5 11.5

Those readings are in broad agreement with the first set, but the differences between cylinders are less worrying. No. 4 is lower than the rest, but within the Haynes recommendations. I don't know what to read into the "after oil" figures. Does the fact that none of the initial readings were low mean that the "after oil" readings are useless? It seems too much coincidence that nos. 3 and 4 have little difference between the two readings, but no. 2 is huge, and no. 1 is inconveniently somewhere in the middle. There's got to be something wrong with no. 2, given the state of the spark plug, but I don't think these numbers tell me what it is.

I also noticed that there was still plenty of white smoke being created, and when I cranked the engine with no plugs in, there was smoke coming from the no. 2 hole. The engine started much better than normal with an intact spark plug in cylinder 2 :-)

Steve has offered to lend me his compressor so that I can do a DIY leakdown test; hopefully I will be able to do that this weekend.

Saturday 30th August 2003: My friend Owen suggested that the smoke might be due to an overfilled sump. I thought that oil-based smoke should be blue, but this theory struck a chord because I had recently topped up the oil with the engine running, as currently recommended by Caterham, which is not what I had previously done.

Sunday 31st August 2003: Neil came round, so I asked him for a second opinion as to the colour of the smoke. It had a definite blue tinge, even to my eyes. I can only think that either there's white smoke as well as blue being produced, or the low evening light had made it difficult to tell the colour before.

Monday 1st September 2003: I got up early and drained some oil from the sump. It was difficult to get the level just right, and I ended up having to add a bit more back in. Overall I reckon I drained over a litre. Much less smoke. I decided to chance a drive to work. Halfords didn't have the right spark plugs yesterday (NGK BCP7ES) so I just tweaked the gaps for now. I've ordered new ones from Spark Plugs UK. Their site has a useful guide to the meaning of the NGK plug codes.

The drive to work was uneventful, with the engine running great. For the drive home the engine again started fine with no smoke. After a short period (maybe a minute) there was some smoke which was definitely water because drops were ejected from the exhaust for about two feet. This only lasted a few seconds though, and I think this is normal. Looking good ;-)

Tuesday 2nd September 2003: We pressure-tested the cooling system at lunch time using a kit that Gurpreet brought in, pumping it up to 1 bar (the Haynes manual says that the pressure cap opens at 0.9 - 1.0 bar). Paul got slightly wet before we realised that at that pressure you can't rely on the gadget sealing correctly, and so he volunteered to hold it in whilst Gurpreet pumped it up. There was a very slow (but noticeable) leak, but I'm confident that it was from the seal between the expansion tank and the pump, so that's looking good.

Sunday 9th November 2003: I've been getting the occasional plume of white smoke when starting the engine, particularly when restarting when warm (like when I stop for petrol). I think the smoke is white, but there's definitely lots of it. I phoned Dave Andrews today to discuss my options after the post-Donington engine failure. He gave an explanation for this first problem, which sounded reasonable at the time. Something about the liners being steel, and the block being aluminium, and a result of their different rates of expansion due to heat can be that the liners can move and push the head gasket up, allowing coolant to seep down around the fire rings giving the effect of lots of white smoke when the engine is next started. Anyway, that combined with the bottom-end problem means that my engine is toast.

This page is http://www.strangely.org/diary/200308/kaput.html. It was first published on Wednesday 20 August, 2003 and last updated on Sunday 9 November, 2003.