General compression test

Currently reading:
General compression test

Fivehundred said:
How can air get in by having all the plugs out? The valves are still operating so there isn't suddenly some way for a cylinder to suck air in through another cylinder's spark plug hole!

When the engine is turning, and all the plugs are out, air can enter down any hole that hasn't got a compression guage connected to it.

The 3 other cylinders will suck air down their plug holes rather than from the manifold if the throttle is closed. There certainly won't be an significant manifold vacuum.

There will be a short period when the cylinder being tested is sucking air from a closed manifold and all the other inlet valves are closed. However, the inlet valve actually closes a long time after BDC (bottom dead center) so by the time the inlet valve closes, one of the other ones has been open for some time, letting air into the manifold.

Also, the manifold is quite large compared to a single cylinder, so there wouldn't be much drop in pressure by sucking one cylinder volume out of the manifold which starts off at atmospheric pressure.

If you don't believe me, jam the throttle open, but I'll be very surprised if it makes any change to the reading. I did a compression test and correctly diagnosed a dead exhaust valve and I didn't open the throttle.

Of course, all of this relies on an engine with 4 or more cylinders and only one manifold for the lot. I know some engines have one throttle per cylinder and for them you have to have the throttles open. God knows how you do a compression test on a 2007 Mini. The'll have no throttle and just vary the valve timing to control the engine.
 
Malin Dixon said:
When the engine is turning, and all the plugs are out, air can enter down any hole that hasn't got a compression guage connected to it.

The 3 other cylinders will suck air down their plug holes rather than from the manifold if the throttle is closed. There certainly won't be an significant manifold vacuum.

There will be a short period when the cylinder being tested is sucking air from a closed manifold and all the other inlet valves are closed. However, the inlet valve actually closes a long time after BDC (bottom dead center) so by the time the inlet valve closes, one of the other ones has been open for some time, letting air into the manifold.

Also, the manifold is quite large compared to a single cylinder, so there wouldn't be much drop in pressure by sucking one cylinder volume out of the manifold which starts off at atmospheric pressure.

If you don't believe me, jam the throttle open, but I'll be very surprised if it makes any change to the reading. I did a compression test and correctly diagnosed a dead exhaust valve and I didn't open the throttle.

Of course, all of this relies on an engine with 4 or more cylinders and only one manifold for the lot. I know some engines have one throttle per cylinder and for them you have to have the throttles open. God knows how you do a compression test on a 2007 Mini. The'll have no throttle and just vary the valve timing to control the engine.

Complete and utter nonsense. It's a four stroke 4 cylinder engine. Therefore each cylinder wil be on a different cycle. At any given moment one cylinder is firing, one is drawing in mixture, one is expelling the exhaust gasses and the final one is on the compression stroke. You have blocked the cylinder which is being tested from drawing in air via the plug hole by screwing the compression tester into it. The others can draw in air that way but their inlet valves which can allow air to get into the plug hole past the valve into the manifold are per definition closed when the cylinder you're compression testing is drawing in air. Yes you can do a compression test with the throttle closed, but it is more consistent to do it with it open.

I don't really understand why you arguing about this, personally I've never struggled to co-ordinate plonking my foot on the throttle and turning the key at the same time, YMMV.
 
When I've done a compression test, I've had to hold the guage with one hand and reach through the window to turn the key with the other hand. Pressing the throttle would have been difficult.

I agree that the inlet strokes of a 4 cylider 4 stroke are all at different times, but there is a large overlap.

Typical figures are inlet opens 8 degrees BTDC (before top dead center), inlet closes 44 degrees ABDC (after bottom dead centre). So the valve is open for 232 degrees (180 + 44 + 8). Of that time, the first 56 degrees is while the previous inlet valve is open and the last 56 degrees is while the following inlet valve is open, so the valve is only open on its own for 124 degrees.

Add that to the fact that the inlet manifold is a lot bigger than a single cylinder, and there really isn't a big effect.
 
The compression test I've had done before was done with one spark plug out (at a time), all the coils disconnected (On a Coupe, with one coilpack per cylinder), and the throttle held fully open.
The Coupe ran slightly poorly on 4 cylinders (out of 5), and poorly on 3 cylinders. I doubt very much that a Cinq would run on one cylinder! :p
H
 
wow this is turning in to a complicated thread.

ive made my mind up to take the head of any way regardless if it is the HG or not, will put my mind at ease its not that once its done.

when people come to mutual decision on how to test a HG will you post up so i know for the future please (y)
 
It's not complicated. Sorry if we made it seem so.

Take all the plugs out. Disconnect the low voltage connectors from the coils.

Put guage in 1st plug hole.

Opening the throttle can't hurt, so do it if you can.

Crank engine.

Write down maximum pressure.

Repeat for each cylinder.

Add a squirt of oil into each cylinder and repeat test for each cylinder.

Post the results here and we'll tell you what they mean.

Much quicker than taking the head off, and you might not be able to see if the head gasket was blown before you removed it anyhow.
 
Compression test must be done with throttle wide open so the cylinder fills with air, not from spark plug hole!!! I would like to know how air gets from other cylinders to the one being tested just using the plug holes!!!Readings should be over 180 and ideally around 200-205 and within 10% of each other. Push fit testers are not very good. Only use a screw in types like Snap-on or Sealey.
Over fueling is possibly a sensor problem and def not a bent valve problem. Might be but very unlikely a broken valve spring or burnt valve.
This engine needs a diagnostic done to find the fault ASAP before it damages cat. Lamba will already be faulty due to high fuel mixture, so it needs replacing with a genuine Fiat one and not a pattern one from say Halfrauds. Hope this helps. :)

Andy.
 
Last edited:
aint got the compression tester no more, so cant do the test :eek:

never took a head off the engine before, so would be good to see how hard it is (malin dixon - incase you didnt know this is my nearly completed project car so dont need it on the road in a hurry, hence my attitude toi taking the head off)

cheers for the info any way (y)
 
sumplug said:
Compression test must be done with throttle wide open so the cylinder fills with pressurised air, not from spark plug hole!!! Readings should be over 180 and ideally around 200-205 and within 10% of each other. Push fit testers are not very good. Only use a screw in types like Snap-on or Sealey.
Over fueling is possibly a sensor problem and def not a bent valve problem. Might be but very unlikely a broken valve spring or burnt valve.
This engine needs a diagnostic done to find the fault ASAP before it damages cat Lamba will already be faulty due to high fuel mixture, so it needs replacing with a genuine Fiat one and not a pattern one from say Halfrauds. Hope this helps. :)

Andy.

had it on the diagnostic machine and it came back with map and o2 sensor last time. will see what happens. cant hurt to replace the HG
 
Ok, i'm not going to say which way is best, but here is the way i have recently done it, and well, this is the best.:D

I have done static tests, i pressume it uses a different machine than you other guys are using. Remove the plugs, screw pipe into the plug hole of the cylinder you are testing first, connect the gauge to a compressor, this pressurises the cylinder, so no need to turn engine over and as far as i have been told, a static test is far more accurate than one with a moving piston. You may have to close the valves, do this by putting car into gear and moving forward or back a few inches until the valves shut. Then read the pressure. Repeat for all cylinders. Again, i've not used other systems, but this works for me, but you will need a compressor.
 
thanks for the reply guys.

just been out and taken the head off, suprisingly easy considering ive never done it before.

the HG hadnt gone, got to get a new HG now and clean up the head. been told to do a test where you check for leaking valves or some thing. any one know how to do it?

had to take off the manifold so i took the cat out too. not to sure why but i dont think its a functional cat. its got weld points in the centre joining the two halfs together

will get pics later
 
99sei said:
been told to do a test where you check for leaking valves or some thing. any one know how to do it?

Read my post at the top of the page, a static compression test will pressurise the cylinder, if it holds pressure then it's fine, if it doesn't then its either head gasket (which you say is fine, but how do you know its fine?) piston rings, or valve seats. Rings don't tend to fail these days, but not sure how to test this. Valve seats can wear, may be worth getting new seats cut. There are ways of checking the sealing of valves seats, best off getting an engine builder to do this. Or, just re lap them in yourself, its not the best solution, but it will work and its a fairly easy job to do.
 
99sei said:
had it on the diagnostic machine and it came back with map and o2 sensor last time. will see what happens. cant hurt to replace the HG
If your map is not functioning, then this will make the lamba sensor read wrong anyhows. Diagnostic should of been your first job before head removal. You are making more work for yourself. Tip - when torquing up the head, torque it to Punto 60 head torque figures. Haynes figures for the cinq/sei are wrong. If i remember rightly, another 5 Nm above published figures. This stops heads from blowing gaskets. Dont use a cheap gasket either. A Fiat or Payen will be ok.
And replace Cambelt and Tensioner too.

Andy.
 
sumplug said:
I would like to know how air gets from other cylinders to the one being tested just using the plug holes!!!
Perhaps the head gasket is REALLY cracked. :p
H
 
well i would hve done another HG test but there is a big confliction as to which is the right way. so by the time every one sorts it out id have had the head off and replaced it in the process checking all the valves were ok.

heads off already and still no real resolution as to which is the correct way to do a compression test :p

as for the map sensor ive tried another one and no change. got three lambar sensors (now all three cant be faulty :p )

will see what happens when i put the head back on. (y)
 
Sorry, didn't read all the thread, was too much bickering etc on it but for what it's woth, when doing compression tests I remove one plug at a time, screw in the guage, turnover and record readings, repeat for all cylinders then put a small drop of oil down each bore to remedy and possible sticking pistons rings.

Liam
 
Back
Top