MR2 Owners Club Forum banner

1 - 13 of 13 Posts

·
Banned
Joined
·
312 Posts
Discussion Starter #1 (Edited)
Gen 4 Caldina Compression #s

Hi I have a quick question. Does anyone know the exact acceptable compression range for a Gen 4 Caldina 3sgte? I found one video with 170 psi each cylinder and another with 215 psi each cylinder. 215 sounds a little high for 9:1 compression but that one isn't available anyways. I just want to know if 170 psi is good and doesn't mean I have to rebuild in the future.
 

·
Registered
Joined
·
1,638 Posts
Compression numbers can be very misleading. There are many variables that can effect them. The main concern when doing a compression test on an engine is the consistency between all the cylinders. If all are within a few PSI of eachother (highest to lowest) then you'll generally be fine. I personally dont like to see more then about 5psi difference between the highest and lowest cylinders. Really, as long as there isn't more then about 8-10psi, its generally fine.

A leak down test will give you a much better idea of engine conditions cause it takes virtually every variable out of the equation.
 

·
Registered
Joined
·
1,638 Posts
Also, just for an example of how misleading a compression test can be (and I've told this story numbers of times).......

I had a girl (old neighbor) bring me her 95 Honda Accord (not that that part really matters) that had a horrid miss at idle to see if I could figure it out before she got raped by the dealer. She said that it had just had a tune-up and the timing belt changed and the miss was there the first time the car started after the work. He BF at the time did the work and couldn't figure out why it was missing all of a sudden. What was really strange was that the miss would start to go away as you rev'd up the engine. By time it was at about 1800-2000rpm, the miss would be virtually gone...

First thing I did was pulled plug wires to find out what cylinder (if not all) was causing the miss. Well cylinders 1-3 caused the engine to stall but when cylinder 4 plug wire was pulled, there was virtually no change in engine running. So this told me that cylinder 4 was the issue.

Next I did a comp test on it cause I didn't have my leak down tester with me (friend had borrowed it) and 1-3 should between 212 and 215psi but #4 showed 180psi so #4 was definitely a problem but one would not think that a cylinder that showed 180psi would be a "dead" cylinder even though it was. So just to make sure, I tried swaping the injectors, plugs and wires around to see if the dead cylinder changed and it didn't.

I then ran over to my friends house and grabbed my leak down tester to do a leak down test on it. After the leak down test, it was obvious where the problem was. 1-3 all had less then 5% leak down but #4 had 100% leak down and you could hear all the air leaking though the exhaust (i.e., exhaust valve(s) on the #4 were not sealing. This lead me to question the timing belt change and after calling her and asking her about it, she told me that the belt was changed cause it had broke but her BF didn't think that the engine was an interference engine so didn't think that there was any problems. Well they are interference and it bent the 2 exhaust valves on the #4.

So the point is, although a comp test can give you an idea, its not really an accurate measure of the engines condition. Case in point, cylinder had 180psi yet was a dead cylinder causing a server miss regardless of the compression reading.
 

·
Registered
Joined
·
2,155 Posts
Gen4 cold compression numbers are usually around 185psi give or take 10psi per cylinder. Compression numbers will increase significantly if there is any oil in the cylinder (leaky valve cover gasket, etc). Compression numbers will also go up with a warm engine.

As for the above story^i cant see a motor with 2 bent valves giving you 180psi compression. Just doesn't make any sense. I've compression tested motors with stuck and bent valves, and they gave me 0 compression. How can you get compression with valves not opening/closing/sealing?
 

·
Registered
Joined
·
1,638 Posts
WrenchswingeR said:
Gen4 cold compression numbers are usually around 185psi give or take 10psi per cylinder. Compression numbers will increase significantly if there is any oil in the cylinder (leaky valve cover gasket, etc). Compression numbers will also go up with a warm engine.

As for the above story^i cant see a motor with 2 bent valves giving you 180psi compression. Just doesn't make any sense. I've compression tested motors with stuck and bent valves, and they gave me 0 compression. How can you get compression with valves not opening/closing/sealing?
There was about .015"~.020 clearance between the valve face and the seat that tapered down to 0 for about 270? of each valve, basically the two valves were just barely not closing and infact were making contact with the seats for about 90? of their diameter. It was nothing more then say a valve that was being held ever so sligh open by thin piece of carbon.

Remember, compression gauges read dynamic pressure (i.e., velocity pressure) and as long as the compresson occurs faster then the leak can release the pressure, the gauge will read a pressure. This is why changes in cranking speed will cause a change in compression reading.

Think of it as an I/C system leak test. As long as the compressed air source can supply air faster then the leak can release it, then you will build some pressure. Yeah, you wont necessarily build as much pressure as the surces but you will build some.

Or even better, think of it as how the factory boost controll system works on the 3sgte, it bleed some pressure off of the WG actuator do the actuator pressure is lower then the source pressure.

Point being, it all depends on how fast the leak can leak out the pressure vs. how fast the source can supply it. And it is entirely possible to have leaking valve (or valves) that will still allow pressure to build. The key thing to look at is the pressure rise rate while doing a compression test. A good cylinder will rise at least 55~60% or so of the total in the 1st or 2nd pressure pulse (depend on the cylinders location when cranking first begins). If the cylinder has an issue, the pressure rise rate will be much smaller. In the example of that honda, cylinders 1~3 rose to about 140psi on the 1st and or 2nd pulse (about 65% of the total) cause they were good sealing cylinders. #4 on the other hand only rose to about 70psi on the 1st / 2nd pulse (or about 40% of its total) because as its trying to build pressure, a large portion was leaking out of the valves. If the cranking speed was reduced, the compression reading would have been reduced and it would have been very possible to see on pressure at all cause if the cranking speed was low enough, the leak could have release pressure as fast or technically faster then the piston could have produced it.

A leak down tester if its correctly calibrated has a .040" orfice in it so it cant supply a very large volume of air and thus will more accurately show small leaks in the cylinder that wont necessarily be seen by a compression testor. Again, this because of the ratio between whats being supplied vs whats able to leak.

This minor leak was why to miss would go away as you rev'd up the engine. The pressure rise in the cylinder was faster at the higher RPM's then the valves could leak so more combustion pressure was available to drive the piston down the bore. At lower engine speeds (like at idle), the pressure is less so the leak ratio is greater and there is more time for it to leak past the valve so less pressure to support proper combustion and less combustion pressure to push the piston down the bore.
 

·
Registered
Joined
·
37 Posts
Gen4 cold compression numbers are usually around 185psi give or take 10psi per cylinder. Compression numbers will increase significantly if there is any oil in the cylinder (leaky valve cover gasket, etc). Compression numbers will also go up with a warm engine.

As for the above story^i cant see a motor with 2 bent valves giving you 180psi compression. Just doesn't make any sense. I've compression tested motors with stuck and bent valves, and they gave me 0 compression. How can you get compression with valves not opening/closing/sealing?
Im having a bad misfire on my gen 4 bad enough it wont run without throttle, my compression numbers are around 140-145 so do you think my timing is off?
 

·
Registered
Joined
·
5,832 Posts
Those numbers are low why don't you just check your timing? Take the top timing cover off and line up the harmonic balancer with the 0 mark on the lower timing cover.
 

·
Registered
Joined
·
37 Posts
Those numbers are low why don't you just check your timing? Take the top timing cover off and line up the harmonic balancer with the 0 mark on the lower timing cover.
I will do that 1st thing tomorrow but before I put it in the car i set the timing and then hand cranked the motor to make sure it didn't move around, the crank timing mark was less than half a degree off and i thought there's no way it mattered. Could this be my problem? Or did it somehow jump timing a lot assuming it is timing
 

·
Registered
Joined
·
5,832 Posts
You'd have to show me what you mean by half a degree off. Did you replace the timing belt? What is the part number if you did, you can't use a gen2 timing belt.
 

·
Registered
Joined
·
5,832 Posts
Yeah that's not going to fly, that's definitely a tooth off. Pull the top timing cover and timing belt tensioner and retension it, then fix your timing and reinstall the tensioner. You don't have to remove the engine or lower timing cover.
 

·
Registered
Joined
·
37 Posts
Yeah that's not going to fly, that's definitely a tooth off. Pull the top timing cover and timing belt tensioner
I have solved the problem and I feel dumb as **** I was using my 5s crank pulley and automatically thought that the Mark's would be the same...... they are far from it on the actual crank gear there is a mark that lines up with a mark on the block thank you guys for the help
69617
 
1 - 13 of 13 Posts
Top