sancho

Hey guys,

I was driving my car tonight with the radio off (in awe at all new, unpleasant noises my car makes) when I realized a particular sound which has me concerned. It's really subtle and you wouldn't notice it if you weren't really listening for it, but it's a semi-high pitched knocking/tapping sound which is in time with the engine speed. It happens regardless of whether or not the transmission is in gear, and speeds up when I give the engine gas.

I listened under the hood after I parked and it's pretty difficult to pick up from the engine bay between valve-train noise, injector noise, and fans. It's inaudiable when I give it gas, but I noticed that I could feel a vibratory shock in the throttle body with my hand which was in time with this sound.

Oil pressure is normal.

I've never had first-hand experience with rod-knock before and I hope I'm not going to anytime soon, but does anybody know of any good tell-tale signs which will help me discern whether or not this is really rod knock (short of pulling the engine out/apart)? Does rod knock usually get louder with engine speed? Can it be heard from within the car?

Thanks

Elephantismo

i'm pretty sure your car has a knock sensor. It's setup to go off if the motor is pingin so it can adjsut timing and fuel, but i think it would pick up rod knock too. Enging knocks are loud and deep; u cna feel them through the whole motor. Get a stethescope and listen for the tick on the valve covers, the heads, the front of the block the back of the block, the oil pan, the oil pan mounting bolts. If its loud on the bottom its prolly rods or bearings. If its loud at the valve covers its prolly a lifter. . Lots of stuff can tick tho. Transmissions can tick(auto and manual though different noises) even when out of gear. Exhaust shakes can create ticking noises. The motor shaking can physiclally hit stuff to make ticking noises. A/c compressors and other accesories can tick, lots of stuff. My most recent tick was a piece of slag rattling in my exhaust from the last time it was welded up. i also had a dipstick that was bent wrong and clicked on the crankshaft(Yikes!).

sancho

Thanks for the reply, Elephantismo!

You're correct about my engine having a knock sensor... as to whether or not it does anything, I'm not sure. I rebuilt the engine but never did change the knock sensor. Also, I never had torque specs on it, so I'm not sure if it's in there too tight/too lose, or if any of that will affect its operation at all. What would be the effects of running with it disconnected? If it is indeed making the ECM retard timing, would disconnecting it restore "correct" timing?

I will say that I have noticed a loss of power at the low end lately. But then again, this car has always been very "finnicky" with if you ask my ***-dynomometer. Sometimes I give it just a little bit of throttle and it will pull hard, then other times I can smash the pedal to the ground and not even loose traction. But that's almost another topic... I just thought I'd toss that in.

Elephantismo

I'm thinkin the knock sensor has got to send a signal whenever a mechanically triggered switch inside is triggerd. Whenever the sensor gets shaken as though a "knock" would do it, it completes a ground through the block and the ecm detects that. Additionally it could use the movement to generate a small voltage which the ecm would detect. Then it adjusts mixture acc'dingly to eliminate it and throws a code.
Either way, with the sensor disconnected(or broken) and the ECM's normal mixutre being the neccesary mixture, it will work fine. If the sensor is bad, it could knock without throwing a code.
Usually motors wont ping or knock at idle, they just die. It is neccesary to use a stethescope to find the ticking i beilive. Pinging at idle would be pretty loud, as it will echo out of the exhaust and intake. A ticking lifter doenst get as much reverb off air and a rod knock gets even less. A stethescope will point you in the right direction.

sancho

I see. Yeah, at the time when I rebuilt this engine a couple of years ago, I wasn't nearly as knowledgable about engines as I am now, so I wasn't fully certain of what the knock sensor did. I still don't know the proper way to put it in, but my guess is that putting teflon tape around the threads might not have been the smartest idea.

So if the knock sensor is detecting excessive knock--either appropriately or inappropriately--would disconnecting it restore optimum timing and, thus, improve performance? I know that's not the most insightful means of diagnosis, but it's about the only way I can think to see what the knock sensor is telling the ECM without a scanner.

Vader

Well, that's all great for the detonation sensor, but the sensor isn't going to cause any audible sounds.

If the stethoscope is inconclusive, try removing the spark plug wire on each cylinder - With plug wire pliers, of course. You can disable cylinders one at a time to see if the sound changes. Wrist pin, and sometime rod knocks, can actually get louder with a disabled ignition on the damaged cylinder.

Elephantismo

I dont think we are looking to knock sensor as the culprit here.
Grab you an automotive stethescope, with the long metal rod on it and point around and see if you hear any thing louder on one spot than another.

sancho

Thanks for the info. I'll try the plug-pulling thing, but it will have to wait until I get back home on the 10th or 11th of May (I'm at school now and don't have any safe ways to get underneath my car).

Incidently, I drove my car around during several different parts of the day on Sunday (again, with the radio off and windows closed) and didn't notice the alleged knock. I even opened the hood and prodded around for the same sound but didn't notice it.

* * * * *

I'm still interested in what the knock sensor is up to, however. While it certainly won't cause knock/detonation, it might give some hints as to why the engine power feels so inconsistent (i.e. strong some days and can't break traction other days). The theory is that it's picking up stuff that either is or isn't there and telling the ECM to do strange things to the timing, adversely affecting performance.

sancho

I remember what else I was going to ask... Instead of disconnecting plug wires, wouldn't disconnecting individual injectors have the same effect? I'd rather have my hands on 12V than 20,000V or whatever it is.

(I should have mentioned that I have an L98...)

Elephantismo

dont pull the wires @ the plugs, pull them at the cap. Get some thick rubber gloves (stuff they sell with paint stripper). Might not stop it, but i havnt been shocked through them. Then agian i usually pull them with my hand and ive only been shocked twice, once when i wasnt even working on the dist. The shock isnt that bad anyways... its just kinda tingly .
If you really dont wnat to touch the wires then use a pari fo pliers with insualted handles for your hands padded teeth for the wires and a screw driver if they are angled head. Pry up gently at the base of the boot and pull the part just past the angle with the pliers.

sancho

Well, maybe I'm crazy, but it's not doing whatever it was doing anymore. I haven't pulled any plug/injector wires yet, but the mysterious periodic tapping isn't there anymore. Guess it's a good thing it wasn't a knock, tho! (I hope)

Thanks for the help, guys.