Interesting! – Loose parts in engine bay caused the vehicle to stall
Read here how a simple and dumb thing was causing a car to completely stall at intermittent times, with no apparent cause. This might be happening to your car…
For some time, we were wondering about the cause of an intermittent power loss on the car of one of our customers, that seemed to have no cure, as all the engine parts and electronics we checked tested ok.
As we made the tests, the problem was tracked down to possible problems with the ignition system, specifically suspecting of too low ignition timing advance. We went to that conclusion, as the reading with a scanner was a sudden drop in degrees, of ignition timing advance, to a lower than usual level when the problem started showing.
Logically, we first checked the spark plugs, ignition wiring and ignition parts for damage and since those parts tested good too, we took out the ecu (engine computer) for inspection, almost sure that the problem was there. To our surprise, the ecu was good too. Running out of ideas for the possible cause of the problem, we observed one more time the collected facts and looked like the problem appeared only right before a drop in the reading of the ignition timing. Of course, lowering the ignition advance by several degrees will cause a loss in engine power. In this case more than a loss in power, it made the car to stall when the problem showed.
Now, with the facts we collected and confirming that it was a drop in ignition timing, we started looking in our diagrams in search of electrical components that could be associated with the sudden drop in ignition timing under normal circumstances. I say normal circumstances because until now, we couldn’t find any damaged or faulty part.
By consulting the diagrams, we found the only thing that might be causing that drop in ignition timing. We saw a knock sensor wired to the ecu. We were not aware that the Korean model we were working on had any knock sensor in its system’s circuitry. We immediately replaced the sensor convinced that we finally hit the problem. We were shocked to see that the problem came back after a few minutes of driving the car. The worst thing about such problem was that it was an intermittent failure and we were not able to duplicate it all the time, making it more difficult to troubleshoot.
Luckily, it occurred to us that if the knock sensor does its work by detecting “knocks” or detonations, then either, detonations might be occurring unnoticed, producing the drop in timing, or something similar to detonations was occurring. We found out that the sensor was really doing its work, but it was not sensing any detonations, but a very similar sound coming from a loose metallic part near the exhaust manifold. Any time that the loose part vibrated in a certain frequency, the knock sensor detected it and the ecu dropped the timing to protect the engine in case that detonations occurred. It all was part of the many program routines integrated in the engine computer for protection of the engine. Detonations might brake the engine and once they occur, can be prevented or at least lessen by lowering ignition timing advance, and so did the “confused” ecu. After several seconds that knocking or detonations cease, the ecu will move back the ignition timing to where it was and will continue doing corrections.
The knock sensor really is a piezoelectric microphone, “hearing” at the engine all the time. A piezoelectric element is used due to its capacity to pickup higher frequencies, around the frequency of the signal that an engine knock produces when a detonation occurs, also known as “pinging”. The signal it generates by “hearing” the engine, is then processed by the ecu. A complex process is done by a microcontroller in the ecu in the “knock sensor amplifier” circuit, to filter out the “heard” signals. This process will discard regular occurring noise and take action on real knocks or detonations. Even when the filtering process is complex enough, the vibrations of this metallic shield were either, not filtered out because of a glitch in the filter circuit design, or the generated noise was simply too similar to engine knock, fooling the filtering circuit.
Below are two examples of knock sensors. A knock sensor, being a piezoelectric part, when tested with an ohmmeter, it should measure infinite resistance or a very high value, from 500,000 Ohms to several MegOhms (1×10^6 Ohms).
Engine knock is very specific in every car or application and it will not “sound” equal in all models. For us, all engine knocking will sound the same, but for a sensitive, high speed computerized circuit, it will have different sound in different models, which offer different environment. That is why, every ecu computer has a different arrangement or programming for the model it was made for. Some car manufactures, use a generic ecu for several car models that are similar, but they put a different programming on each one, suitable for that application. GM for example, used to put similar ecus in different models of the same line, but each one had a different programming and besides that, each one used to have its own detachable knock filter circuit, so it could be transported to a new ecu should the ecu needed replacement.
Well, I think that the bottom line is to keep all mechanical parts tighten, in place and clean, as the minimal thing may negatively affect the engine of your car, due to the complex computerized systems installed on them now a days. Something that looks inoffensive, may cause lots of untraceable and annoying troubles in our vehicle. I hope this experience helps for correcting the same or similar problems on other automotive cases not solved yet.
Also Interesting…
Website's recent related searches:
OtherDeal.Com
Comments
No comments yet.