Tech To Tech
More On Time-saving Check Sheets
Just a little review from last month first. The car we were working
on is a 1992 Suzuki Sidekick with port fuel injection and a mass
air flow (MAF) sensor. The problem is pinging all the time, especially
under load, and loss of power after driving two hours. After eliminating
some possibilities we found the EGR vacuum to be too low to open
the valve properly. Further tests revealed the engine vacuum itself
was too low.
Low engine vacuum seems to be the problem now, so it is time to create a new check list of possibilities. Don't forget to list the possibilities that you may have already eliminated, like the second one listed below. This is important in case you need to consult someone else or double back on possibilities. How many times have we eliminated a possibility only to find later the part we replaced was defective? I always get wary when someone says, "It can't be that, I replaced it." No possibility is truly eliminated until the car runs right!
Possible Problem List:
- Engine timing wrong.
- Ignition timing or advance bad. (This already tested OK. We have not checked the timing marks against true TDC of the piston so there is still some assumption here.)
- Check vacuum waveform for abnormalities. (Broken valve spring? Valve hanging open? Valves adjusted too tight?)
- Plugged exhaust. This is unlikely because the whole exhaust system is new.
- A misfire condition, such as a burned-through rotor that affects all cylinders at a specific KV. Once again unlikely because cap, rotor, plug wires and plugs are new. However, the customer did this work and maybe the plugs are wrong or gapped wrong. Check on scope and check cylinder balance.
- A lean condition affecting all cylinders. (Consider fuel pressure, defective input, or ground or power problems.)
- Check codes.
OK, this list may not be all inclusive, but it gave us a good start to try and corral this problem in. In the process of elimination we have two possible ways to go, the easy stuff and the most likely stuff. On the most likely side, I am leaning toward the engine timing and fuel pressure. I'm going to do the easy stuff first because this is the least labor intensive and therefore gives the greatest value.
First, I checked codes. None. No surprise here. Now I'm wondering about misfire and lean conditions. Remembering to begin at the end, I tested the O2 sensor on a scope while driving. This is a fast, easy test that has the possibility of eliminating many tests. The O2 sensor sweep was good, .1v-.9v, at five times per second. A good O2 signal indicates that the engine isn't too lean, so there is no fuel starvation problem. Also, the O2 shows no problem with inputs or misfiring. This has eliminated many possibilities with one simple test.
It seems we are going back to the most likely cause, an engine problem, valve timing, or a valve opening problem. The owner had replaced his timing belt a few months back with no effect on the problem. He now had enough doubt about this repair to insist we pull the timing cover and check the valve timing. We pulled the cover and everything was perfect. There is a very important point to make here. Just because the timing marks line up doesn't mean all is well. Take the time while you are here and remove any bolts and belt guides to inspect both keyways. They seem to have an insidious way of shearing and only shifting 20 degrees or less, causing driveability problems, but not a "no start." While we were here, we knew that when the crankshaft keyway was straight up the engine was at true TDC. Holding the timing cover in place and slipping on the harmonic pulley showed that the TDC marks on the balancer were correct. One more possibility eliminated.
It seems we are getting closer and closer to the end of our list and valve problems are pretty rare. The customer had adjusted the valves awhile back and found no change in the problem. So far, all the work he did that we checked was perfect. He now requests that we check the compression. He checked the compression in the past and it was perfect, but he wants us to double-check for his peace of mind.
The compression was perfect. This was also a good time to check that the spark plugs and the plug gap were correct. Once again, perfect.
We decided to check the vacuum waveform next. We want to check the cranking waveform, the idle waveform and the 2,000 RPM waveform. By the way, I forgot to mention that not only was the loaded vacuum way too low, but the idle vacuum was never more than 14"/hg. This sure seemed low to me.
The cranking vacuum waveform looked normal (Diagram 1); nice smooth sine waves of cylinder suction. This engine uses a bypass-style idle speed control. Usually when cranking they will be wide open, so the vacuum will be quite low. This is perfectly normal. What we are looking for is variations from cylinder to cylinder.
Diagram 2 is a vacuum waveform taken at idle. Notice the difference in the humps at the bottom of the valleys. This is where the exhaust valve is closing and the intake is opening. A valve adjustment that is too tight may cause a problem here. There is no indication of a bad valve spring or sticking valves, which would cause one cylinder to be very different from the others.
Diagram 3 was taken at 2,000 RPM. (Note the radical difference between cylinders at the bottom of the valleys.) This shows the importance of testing at different RPMs. A small vacuum leak will be dramatic at low RPM and diminish as air flow/ RPM increases. A floating valve or a flattened cam lobe will be more dramatic at higher air flow/ RPM. This is why you want to do cylinder balance tests at both RPMs. More on that later.
So, we decided it was time to check the valve clearances. The factory spec was .003"-.0047" cold. Checking the clearances showed all the valves were too tight. The loosest I found were .0015" and most had almost zero clearance. I set all the valves to the loose end of the spec. If he had truly set the valves correctly in the past, then they must be losing clearance fast, an indication of valve recession in the head.
Retesting the vacuum waveforms showed all cylinders with perfect sine waves. The idle vacuum came up to 18", an improvement of 4". The 40 mph cruise vacuum came up to 5", an improvement of 3", but it still seemed low to me.
Were the ping and power problems solved? Well, this story isn't over. I'm going to have to finish this tale next month.
About the author: Mark Warren is the owner of Mark's Auto, an Automotive Service Association (ASA) member shop in Tucson, Arizona. Warren discusses underhood diagnostics and other service considerations monthly in Tech to Tech. If you have a comment for Mark Warren, you may write him c/o AutoInc., P.O. Box 929, Bedford, TX, 76095-0929, for fax to (817) 685-0225. Mark's email address is 76727.3631@compuserve.com
Regulatory Compliance -- Turn A Pain Into A Gain
Low-budget Marketing Ideas For Big-dollar Results || How To Handle Low VOC Paints
Mechanical Shops Are Going Green || Election Year Impact On The Aftermarket
Guest Editorial -- Control At Point-of-sale Is Not About Leveling The Playing Field
Tech To Tech || TechTips || Stat Corner || News Briefs || News Briefs Extra || Taking The Hill || Directions || Around ASA || Chairman's Message
AutoInc. Magazine ®, Vol. XLIV No. 10, October 1996