Compression test values

[The Hatter]

I'm playing with my 350i at the moment and checked the compressions. Engine cold, open throttle.

On the left hand bank (in PSI)
182,182,179,182 front of car to rear - so far so good.
Right hand bank
189,188,173,190.

Composite gaskets with 0.7mm skimmed off the heads.

I don't like the disparity between left and right; I especially dont like the 173 amongst the 190ish on the right hand bank. However all the values seem very high to me.

Any comments, should I pull the heads off and investigate?

[Jack Valiant]

Hi Martin

The two high and low values next to each other would seem to indicate a small cylinder cross over gasket failure, the other values look in the ball park to me? I would whip that head off to confirm, what do the associated plugs look like compared to the rest?

[Wedg1e]

If the cam followers have drained down you probably won't get full lift on some of the valves which will affect the values anyway.
I'd be ecstatic with 190

[MrPicky]

Those figures don't seem to bad to me with a cold engine, when warm the pistons will expand and get a better seal with the rings, there is also the dry cam followers situation to consider.

Try it with a warm engine and see how they change.

Finally if you suspect a problem with any cylinder then when warm pour a couple of squirts of oil into the cylinder and try again, if the figure improves a lot then the rings are shot or the cylinder scored. If the figures don't improve then it could be a valve issue (or a gasket).

Low figures next to each other could indicate a gasket leak, but I don't see much of that here.

Some things relevant to the temp gauge thread which may also help here.

The jiggle-pin should be at the top. The bulb of the thermostat should also be positioned on the hot side, towards the head not the radiator.

When bleeding the system make sure that the heater is fully on, heaters can hold a lot of air which will only find its way into the main system quite slowly if turned off.

ETA the bit about the thermostat bulb and the other thread.

Edited by MrPicky on Friday 31st October 07:20

[V8 Fettler]

How does the engine run? I would be tempted to retest the "173" cylinder in case the value increases slightly, thereby solving the problem. If it drops then stick with 173.

Isn't 10% a rule of thumb for max variance.

[smash]

The Hatter said:

Engine cold, open throttle.

Erm, wouldn't open throttle be firing the injectors and therefore giving a wet seal to the pistons falsely increasing the readings? The last cylinder tested must've had a fair bit of fuel in it...was the right bank last to be tested?

Also I thought it should always be done warm for comparative purposes? i.e. comparisson to quoted specs or others readings

Edited by smash on Friday 31st October 14:08

[Number 7]

There is a very good guide to comp testing by Dave Baker of Puma Race Engines on the web somewhere (or I have a saved copy if you cant find it). A healthy engine should be giving around 17-20 times the CR, so for a 9.75:1 engine, you should see 165-195 psi. Note that the test should be done with a hot engine, throttle wide open, ignition and fuel disabled, all plugs out and air filter removed. Readings should really be within 10% of each other, and the max reading should be seen after only 2 or 3 cycles.

Just for comparison, my 4.3 (probably only about 9.5:1) gave readings in the range of 190-220 last time I did it.

[adam quantrill]

This does look like a high CR.

To diagnose further, you do it warm dry, then cold wet (a teaspoon no more of oil down the plug hole), then warm wet.

Then let us know the figures.

[The Hatter]

Thanks for all the comments, it helps the old grey matter move along...

Plugs are all sooty due to ECU/AFM/wiring shananigans so no clues there, certainly no steam cleaning;
There's no water passages close to the 'low' cylinder so I don't think it's related to cooling system issues (although if one pot is dodgy maybe another is too?);
Engine was running OK when the ECU/AFM/wiring was playing ball (I think the compressions are a long term issue and it has run quite well relatively recently);
Compression testing was done without fueling (wiring's been removed).

Unfortunately I pulled the top of the engine apart for the megasquirt exercise before checking the compressions, so I can't now run the engine and do a hot test. When I had the heads off last time the bores were pretty much perfect and I've got no reason to think that it's had a reason to crack/break the rings since. All the valves were ground/bedded in at that time so I think they're OK too.

I'm still a bit curious as to the 180psi ish readings on a cold engine; it does seem pretty high but I guess the gauge calibration may have something to do with it. I'm thinking it's the variation that's the problem rather than the actual values; I did try several times and got very consistent readings on each cylinder.

I think I'll have to bite the bullet and pull the head(s) off to investigate. My suspicion is that the head gasket is dodgy; I fitted composites from Ri**ers and I wasn't particularly happy with them at the time. I suspect I have two problems, the 170 psi pot is blowing into the valley area and one of the outermost water passages is giving the cooling system issues. Does anyone have experience of Felpro gaskets or any other make/suppliers?

[jeff m2]

I think a leak down test may be better.
It'll give you a more reliable result.
It will also tell you which valve. (if it is a valve!)

[The Hatter]

Head gaskets confirmed as dodgy, with corrosion on the head/block faces caused by coolant leakage. There was also a load of carbon deposits on the head dowel nearest the dodgy cylinder indicating combustion gasses leaking out, I assume that was the reason for the low compression. I have Felpro (Federal Mogul) gaskets to go back on, I hope they do a better job!

[adam quantrill]

Are you gonna reassemble with ARP studs too - if so get the 20-stud set.

[The Hatter]

I'm tempted by the ARP studs... However this is a standard 3.5L engine so I think it should be OK with bolts; I've spent some time cleaning and checking the bolts/washers and they look OK to me. I'm planning on using the ARP studs on the SEAC engine though... I have two V8s in pieces ATM!

[Wedg1e]

I used a full set of bolts on my 390's heads in the 2001 rebuild and no problems that I know of so far: I did drop the compression ratio from 10.5 to 9 point something in the process though.

[ed_crouch]

ARP studs are great and everything, but if you go for studs, removing a head will be an ENGINE OUT JOB! On studs the heads dont clear the body at the rear of the engine bay!!

Edit: not exactly a problem unless you have an uncontrollable valve grinding fetish, but beware!!

[ed_crouch]

My 390SE compressions were all up at about 190psi. NCK093.

Edit: mighta been 091..... cant exactly remember, but anyway it was a boggo NCK engine rather than the Andy-Rouse-rocking-horse-st-special.

[adam quantrill]

Really? Mine went in ok with the engine in situ.... I did raise it up a little bit off the mounts.

I have since realised that you can put the studs in after with a hex key, so maybe if they haven't seized you can get them out too before lifting the head.

[ed_crouch]

Ah fair one - I didnt look at that with mine. I put a straight edge up against it with the engine on its mounts and decided it wouldnt clear in situ.

[Number 7]

ed_crouch said:

ARP studs are great and everything, but if you go for studs, removing a head will be an ENGINE OUT JOB! On studs the heads dont clear the body at the rear of the engine bay!!

Mines had ARP studs for years, and I never had a problem taking the (N/S) head off a few years back - no jacking involved.

[adam quantrill]

I hope not too (maybe I didn't) as I will be whipping off the right hand one to replace the gasket sometime.