I've been following several threads involving members running compression tests and then wondering what their results mean in terms of potential limitations to engine performance. I've derived some simple equations to predict ideal pressure values for compression test readings and the resulting ideal air temperatures, and an equation to give the actual air temperature associated with actual compression readings.

Using actual compression measurements, one can define an effective compression ratio, and compare predicted actual air temperatures with the auto-ignition temperature for #2 diesel of 558 K to see if the measured compression will produce hot enough air to start and run the engine without energizing the glow plugs.

I could just give the results, but it's more fun (at least for me) to give a little background as to how I derived these equations. The ideal gas law is PV=nRT, but since we're only concerned with before and after conditions we don't have to address the n and R parts, just that (P1V1)/T1=(P2V2)/T2 where #1 are values at the bottom of the intake stroke, and #2 are values at the top of the compression stroke.

As the air is compressed, its volume decreases from V1 to V2, the pressure increases from P1 to P2, and the temperature increases from T1 to T2. The ratio (V1/V2) is the compression ratio CR which is 17.5 for the 7.3L PSD. We also know T1, the ambient temperature of the air being pushed into the cylinders, which for an 80 F day is 300 K, and P1 the ambient pressure in the cylinders, which is about 14 psi. That leaves two unknowns, P2 and T2, but only one equation!

The second equation is also an ideal one for relating the adiabatic temperature change to pressure change as the air is compressed. An adiabatic process is one in which heat does not enter or leave the system. Though this is not strictly true in a compression engine, it's a close approximation over the very short time period of a compression stroke, and it gives... (T2/T1)=(P2/P1)^0.286. The exponent (0.286) has to do with the ratio of air's heat capacity at constant pressure to air's heat capacity at constant temperature, and the exact value is due to air consisting primarily of diatomic gasses like N2 and O2.

If you combine these two equations, you can derive the following.

P2ideal=P1(CR)^1.4 = 14(17.5)^1.4 = 769.8 psi

T2ideal=T1(CR)^0.4 = 300(17.5)^0.4 = 942.6 K

T2actual=(T1/P1)(P2actual/CR) = (300/14)(475/17.5) = 581.6 K

CReffective=(P2actual/P1)^0.714 = (475/14)^0.714 = 12.4

P2ideal and T2ideal give the ideal maximum values for a perfect engine (complete filling of cylinders, no blow by, etc..) operating according to ideal equations. T2actual is predicted from a presumed compression measurement of 475 psi, and such a measurement could be used to define an effective CR of 12.4 which if substituted into the equations for ideal values (P2ideal and T2ideal) would give the values that were actually measured. What I'm saying is that an engine with a measured 475 psi could be considered to be an ideal one, but with a CR of only 12.4 instead of the actual 17.5.

If one plugs 558 K into T2actual and solves for P2actual, you get the minimum value for auto-ignition without glow plugs being energized.

P2actual,min=(558)(17.5)(14/300) = 455.7 psi.

So if an engine cranks 456 psi or higher, and ambient is 300 K = 80 F or higher, it should start without the use of glow plugs!

 

AHHHHHH.... I'm having flashbacks of Thermodynamics in college

 

Just after I posted the above, I checked the 2fun4u thread to see how their compression test turned out. They only got 300 to 315 psi, but commented that the engine was cold at 45 F when they ran their test, and did that make any difference? If you check the above equations, the one for P2ideal=P1(CR)^1.4 doesn't depend on ambient temperature! At first that seemed counter intuitive, but when I checked the derivation it was clear that only the ratios of pressure, temperature, and volume were important. I guess the result only depending on the volume ratio = CR is the most intuitive. After some additional thinking about what's really going on, I came up with the following.

My equations were derived assuming an adiabatic process in which heat does not enter or leave the system. The system in this case refers to the piston, cylinder walls, and cylinder head, all of which were evidently quite cold and thereby can more readily absorb heat from the compressed air, so a pure adiabatic assumption is probably better suited to a compression test on an engine that's been warmed up prior to the test.

It's informative to compare the results for an isothermal process with those for an adiabatic one. As air is compressed in an engine, the pressure increases by two different mechanisms. First, the volume is decreased and this increases the pressure according to (V1/V2)=(CR), and second, as air is compressed it heats up and this increases the pressure according to (T2/T1). In an isothermal process all of this "compression generated heat" is removed from the system, and this results in ideal compression readings given by...

P2ideal=P1 (V1/V2)=P1(CR) = 14(17.5) = 245 psi, and...

T2ideal=T1, since all the "compression generated heat" is removed.

In summary, both methods (adiabatic and isothermal) for estimating maximum compression readings are idealized in that they're based on perfect seals so that no air molecules escape during the compression process, and the adiabatic method also assumes that none of the "compression generated heat" escapes. What we're really trying to determine with a compression test is the integrity of the seals, but as the above illustrates, interpreting the readings is complicated because a large portion of the pressure reading is due to compression generated heat. Also due to the inertia of air, when some leakage occurs the compression readings depend on cranking rpm and this further complicates interpreting the results.

 

Wouldn't it be alot easier just to look at the compression tables in the service manual?

 

These posts confuse me way too much.....

 

I'm with you man. I'm better off with the "Dino -vs- Synthetic" or "what is the Zoo Dad mod?" posts.

 

I am with you Gene, I may not understand half of what your peddlin(at first) but after I let it sink in awhile it makes sense. Like Tenn01, I am no math stud, so please keep talkin in slow deliberate words( we're from the South and can't understand some of them fast talkin Northern folk) . Although I am amazed, at your disregard for engineering principals, when it comes to the mounting of your peripherals and subsystems and the Duct Tape. Keep this stuff coming.

 

But gene's from TEXAS.

 

Exactly ,so he knows what I mean. I was not indicating that he was one of those fast talking Northerners, just merely stating that....AWWW, Hell, Its a southern thing, you just wouldn't understand.

(disclaimer: Any Yankees, northern folk, Snowbirds, fast talkers,etc... please do not take offense to any previous or future comments about said heritage, as no harm was meant by afore mentioned comments. Any similarities between any persons living or dead are purely coincendental)

 

Your conclusion is possibly that glow plugs are required for high time engines ?

yes the 45 degrees should make a difference in the cylinder seal since the pistons and rings ought to be about .005 smaller, and it will take a bit of time to get them growing properly. might even loose up to 100 psi

my $.02 your mileage will vary

 

im confused as hell, but my truck starts without GP cycle on the first crank in 65* wether, so i think my engine is fine?

i hope so

 

So,, Let me see if I understand,,,If you,,,,,and then you,,,,,but if you have to,,,,and it starts,,,,Hmmm,,,,I think I need a beer.

 

Could you please post them here, so I and others can see how they compare with my theoretical predictions. Also if there are any special procedures prior to making the measurements like warming the engine, etc... If you compare the two methods 770 psi vs 245 psi, you see that most of the pressure in a test comes from "compression generated heat" so I'd like to see where a "good" and even a "new" engine falls between these two values. Also, if anyone has actual data from their own tests please post it.

 

We did the "wet test" by putting a teaspoon of oil down the hole, then doing the compression test again.

Dry we got 300 psi from this cylinder very cold - about 40-45 degrees.
Wet we get an additional 20 lb psi from the same cold cylinder.

What does this prove? Not sure but interesting!

 

Well I have some difficulty finding things in the shop DVD from time to time. I did find this:

from sweethaven on a quick search at this site.