high pitch noise at idle

[mokmok]

Hi,

I've waited a long time before posting about this problem trying to fix it on my own but now i'm at a dead end and im open to pretty much any advice,

I have a 2007 hatch and when the engine is warmed up and i'm stopped at idle for about a minute I get a high pitch noise that comes from the engine bay. Ive tried to film it (http://www.youtube.com/watch?v=vTiOPe6qmAg), without much success.. It sounds like a vacuum leak but even when I sprayed starter fluid on the vacuum lines it didn't affect anything ...the car tries to stall and the engine shakes in a way i dont really like :frown:

Then when the radiator fan comes on the idle goes up a bit and the noise disappears. As soon as the fan shuts down it starts again

I look foward to your advice

thanks :smile:

[Bluevitz-rs]

So you fixed it?

[mokmok]

So you fixed it?

Not yet, still looking around for possible issues linked to that, i notice it runs lean when it runs badly, checked my upstream O2 and the resistances aren as specified in the repair manual so that might be it

have you ever heard of something like that?

[Bluevitz-rs]

Have you checked to make sure the. Heat shield isn't loose. It'll make a squeaking sound like that.

And you need to check voltage on the front sensor not resistance to determain your reading.

[mokmok]

Have you checked to make sure the. Heat shield isn't loose. It'll make a squeaking sound like that.

And you need to check voltage on the front sensor not resistance to determain your reading.

the repair manual states the values in Ohms not in Volts :confused: and the heat shield does not move at all :frown:

[Bluevitz-rs]

Those resistance values are for an unplugged sensor to make sure it's within static value. Those are not for when the car is running plugged in. So in order to read rich or lean you need to measure the output voltage of the sensor.

[mokmok]

Those resistance values are for an unplugged sensor to make sure it's within static value. Those are not for when the car is running plugged in. So in order to read rich or lean you need to measure the output voltage of the sensor.

That's exactly how i tested it, engine off and in my garage, so i was around the 20 degrees they recommend for testing. The car seems to run lean when it tries to stall my obd2 with bluetooth to my phone reads 15.5-16 when it makes the noise and stalling :confused:

[CTScott]

Those resistance values are for an unplugged sensor to make sure it's within static value. Those are not for when the car is running plugged in. So in order to read rich or lean you need to measure the output voltage of the sensor.

The ECM maintains a constant voltage on the sensor and it measures the current required to maintain that constant voltage to determine AFR. So, if you measure the voltage on the sensor it will always read the same unless there is something wrong with the circuit within the ECM.

[Bluevitz-rs]

I guess I'm not up to date with how the Yaris ECU works, just the Echo.

[mokmok]

If I unplug the AFR meter on the exhaust manifold I'm assuming that the ECU will switch to fault mode and run from a baseline set by Toyota ? If so, if I unplug it and the car runs fine it just means that I have to change the sensor?

[CoryM]

Hard to tell much from the video. Try hitting the exhaust with your hand or a soft mallet. See if you get any rattles. Also while cold, take your finger and pull on the manifold heat shield etc and try to "twang" it like a guitar string to get a noise. Often the best way to find loose heat shields. Just watch for sharp edges.

Other things that come to mind:
-cracked exhaust
-plugged cat or broken baffle in muffler causing intermittent plugging (but would expect to notice this driving, not at idle)

Any other engine problems with the car? Is it stock?

[mokmok]

Hard to tell much from the video. Try hitting the exhaust with your hand or a soft mallet. See if you get any rattles. Also while cold, take your finger and pull on the manifold heat shield etc and try to "twang" it like a guitar string to get a noise. Often the best way to find loose heat shields. Just watch for sharp edges.

Other things that come to mind:
-cracked exhaust
-plugged cat or broken baffle in muffler causing intermittent plugging (but would expect to notice this driving, not at idle)

Any other engine problems with the car? Is it stock?

Yes, it is stock, the only difference is a straight pipe axle-back , I will try to check for a heat shield when i get home tonight after work, as for cracked exhaust my exhaust headers are RUSTED worse than i ever could've imagined so it might be leaking

[Bluevitz-rs]

Ok, I have another idea... I too ran a straight pipe axle back for a while and tho it was cool and loud, It also set missfire code because the joint was loose, at idle. Make sure all the joints in the exhaust system are tight. It was doing the exact same thing at idle, minus the noise. The revs were dropping because it was pulling timing back.

[CoryM]

Ok. Leaking exhaust before, or sometimes up to a foot behind the pre-cat O2 sensor can suck in oxygen and confuse the computer. Also rattles can be picked up by the knock sensor, which will then cause the computer to pull timing out. Either could cause the poor idle issue.

[mokmok]

Ok, I have another idea... I too ran a straight pipe axle back for a while and tho it was cool and loud, It also set missfire code because the joint was loose, at idle. Make sure all the joints in the exhaust system are tight. It was doing the exact same thing at idle, minus the noise. The revs were dropping because it was pulling timing back.

The joints are tight and the exhaust heat shild is well held in place, but i'm starting to think that my rusted out headers are somehow leaking but why wouldnt it leak when cold... :confused: and the timing is pulled back to 2.5 degrees instead of 5 like it normally does

[Bluevitz-rs]

try re-installing the stock muffler and see if that fixes the problem. Maybe the knock sensor is overly sensitive.

[mokmok]

try re-installing the stock muffler and see if that fixes the problem. Maybe the knock sensor is overly sensitive.

Already tried that, no change :-(

[mokmok]

Is there any way to bypass the knock sensors for testing purpose?

[CTScott]

Is there any way to bypass the knock sensors for testing purpose?

Try connecting a 120K to 280K resistor to the knock sensor connector, in place of the sensor.

[mokmok]

Ok I tried that, still no change at all:frown: I also noticee when I am on the highway if i press the clutch the rpm drop to 500 then go up to 1500 then 1000 then sudden drop back to 500 :s