Hallo, vor einiger Zeit habe ich versucht meine Drehzahlproblem bei warmen Motor (etwas niedrig) zu schildern. Ich hatte daraufhin im Forum nichts wirklich meinem Problem dienliches gefunden. Da ich auch keine Fehlercodes habe, habe ich mit meinem bescheidenen Wissen versucht das Problem um die Thematik Luft/Benzin Gemisch einzukreisen. Resultierend daraus habe ich vor einiger Zeit schon Vacuum Schläuche wechseln lassen, Drosselklappen reinigen lassen. Das bescherte Besserung aber keine Zufriedenstellung. Vorgestern habe ich das PCV Ventil getauscht, aber auch das brachte mich nicht entscheidend Weiter. Bis ich gestern den gleich erscheinenden Artikel gefunden habe. Ich habe zwar nicht die Testprozeduren nachvollziehen können (a) fehlt es mir an Know How und (b) an den Geräten, aber das Problem beschreibt exakt meines und einen Sauerstoff Sensor habe ich mir vor einiger Zeit mal auf Verdacht gekauft und siehe da. Kein Problem mehr. Hier nun der Artikel. Der dürfte vor allen die versierten unter Euch interessieren: PS: Mein Proble war Problem 1: ------------------------------------ I was playing around with my 1986 Le Baron (2.5 liters TBI with a/c) and thought that some of the more technically inclined readers may be interested in my experiences. Here are the problems I was addressing: 1. On a hot engine, the idle would "randomly" drop abnormally low (but not low enough so as to stall the engine) and would be corrected. During the correction, the idle sometimes overshot what appeared to be the desired idle rpm. This problem has been around for the past few months, but it's been COLD in that garage... 2. Intermittent engine knock developed in the past two weeks. 3. Records of fuel efficiency showed a marked decrease in efficiency (in miles/gallon) in the past two weeks. Now, for the first problem... A weird idle problem on a hot engine (or, for that matter, ANY drivability problem under this condition) usually implicates the feedback control system. However, this car had a new oxygen sensor (replaced 13 months ago). As expected, no trouble codes were stored. I connected an oscilloscope to the O2 sensor and let the engine idle. My opinion was that the sensor's output wasn't "steep" enough (the transition from a high reading to a low reading and v.v. should be FAST -- almost a step response). The reference voltage the computer was feeding the sensor was 438 mV, not the 450 mV I'd expect (but this isn't a big deal -- computer systems put an "error band" around the 450 mV threshold and ignore sensor readings in this band -- this is like trying to operate a digital logic gate in the "forbidden region"). I next decided to check the average a/f mixture. The easiest way to see what the a/f mixture is is to use an ignition oscilloscope. Since I don't own one, I did the next best thing -- I 'scoped the primary side of the ignition coil. On a well operating engine, here's what I'd expect to see: 1. Firing line (ignition voltages) about the same for all cylinders -- no more than +/- 20% variation between the highest and lowest. If an ignition oscilloscope was available, I'd expect to see the maximum differences to be less than 3 kV. If the firing line is jumping all over then the a/f mixture is screwed up. 2. Spark time ("burn time) to be in the range of 0.85 ms to about 2.2 ms. Also, the spark line should be somewhat jagged AND FLAT. A line which rises upward toward the end of the burn period indicates lean misfire. Typical spark time is around 1.6 ms. 3. Firing line to spark line voltage ratio should be about 4:1. 4. At least 5 coil oscillations after the spark ceases (some computerized diagnostic gear -- e.g. Sun's "Modular Computer Analyzer" -- look for a minimum of 3 oscillations). In my case, the firing line to spark line ratio was substantially higher than 4:1. This is attributed to using non-Chrysler ignition wires which have MUCH more resistance than Chrysler's specs (these are Autolite wires). The firing voltages were almost all the same (indeed I couldn't tell the difference between the cylinders). Spark time varied from 1.9 to 2.1 ms and the spark line was almost ruler-flat. Around 10 coil oscillations were present. This is a classical example of (1) abnormally high secondary resistance (2) perfect ignition coil (by the way, this is a Wells coil -- about a year old) and primary circuit, and, possibly, an engine running rich. Temporarily shorting out one of the cylinders showed that the firing voltage for that cylinder did not change appreciably -- hence, confirmation for the fact that the secondary resistance is too high. Also, the spark voltage did not change appreciably. Based on these observations, I replaced the oxygen sensor. The new sensor's response is much better (closer to a step function). Why did the previous sensor fail? The most probable answer is that the engine blew its head gasket a month after the previous sensor was put it. Silica in the coolant contaminated the sensor and caused its response time to increase. Eventually, the computer would have caught the problem ("eventually" meaning at least a year from now). Now, on to the engine knock problem. When diagnosing engine knock, I look at two things: fuel quality and EGR operation. Only if these two are eliminated as culprits do I look to see whether a lean a/f mixture exists. Since the idle looked ok in open loop, I discounted fuel quality. I tee-d a vacuum gauge into the EGR valve's vacuum line and, with the engine running, slightly opened the throttle. With no appreciable engine load, I expected to see about 5" Hg control vacuum. Instead, I measured full ported vacuum. To understand what this means, one has to understand how the EGR system meters exhaust gasses. The backpressure transducer has two vacuum hoses and a stainless-steel backpressure pipe feeding it. The pipe feeds exhaust gas pressure to the transducer. One of the vacuum ports goes to ported vacuum (or, the EGR solenoid if this is the 2.2 turbo or multipoint injected engine) whereas the other feeds the EGR valve. If the exhaust pressure is high, the transducer opens and allows control vacuum to move the valve's pintle up. This increases EGR. If the exhaust pressure drops, the transducer closes and allows atmospheric pressure to enter the EGR valve and close it. By oscillating back and forth very quickly, the backpressure transducer allows the EGR valve to see a smoothed control signal and thus open only a certain amount. The amplitude of this (pneumatic) signal is a function of engine load, rpm, etc. Since the transducer was applying full ported vacuum and since the engine didn't bog down when the throttle was slightly opened, the implicated culprit is the EGR valve. I connected a vacuum pump to the valve and pumped up about 15" Hg vacuum. The valve's pintle did not move. A new valve/transducer assembly will be put in within the next two weeks.