Compression test results, errr....

[mr brightside]

Invested in a Sykes Pickavant compression test kit over the winter to see if i could find out why the 924S smokes on hard accelleration, the car has been under wraps all winter so the engine was cold. Plugs were wet and crusty black on all cylinders but the dampness smelt like fuel that had bled from the injectors over winter. After i'd cleaned them up they weren't too bad, no pitting.

No 1 (Front of engine/NDE): 155psi, generated within 2 strokes.
No 2: 86psi, generated within 1 stroke.
No 3: 58psi, generated within 1 stroke.
No 4 (Back of engine/DE): 170psi, generated within 3 strokes.

The middle two cylinders were so far down i had to reperform the test but all was correct, my kit doesn't perform leakage tests but i don't think it's really necessary given the results. It strikes me as more than just a coincidence that the middle two pots are so bad, is this indicating anything in particular? Any comments appreciated, go nuts!

 

[homesea]

Put a little oil in the cylinders and test again!

 

[mr brightside]

Hang on then....

 

[mr brightside]

Test 2 performed with 9ml of mineral oil added to each cylinder and the engine cranked for 10 seconds to ensure some type of distribution.

No1: 196psi, 3 strokes
No2: 152psi, 2 strokes
No3: 110psi, 2 strokes
No4: 167psi, 3 strokes

Analysis of the two problem cylinders' results shows that in test 1, No2 achieved 51% of the value of the highest compressing cylinder; wheras in test 2 with the oil it achieved 77% so benefitted from lubrication. No 3 however achieved 34% in test 1, and 56% in test 2 with the oil so is still well down dispite generating a significantly better result. No2 is up by 26%, but No3 is only up by 22% giving rise to a conclusion that No 3 is still suffering badly for some reason. Any comments?

 

[barks944]

I had a problem with yy lux smoking under hard acceleration , at the time it made ~120 is psi in compression tests. Actually drove fairly well for the most part, but it was blowing a lot of oil through the breather hoses and really did smoke like crazy under hard acceleration at times. It was a strange problem as we couldn't really think why it was going well but smoking so much under acceleration. Turns out it had 3 broken top rings.......

 

[zcacogp]

Putting the oil in isn't to lubricate; it's to add something to help the sealing of the pistons against the cylinder bores. The rationale is that if you have poor compression then it is leaking, either from the valves or the bores. If the oil helps the compression then it implies that the bores are leaking, and the oil has helped seal them for the second test.

I'm sorry to say but results like that suggest that there is something wrong with the engine bores. You really need to get the head off and the pistons out to have a look at what is going on - damaged piston rings are one possibility. It's a big job, but needn't be expensive if you wield the spanners yourself (which isn't hard.)


Oli.

 

[Sly]

A correct compression test isn't do as you made.

You must crank until the compression on the gauge is stabilised (approx 5 to 10 strokes), the engine must be hot, you must floor the accelerator remove the dme relay (to not inject fuel).

your test, isn't good because the process isn't good []

 

[barks944]

ORIGINAL: Sly

A correct compression test isn't do as you made.

You must crank until the compression on the gauge is stabilised (approx 5 to 10 strokes), the engine must be hot, you must floor the accelerator remove the dme relay (to not inject fuel).

your test, isn't good because the process isn't good []

I've done compression tests a number of times and it never occured to do it with the throttle open. Makes a lot of sense actually .

 

[Indi9xx]

Run the compression test as advised, as well as that, get a leakdown compression test performed as well, from the two tests you can discover volumes about the engine condition and what the faults are.

A good example would be an engine with poor compression but brilliant low leakdown figures would point to a worn cam, bad cam follower or even a coked up valve... a compression test alone would just point you in the direction of a bad engine.

A leakdown test will also show you were the compression is escaping to, as you can listen for hissing noises from the throttle body (intake valve) exhaust (exhaust valve) oil filler (rings/bores) and the coolant level rising (headgasket or cracked head/bore).

Follow that up with an endoscope examination (not so easy to get hold of for a DIY job) and you can pinpoint the problem without even taking anything apart!

If it does indicate rings and bores, the M44 engines can have a habit of building up a lot of carbon in the piston ring lands, a dose of a good engine flush down the spark plug holes and left for 48 hours, then an oil change, followed by a good hard run and another compression test can often see the problem cured (but in that case then have your injectors flow tested as often this causes washed/glazed or carbon coked bores/pistons/rings)

 

[barks944]

ORIGINAL: Indi9xx

Follow that up with an endoscope examination (not so easy to get hold of for a DIY job) and you can pinpoint the problem without even taking anything apart!

Theres some really cheap ones on ebay that are basically a small webcam with an LED light on them. Can't speak for the image quailty but decent small digital cameras are pretty easy to come by these days. Think what your average smart phone camera can do. Might be worth a pop...

Edit: I just bought a 30 quid one, will report back on what its like!

 

[mr brightside]

Thanks for the replies. I hadn't thought of testing the engine warm, and i've noticed in the instructions for the test kit it advises this. It's not out from hibernation until May, so i'll get it running again and try another test warmed up.

I like the suggestion of a more comprehensive leakage test, it would definitely be worth it and probably a while-u-wait service. I'll send Hartech an enquiry, they're one of the best specialists around and only a short drive over the Pennines.

Unfortunately i don't have the facilities to strip out the engine internals myself. I only have a drive and a jack, and one piece of grit in the wrong place would be all it would take to ruin the job for good.

 

[Alexw]

ORIGINAL: barks944

ORIGINAL: Indi9xx

Follow that up with an endoscope examination (not so easy to get hold of for a DIY job) and you can pinpoint the problem without even taking anything apart!

Theres some really cheap ones on ebay that are basically a small webcam with an LED light on them. Can't speak for the image quailty but decent small digital cameras are pretty easy to come by these days. Think what your average smart phone camera can do. Might be worth a pop...

Edit: I just bought a 30 quid one, will report back on what its like!

I bought one a while ago for exactly the same job - you attach a little mirror to the end so you can see round corners, problem was with the mirror attached it won't fit down the spark plug hole [] Fits fine without it, so i had a great close up of the top of the piston.... I did use it to check on the sills as well though so not a total failure.

 

[peanut]

ORIGINAL: Sly

A correct compression test isn't do as you made.

You must crank until the compression on the gauge is stabilised (approx 5 to 10 strokes), the engine must be hot, you must floor the accelerator remove the dme relay (to not inject fuel).

your test, isn't good because the process isn't good []

eaxactly right Sly
I usually crank 10 times each cylinder but 8x would probably be sufficient to be sure of max pressure buildup.

There are three likely reasons for the low compression you have as jon points out. Its probably broken piston ring/s but it could possibly be a small head gasket leak starting or valves not seating properly either the start of a burnt valve or seat.

I would suggest you carry out another test yourself cranking 10 times per cyclinder and note results.
Then you could put a little thin 3:1 oil in the sparkplug hole of the cyclinder that has the low pressure and test again.
If the pressure comes up 10-15psi it would confirm rings. No change would suggest valve/seat or head gasket.

Re the smoke on acceleration I would want to eliminate ignition and fuel delivery issues before blaming the low cyclinder pressure too quickly. A check on your spark plugs should reveal any rich mixture issues in all or any of the cylinders.