Info:Diagnosing

Info:Diagnosing


Up

 

Compression Problems

Compression: You can't leave home without it. It's one leg of the tripod that supports internal combustion, just as important as fuel and ignition. Regardless of all the geewhiz technology in modern engines, that basic physical fact is exactly the same now as it was well over a century ago when the Otto cycle was invented. So, underneath that hightech exterior there's still just a piston pump.

Sure, today's cars typically go a long, long way before they develop conditions that reduce compression. But the key word here is "typically" -- there will always be plenty of exceptions. Then there are vehicles that have exceeded that long, long way, but are still worth fixing.

The following is intended to help you cope with cases of weak compression. Since nothing will damage the pride you take in your knowledge of cars more than a misdiagnosis, I'll devote plenty of space to testing procedures. And I'll add some practical advice on policies and decision-making.

Hark

 

The first thing to do is listen carefully, both to what your brain is trying to tell you about the symptoms and to the power-plant itself. Asking a few question in your head will help: Is the idle uneven? Are power and fuel mileage declining? Does it smoke and use a lot of oil? How about backfiring, hard starting, or high emissions that resulted in failure of a state pollution test? Has the engine been properly maintained? What work has been done on it? Take the time to be sure of your answers.

Start it up and note if it cranks unevenly or takes a long time to fire up, both of which suggest poor compression. Go for a ride to confirm the complaint.

If you happen to have access to a scope, use it to find out if the ignition system is in good shape -- it would be pretty embarrassing to pull a head only to find that a bad plug, wire, or cap was really the culprit. If you had a computerized engine analyzer, you could do electronic or relative compression and cylinder balance tests.

Of course, most do-it-yourselfers, and even some technicians, like to get an initial idea of the situation with a manual cylinder balance check. You know, where you pull one plug wire at a time to find out if a particular hole has little or no effect on idle quality and speed when disabled. This is still about the most useful troubleshooting trick known to man, especially where there's a definite miss. That is, if you remember to disable whatever computerized idle stabilization device is present. And use well insulated boot pliers or that hotstuff electronic ignition might blow your pacemaker. Where possible, kill injection instead of spark.

Sucker test

 

A vacuum gauge can be helpful at this point, although its readings may be inconclusive or ambiguous. You'll get the most useful results at curb idle speed with the engine fully warmed up. A typical healthy power-plant will produce 1520 in. Hg.

A steady low reading may be caused by a vacuum leak or late valve timing due to a worn or jumpedcamshaft drive mechanism. If the needle drops at regular intervals, suspect a leaking valve, whereas if such drops occur irregularly, a sticking valve is indicated. Floating over a wide range suggests a bad head gasket seal. Rapid needle vibration is evidence of loose valve guides.

Since back-pressure can interfere with cylinder filling, check for a clogged catalytic converter or crimped pipe by holding 2,500 rpm. The reading will drop when you first open the throttle, then stabilize. If it starts to fall afterwards, there's probably an exhaust restriction.

PSI

 

Whether or not you've isolated a cylinder or two as the source of the problem, it's time to unscrew the spark plugs and do a traditional dry/wet compression test. You probably already know how to do that, but here are a few subtleties that'll help you avoid a costly mistake:

  • On aluminum heads, better loosen those plugs with the engine cold, then just snug them back down enough to fire it up.
  • Readings will only be accurate at normal operating temperature.
  • Blow out the plug wells or flotsam and jetsam could hold a good valve off its seat.
  • Pull all the plugs at once to make cranking easier.
  • Make sure the battery and starter are up to the task of achieving normal cranking rpm.
  • Block open the throttle plate and, if you're working on a car with a carb, the choke.
  • Disable the ignition, preferably on the primary side. Letting those dangling wires arc won't only zap the secondary circuit components with more voltage than they might be able to take, it's also asking for an explosion.
  • Even though the clear-flood mode is supposed to halt injection during WOT (Wide Open Throttle) crank, you can be doubly certain to eliminate gasoline spray by shutting down the fuel pump and blowing the residual pressure through the rail's test Schrader into a rag, or by unhooking injector connectors.
  • You need at least four pulses per cylinder.

Interpretation

 

If a low reading jumps substantially after the addition of a few squirts of oil, you've got a ring/bore problem. On the other hand, if wet readings are only slightly higher, and this rise is roughly the same for all cylinders, valves are implicated.

The difficult part is judging how much variation among cylinders, or between dry and wet readings, represents a serious problem. Say you've got 80 psi in one, but about 120 in the others, and adding oil brings them up only five psi or so. Is a valve job necessary?

That depends. Obviously, the low one is leaking somewhere, probably through an exhaust valve, and erosion is going to make it get worse pretty rapidly, so the proper thing to do is get in there and attend to the seats and faces. On the other hand, if it's not bad enough to cause a miss yet, and you've been frittering away your money on luxuries like food and shelter so you can't afford major work this month, or if the car is a nearly worthless heap, maybe you can simply live with it. Just make sure you understand that no amount of tuning or other external attention will make that engine run any better or go any farther before that cylinder loses it altogether. At least there's some good news -- the rings are okay.

Poor pressure in two adjacent cylinders should make you think about a blown head gasket. Confirm this by looking for coolant in the oil or on the spark plug, and by checking for evidence of compression in the cooling system. Professionals hold the probe of an exhaust analyzer over the radiator filler neck to see if they get an HC reading, whereas a do-it-yourselfer might remove the thermostat housing and water pump belt, then watch for bubbles.

Pump it up

 

Gauging compression isn't the only time-honored procedure that's still useful for assessing an engine's ability to squeeze air. There's also the cylinder leakage test, which is done by pumping maybe 90 psi into the spark plug hole with the valves closed (remove the rockers, or lock the crank at TDC of that cylinder's compression stroke), and listening to where it escapes. Hissing at the intake points to the inlet valve, and the same sound at the tailpipe indicates the exhaust.

There'll always be some noise at the oil filler hole because even the best rings can't seal completely (gaps, you know). The trick is to tell when it's excessive, which you can probably do by comparing cylinders. And this test is great for fingering a leaky head gasket -- remove the radiator cap and listen.

An improvement on this theme is the use of a gauge that lets you know what percentage of the available pressure is escaping.

No conclusions yet

 

There's more to throw into your mental threshing machine before you make your grand diagnostic pronouncement. Anything that holds a valve open, such as a broken spring or a sticking guide, will certainly cause a low compression reading. While these can usually be fixed without removing the head, chances are the valve is burned to a crisp (it can only cool when closed, after all) and/or bent. You can try making the repair and see what happens, but there's no guarantee of success.

Low cylinders that don't produce more pressure after oil has been introduced into them may not have valve sealing trouble. A wiped cam lobe or a pulled rocker stud can result in a miss because the cylinder isn't being properly packed. Check lift and stud protrusion before you start unscrewing head bolts.

Absolutely no compression in a cylinder does not necessarily mean a valve is stuck open or burned away, either. There could be a hole in the piston, and I remember one Triumph engine that was still running, albeit roughly, even though the piston was entirely gone along with the whole rod so that when I yanked the head was looking down on what was left of the crank pin.

In cases where the car suddenly refused to start and you got weird compression readings, a snapped OHC drive belt or a jumped timing chain is a more likely possibility than bad valves, which deteriorate gradually. If the power-plant isn't freewheeling (that is, the valves hit the pistons if they're out of synch), however, you're going to have to remove the head anyway for the replacement of some bent stems.

Probabilities

 

As far as probabilities are concerned, the most likely culprits are burned valves, a violated head/block seal, valve train or lift troubles, failed rings, and damaged pistons, in that order.

With everything except failure of the parts that actuate the valves, the head will have to come off, and it makes sense to service the valves and replace the stem seals even if the head gasket was the problem. In cases where only a couple of valves were leaking, it would be pretty silly to just fix them since the others are probably on the ragged edge of survival. Do a whole valve job.

Do it yourself?

 

That brings up another subject: Should you do the head in your own garage, or ship it out to a machinist?

Your particular skill and equipment situation will have to determine that, but we can offer some general observations. Commonly, you'll find one or two burned valves, the others starting to pit, and seats that look pretty good. The guides will often be a little loose, but usually not beyond what knurling can rectify. A head that falls into this category is a prime candidate for your personal treatment.

But you'll also run into cases where the guides are terrible, and, especially with aluminum heads, the sealing surface is about as flat as a washboard. Cracks are relatively rare, but certainly not unheard of. When you see any of these serious problems, it's logical to make a trip to the machinist.

For instance, a crack isn't necessarily the kiss of death. Heads are often repaired using various welding techniques, but you've got to find somebody who knows what he's doing or you're risking a monumental failure. If the engine is common, you might do better shopping for a used unit at your favorite wrecking yard.

Then there's sealing surface flatness. Everybody knows about milling and grinding, but I never cease to be surprised at how few people out there, professional or otherwise, are familiar with heat straightening, wherein a warped head is bolted to a plate and baked in an oven until it relaxes into its original shape. This can save an expensive casting that would otherwise have a future as raw material for about a million beer cans.

Conditions that interfere with valve lift are next in frequency. Sometimes, there'll be a rash of wiped cam lobes with a particular engine -- the 305 Chevy and the Pinto/Capri 2L come to mind -- even though regular oil changes were done and the rest of the power-plant is in excellent shape. Here you won't be risking much by just replacing the cam and lifters (with the roller type, you can reuse the old ones).

When it's the rings

 

It's sensible to make a repair when low compression is due to burned valves, a blown gasket, or a wiped lobe, but defunct rings are another story. Simple ring jobs were common in the past, but those circular cylinder seals seem to last just about forever in modern engines. When you run into a case where they're worn out or jammed in their grooves by deposits, you've probably got a completely trashed power-plant on your hands. Regardless of whether the cause was neglected lube changes or severe overheating, you can't just get in there, hone, and install new rings and expect much more service out of this specimen.

That means you've got to consider several alternatives: the patch job overhaul just mentioned, a proper rebuild (boring, new pistons and pins, reconditioned rods, reground crank, etc. -- safer and more profitable), or the installation of another engine, which may be factory remanufactured, brand new, or even recycled from a wrecking yard.

Choosing among these alternatives is a complicated mental process because so many factors are involved. Each individual case will have its own unique set of circumstances that you have to weigh carefully along with your capabilities and time situation.

If you insist on doing an old-fashioned ring job, at least make sure the bores are round and straight, and check the pistons for scoring and excess skirt clearance. Select fit those rod bearings, too, and don't be tempted to slip them over grooved, tapered, or out-of-round journals. While the pan's down, you should probably think about replacing that oil pump.

Greg McGoniga, former NAPA/ASE Technician of the Year, put the situation in perspective for me from a professional's point of view: "Remember, the technician's the guy the customer's paying to tell him how it should be fixed. The tech shouldn't let the customer's finances dictate the quality of his work."

Wretched excess

 

I'll conclude with some notes on the opposite situation: too much compression, which results in detonation and all the mechanical havoc that condition can wreak. There are two causes. One, deposits thick enough to significantly reduce the volume of the combustion chamber. And, two, head machining or cylinder overboring, both of which boost the squeeze ratio.

I'm old enough to remember a job called "manual decarbonizing," which entailed removing the head and scraping away those sooty black deposits. Today, you'll probably start off with some top engine cleaner, or perhaps take it to the rare shop that does walnut shell blasting. But in really bad cases, you may have to get inside.

When it's the result of machining, the most cost-effective cure is the installation of a proper shim or a head gasket that's thicker than stock, depending on the application. Our responsive aftermarket has made this simple solution widely available.

As knock sensor circuits appear on more and more cars in our vehicle population, however, automatic timing compensation will make this problem fade away.

 

 

DL3SJB's World


What's new ?
Projects
DB0SAO
DL3SJB
Feedback
Impressum


Last updated 05/31/04. Best viewed with your eyes wide open. All information provided is of a general nature and is not intended to address the circumstances of any particular individual or entity. Although I endeavor to provide accurate information, there can be no guarantee that such information is accurate as of the date it is received or that it will continue to be accurate in the future. No one should act upon such information without appropriate professional advice after a thorough examination of the facts of the particular situation. You are visitor number