Compression testing

I was just doing a compression test on a car to check the engine condition before selling it. I picked up a new OTC tester kit and went through the procedure to prepare the car.

My old tester was the kind you have to hold against the plug hole manually, I've probably had it for 30 years. The new one has the cool assortment of hoses to screw in.

The car is a E46 328i,with an M52B28TU, about 180k miles. Tested after being warmed up. I cranked it until the gauge stopped increasing pressure, probably 6 or 7 compression strokes. All the cylinders read within a few lbs of 230psi.

I've been running rotaries for a long time and we never see those kinds of pressures. I don't recall ever having a stock piston engine come up that high, all the stuff I'm finding online suggests 160psi is good, 140psi is acceptable. I know cold testing would be lower but 70psi?? Engine is supposed to have 10.2:1 compression ratio.

Anybody have tips? Am I cranking too long? Does it matter?

Anybody need a BMW engine? lol

jgrewe said:

I've been running rotaries for a long time and we never see those kinds of pressures. I don't recall ever having a stock piston engine come up that high, all the stuff I'm finding online suggests 160psi is good, 140psi is acceptable. I know cold testing would be lower but 70psi?? Engine is supposed to have 10.2:1 compression ratio.

Rotaries really need more revs than pistons to make good compression, more than the starter is really capable of turning.  That's why the RPM is such a critical number when doing a rotary compression test and part of why they flood more easily.

As for the numbers, I've always been told that for generic compression testers the absolute number can vary quite a bit from gauge to gauge, and that's why it's the relative numbers that are important to look at.  One variable is the volume inside the gauge and hose -- that will get added to the effective combustion camber volume, and so will have a different effect on the reading on different cars and with differently-designed gauges.

jgrewe said:

I was just doing a compression test on a car to check the engine condition before selling it. I picked up a new OTC tester kit and went through the procedure to prepare the car.

My old tester was the kind you have to hold against the plug hole manually, I've probably had it for 30 years. The new one has the cool assortment of hoses to screw in.

The car is a E46 328i,with an M52B28TU, about 180k miles. Tested after being warmed up. I cranked it until the gauge stopped increasing pressure, probably 6 or 7 compression strokes. All the cylinders read within a few lbs of 230psi.

I've been running rotaries for a long time and we never see those kinds of pressures. I don't recall ever having a stock piston engine come up that high, all the stuff I'm finding online suggests 160psi is good, 140psi is acceptable. I know cold testing would be lower but 70psi?? Engine is supposed to have 10.2:1 compression ratio.

Anybody have tips? Am I cranking too long? Does it matter?

Anybody need a BMW engine? lol

 

 

Somewhere I have a chart that would tell you the ratio versus psi.  I have done a bunch of work on this subject and tested 100s of cars with both piston and rotary's.  I would be suspishious of the tester.  I would connect the tester to a air source with a known psi and see what you get.   
 

Another thought is that you have a cam with very little overlap. Loosely speaking static the compression ratio determines the theoretical maximum psi you should test to.  Cam overlap will lesson the actual compression you ultimately test to. 
 

I don't know the specifics of the motor you are testing. Does it have any kind of VVT?   Spitballing a little on my part here. I wonder if cold cranking a motor with the VVT not doing its thing could possibly greatly reduce the valve overlap and get you these high numbers?  

I did find this on the inter web.  Not sure of its accuracy. 
 

5:1 100 psi
6:1 120 psi
7:1 140 psi
8:1 160 psi
9:1 180 psi
10:1 200 psi
11:1 220 psi
12:1 240 psi
13:1 260 psi
14:1 280 psi
15:1 300 psi

That's insane.

IIRC, my S52 was reading around 175 psi on each cylinder and it probably has a higher or equal compression ratio.

But like Codrus mentioned, consistency/repeatability is key here as gauges might not be the most accurate.

So what are the details on this BMW? year, transmission, price.

My S52, none was higher than this on a cheap Harbor Freight gauge

Slippery said:

That's insane.

IIRC, my S52 was reading around 175 psi on each cylinder and it probably has a higher or equal compression ratio.

But like Codrus mentioned, consistency/repeatability is key here as gauges might not be the most accurate.

The S52 likely has bigger cams, which will mean lower cranking compression for a given static compression ratio.  At low RPM, a cam with a later intake valve closing will leak off more compression. 

Gauge matters. Heat matters. Cranking rpm matters more than you would think.

My 12:1 race engine with VVT only does 180psi hot because the cam is retarded. If I manually lock it I get 250. 

If the numbers are even, you are good.

Slippery said:

So what are the details on this BMW? year, transmission, price.

2000 328i in front of a ZF automatic. Price would depend on what you want to go with it. I don't think I need anything between the MAF and cats. I'm not pulling it until its sold, it can be test driven. People are pretty proud of them on ebay and marketplace but there would be a definite GRM discount. Send me a private message, let's make a deal!

In reply to Paul_VR6 (Forum Supporter) :

So, I'm not crazy but I'm going to check my gauge against some others with my compressor.

jgrewe said:

Slippery said:

So what are the details on this BMW? year, transmission, price.

2000 328i in front of a ZF automatic. Price would depend on what you want to go with it. I don't think I need anything between the MAF and cats. I'm not pulling it until its sold, it can be test driven. People are pretty proud of them on ebay and marketplace but there would be a definite GRM discount. Send me a private message, let's make a deal!

Are you selling the whole car or just the engine? I am confused.

A healthy M54 or M52TU should make around 180-190psi +/- 10% between cylinders, in my experience.

I just bought a healthy M54b30 with 118k miles and a 90 day warranty from a local yard for $1370, for a data point.

Just selling the engine

When I tested the Rio, he 11.5:1 GDI engine only registered about 120psi. I checked it with a known good one (wife's rio) and they were the same. Whether the gauge reading is the actual or not I cannot say. I imagine with CVVT on both cam's it may matter what position they were in etc on the actual psi.

Industry standard used to be four pulses.  

I can get compression numbers 20% different on different gauges.  Some of them are all inaccurate.   What you want is for all the cylinders to be roughly the same. Within 5%? All good.

They are all within a couple pounds 228-230 on the OTC 5605. I'm going to retest it cold and see what it does. I've also got an M54B30 to check, I hope its as good as the M52.

One other note is that I have been able to diagnose wrong cam timing (sometimes) by looking at compression numbers. If you are far off high or low, it's worth double checking timing of the intake cam on a dual cam motor. A tooth off can produce some interesting results.

Valve lash can also affect compression psi.  If the lash is excessively high, the PSI numbers can read high.  On an old solid lifter Volvo engine I did some experimentation, and found 0.001" of lash change can affect pressure by 5 psi or so either way.