ECU code 9 cam drive problem
Posted: Sat Mar 15, 2014 12:25 am
I bought a cheap 1988 Si 4WS about a year ago and dealt quickly enough with the oil leaks, but I was left with the problem of severe bogging below 3000 rpm when the engine was warm. The fact that it ran well when cold gave me some confidence in the general condition of the engine and a compression test came up OK. Anyway, it took me months to get to the bottom of the problem, and having now fixed it I thought I'd report it in case anyone else encounters something similar on high-mileage engines.
The code 9 (cylinder sensor) appears to set when there is less than about 340 degrees (in crankshaft terms) between the CYL pulse and the TDC pulse from the distributor. That should pick up one of the camshaft pulleys being one tooth out on the belt, but that was not the issue here. I found that when i went to set the ignition timing, it would drift further and further retarded until I ran out of travel in the adjustment slots of the distributor. Being new to fuel injection systems I blamed this on the ECU and spent ages looking for faulty inputs, but to cut to the chase, it turned out to be a purely mechanical problem: the intake camshaft pulley was loose on its shaft, and the Woodruff key was flogging out the slots in the pulley and shaft. There was a good 15 degrees of play, which is 30 degrees at the crankshaft. So the main effect was to retard the ignition, but it may also be that the inlet valves were lagging with respect to the injection pulse, which i believe is triggered by the CYL sensor on the exhaust camshaft.
Now you may wonder why it was not obvious that the camshaft pulley bolt was loose, assuming that was the root of the problem. It actually took a fair effort to release the bolt. It seems that the head of the bolt was stationary with respect to the pulley and the camshaft was jerking back and forth on the thread of the bolt. With the threads being fairly dry, the resultant galling prevented the bolt from unscrewing.
This is obviously a rare problem, since a Google search didn't show up any other mentions of it. On the other hand I did come across one or two mentions of symptoms very close to mine, and it could account for obscure drivability issues in less extreme cases. The main giveaway would be the appearance of brown metallic dust on the back of the pulley (from fretting between the pulley and the camshaft) and it might also be detectable by putting a large shifter on the 26mm hex part-way along the camshaft and looking for play between the pulley and the shaft.
The code 9 (cylinder sensor) appears to set when there is less than about 340 degrees (in crankshaft terms) between the CYL pulse and the TDC pulse from the distributor. That should pick up one of the camshaft pulleys being one tooth out on the belt, but that was not the issue here. I found that when i went to set the ignition timing, it would drift further and further retarded until I ran out of travel in the adjustment slots of the distributor. Being new to fuel injection systems I blamed this on the ECU and spent ages looking for faulty inputs, but to cut to the chase, it turned out to be a purely mechanical problem: the intake camshaft pulley was loose on its shaft, and the Woodruff key was flogging out the slots in the pulley and shaft. There was a good 15 degrees of play, which is 30 degrees at the crankshaft. So the main effect was to retard the ignition, but it may also be that the inlet valves were lagging with respect to the injection pulse, which i believe is triggered by the CYL sensor on the exhaust camshaft.
Now you may wonder why it was not obvious that the camshaft pulley bolt was loose, assuming that was the root of the problem. It actually took a fair effort to release the bolt. It seems that the head of the bolt was stationary with respect to the pulley and the camshaft was jerking back and forth on the thread of the bolt. With the threads being fairly dry, the resultant galling prevented the bolt from unscrewing.
This is obviously a rare problem, since a Google search didn't show up any other mentions of it. On the other hand I did come across one or two mentions of symptoms very close to mine, and it could account for obscure drivability issues in less extreme cases. The main giveaway would be the appearance of brown metallic dust on the back of the pulley (from fretting between the pulley and the camshaft) and it might also be detectable by putting a large shifter on the 26mm hex part-way along the camshaft and looking for play between the pulley and the shaft.