Tech To Tech
Relative Compression Vs. Trouble Trees
I have never been a fan of diagnostic trouble trees and I have
met very few techs who have used them with any success. This is
why I refer to trouble trees as "the trail of broken dreams." Too
many variables are not considered by trouble trees and they are no
good for intermittent problems. A tech with a good understanding of
electronic principles and a circuit diagram should always fair better.
One of my primary complaints with trouble trees is the disclaimers they start with -- check basic compression, check all grounds, check charging system, check fuel quality and fuel pressure. This appears to me to be about three hours of work or more on many late-model engines. It would be nice to be able to bill these three hours before we start our diagnosis, but let's get serious -- most customers want a faster, less expensive path to the solution. You also want to bill appropriately for the time you spend on a job, so the faster the solution, the better the billing time.
OK, I probably lost a few of you, and some people are probably starting to get upset. Before the letters start coming in, let me clarify what I mean. First, I think we can all agree that the No. 1 enemy in the shop is wasted time. Time is the main thing we sell--parts follow time. While we sell our time, what we really sell to the customer is value. Let's say you invest 40 hours and $1,500 in a customer's car and still have not solved the problem. Is there any value here? Is this billable time and parts? Now on the flip side of this coin, let's say you purchased a $20,000 machine that helped you to fix this same car in half an hour. Is this job only worth half an hour? Or is there more value here? Yes! We need to charge for driveability work based on value and not time. So when I say repair in less time, I don't mean charge less.
Now back to trouble trees. Basic compression ... are you going to test this on every car? Not in my shop. The probability of the problem being here is so low we don't feel we can justify this charge on every driveability problem. Besides, who wants to call customers to tell them they need new spark plug wires from all the torn boots from doing a compression test?
So what's the solution? A "relative" compression test using a scope and a current probe on the battery cable. Let's consider Figure 1. This is a relative compression test. If all the cylinders are low compression, the waveform will have reduced amplitude, but all cylinders will look the same. Usually we are dealing with a single cylinder compression problem that might be due to a problem such as a burned valve or a flat camshaft. The waveform in Figure 1 was taken from the positive battery cable on my 300 CID Ford van with the ignition disabled. Despite the poor service I give my own vehicles, it has very good compression. This is evident by the evenness of the peaks and valleys (all cylinders are the same), and the difference between the height (amplitude) of the peaks and valleys.
With this method we're talking five minutes max and you've got compression behind you. This is nothing new -- tests like this have been available in scopes like the Allen SEA for years. What is new are the hand-held digital storage scopes combined with a current probe that make this test faster and easier (you don't have to roll over the giant machine). Also the equipment to do this is much cheaper than the old giant machines, making this more available to every shop and tech.
The waveform in Figure 2 is my Ford van with the No. 1 spark plug removed. Notice the large dip where a peak should have been, showing this cylinder has no compression at all. Obviously, reduced compression would fall somewhere in between and the relative compression of that cylinder can be estimated from the difference in the peak height. The high point in Figure 1 is 220 amps and the low valley is 175 amps. I have not tested enough vehicles to give good general values for this test -- perhaps someone out there has a tip on this?
There is more information here than may be immediately obvious. The cranking RPM can be determined using the following formula. This van took 600 milliseconds (ms) to complete all cylinder compression strokes or 2 RPM. Now divide 1000 (that's the ms in a second) by 600, and you get 1.67. Now take that 1.67 multiplied by two (that's the RPMs completed) multiplied by 60 (that's seconds in a minute). The result is almost exactly 200 RPM. This technique is invaluable on a diesel engine where compression fittings are hard to come by and cranking RPM is so critical to starting. In addition to engine data, this also gives a good idea of the starter condition.
There are some exceptions. Some Hondas have a problem with carbon on the intake valves getting so bad that the car fails to develop enough compression to start. You need to take into account total current draw and the amplitude to find these problems, as all cylinders on this car will look the same. Be aware that battery, cable or starter problems may cause erroneous results.
Figure 3 is a recording with all spark plugs removed. Note that some compression is still achieved against the open spark plug hole. Also notice that it still takes about the same time (three scope divisions) to complete six compression cycles. It's also interesting to note the current draw isn't that much less with all the plugs out. Why isn't the current draw less and why isn't the RPM a lot more? Motor current draw (on a good non-shorted motor) is relative to its RPM. As long as the load is not so great that it slows down the motor, the current draw should be fairly constant. If a piston drag existed from a partial seizure, it should show in this waveform. If partial seizure had occurred you would probably notice lower draw on the affected cylinder in the first pattern (with plugs in) and higher draw on the last pattern (plugs out).
Figure 4 shows a higher resolution pattern showing good starter windings and brushes. At an average starter to flywheel ratio of 15 to one and cranking speed of 200 RPM, the starter is turning at 3,000 RPM. If you do the math on this waveform, knowing there are two power brushes, it computes out to this starter having 20 commutator segments. I love these current probes!
AutoInc. || More Info About AutoInc. || Tech Tips || Tech To Tech
Stat Corner || Chairman's Message || News Briefs || News Briefs Extra
AutoInc. Magazine ®, Vol. XLIII No. 12, December 1995