Knock sensor affected by piston slap?

[pacey]

I have seen a couple of references here on the forum that piston slap might be picked up by the knock sensor and interpreted as preignition knock and cause the ECU to retard the ignition timing. Can anyone shed any more light on this?

 

 

The reason that I am wondering is that I have a 2002 Legacy GT with 40,000 miles that exhibits a cold startup noise (not yet confirmed as piston slap) and also idle hesitation. We have scheduled the car at the dealership next week at a time that a SOA tech rep can observe the car during a cold startup. The service manager at the dealership has told me that piston slap will likely not be considered a warranty issue as it is considered nondestructive if it goes away when the engine warms up. He did say that they might be able to reflash my ECU to get rid of the idle hesitation. I’m wondering if anyone knows if this reflash makes the ECU less sensitive to the knock sensor so that it doesn’t respond to piston slap (and also possibly less sensitive to preignition knock as well). If this is the case, I’m reluctant to let them reflash the ECU, which seems like a Band-Aid to cover the more serious problem of piston slap. Comments?

Pacey

[forester2002s]

He did say that they might be able to reflash my ECU to get rid of the idle hesitation.

 

I would be interested to know if this works to cure the idle hesitation. And what exactly will the dealer do to 'reflash the ECU'? Will they just disconnect the power to the ECU for a while, or will they actually 'reprogram' the ECU with new software?

 

Let us know what transpires.

[sregor13]

Whats your elevation?

[mrk610]

I have an 02 outback had the #2/4 pistons replaced for slap . My noise was very bad at all temps . I still have slap but not as bad and no noise when warm. I also had the ecu reflashed for hesitation about 2 months ago. The hesitation is gone but now my car pings under load . Taking it in for some of the 60k service and going to have my TCU replaced for harsh tanny shift and harsh power transfer to rear wheels .I'm going to see if I can have them look into the pinging problem . I have had all work other then regular service done under warranty .

[pacey]

The pinging under load after the ECU reflash reported by mrk610 makes me wonder even more if the reflash reduces the sensitivity to engine knock. The onset of pinging after the reflash makes this seem plausible.

 

In response to sregor13, our elevation is 1100 feet.

 

Since first posting, I have installed new spark plugs, spark plug wires, PCV valve, and O2 sensor and the hesitation seems to be much less, but I have only driven it a very limited amount. I did these changes in steps, and the biggest improvement seemed to come after installing the new O2 sensor.

 

Does anyone else have any experiences, good or bad, related to getting the ECU reflashed?

 

Pacey

[rysbrnr]

In regards to piston slap, does the 2.5L have cast or forged pistons? Piston slap is is kind of a trait that comes with having forged pistons. You don't usually hear of imports with stock cast pistons slapping.

[friendly_jacek]

In regards to piston slap, does the 2.5L have cast or forged pistons? Piston slap is is kind of a trait that comes with having forged pistons. You don't usually hear of imports with stock cast pistons slapping.

 

I thought piston slap was caused by short skirts and clearance in the cylinder.

 

Pacey, I have no short answers here, but if one uses OBD2 interface, wouldn't the excessive retard show up? Subaru's ECU reports degrees of ignition timing. Unfortunately, don't know the spec/expected number.

[nipper]

The knock sensor and the computer behind it are fairly inpressive peices of equipment.

The knock sensor isnt listening for any old thing. Its purpose os to react before ping is even audible. The sensor, which is a tranducer, is looking for a specific accoustic profile. Piston slap, bearing knock, slap, ro any of the other many noises that can go wrong or are supposed to be in an engine do not have an effect on the transducer. It also a limited capability on how much it can deal with. If the pinging is audible, its most likely out of the range that the computer can adjust for without making it a huge driveability issue. Thats when you should get (if the computer is set up for it) a check engine light.

 

nipper

[friendly_jacek]

The knock sensor and the computer behind it are fairly inpressive peices of equipment.

The knock sensor isnt listening for any old thing. Its purpose os to react before ping is even audible. The sensor, which is a tranducer, is looking for a specific accoustic profile. Piston slap, bearing knock, slap, ro any of the other many noises that can go wrong or are supposed to be in an engine do not have an effect on the transducer. It also a limited capability on how much it can deal with. If the pinging is audible, its most likely out of the range that the computer can adjust for without making it a huge driveability issue. Thats when you should get (if the computer is set up for it) a check engine light.

 

nipper

 

If so, why people in late 90's subarus were getting the infamous acceleration hesitation fixed by knock sensor mods?

[nipper]

If so, why people in late 90's subarus were getting the infamous acceleration hesitation fixed by knock sensor mods?

 

Could have been a poorly designed knock sensor, just like poorly desgined head gaskets. Just because one mfg screwed it up, it doesnt mean thats not how its work.

The knock sensors were mis-calibrated and hearing the wrong noises. What I described is the automotive enginnering explination on how they work. Subaru just had improperly calibrated sensors (or a bad computer algorythem). It happens.

I have gotten them where they were designed for the wrong frequency (for a different application with explosives but the same type of sensor). Just when i got it i sent it back to the vendor. When Subaru got they got 1000's of them.

No one argues what a head gasket does but they seemed to have had 1000s of bad ones of those too .

 

nipper

[Setright]

One large batch of knock sensors had a tendency to crack their casings. Not good for accuracy!

[blitz]

One large batch of knock sensors had a tendency to crack their casings. Not good for accuracy!

I gotta argee with Setright. The knock sensor is pretty crude (it's essentially a piezo accoustic-guitar transducer stuck into the engine block). It works OK for most apps, but needs two (one in each cylinder) on a boxer rather than a single, solitary unit mounted on the crankcase - while the cylinders extend WAY out into the ether. The way Subaru does it, it picks up a lot of extraneous noises to the exclusion of spark knock (poor S/N ratio).

 

Also, the current implimentation has a difficult time distinguishing piston slap from spark knock (they both clang the bore fundamental frequency).

 

The idea of knock-sensing isn't bad, but the current implimentation is.

[syphon]

Does anyone have a link with all the symptoms and info about the infamous "Piston Slap"? It's really frustrating to always see threads with people talking about it, but assuming that everyone else knows about it

 

I own a '98 Legacy GT and am curious if my car is afflicted with this.

[nipper]

i found this

 

http://resnet.uoregon.edu/~dwhite/pages/subnoises.html

 

lets see it says a possible 10% reduction in engine life. Considering almost every subaru has it to some degree, unless its really noisy in the morning i wouldnt worry about it. My 98 OBW does it when its cold for about 2 seconds then it goes away and it has 180K on the clock.Well from what ive seen on the boards, if you dont blow a headgasket, seems like the boxer runs for ever, so whats 10% of forever.

[friendly_jacek]

I have seen a couple of references here on the forum that piston slap might be picked up by the knock sensor and interpreted as preignition knock and cause the ECU to retard the ignition timing. Can anyone shed any more light on this?

 

Pacey, have you figured out the answer?

I measured timing via OBD2 in mine and the advance is 15 deg on idle and much higher under load (seen higher than 30 deg at high speed).

If your numbers are much diffrent, you could have a problem.

[nipper]

I actually found a service bullitin put out by subaru (i'll hunt it down) that said if the customer complains about it, and a whole bunch of other stuff checks out, to replace the ECU with a ECU that has differnt algorarythem to fix the problem.

 

I'll try to find it later today.

 

nipper

[nipper]

I have seen a couple of references here on the forum that piston slap might be picked up by the knock sensor and interpreted as preignition knock and cause the ECU to retard the ignition timing. Can anyone shed any more light on this?

 

 

 

The reason that I am wondering is that I have a 2002 Legacy GT with 40,000 miles that exhibits a cold startup noise (not yet confirmed as piston slap) and also idle hesitation. We have scheduled the car at the dealership next week at a time that a SOA tech rep can observe the car during a cold startup. The service manager at the dealership has told me that piston slap will likely not be considered a warranty issue as it is considered nondestructive if it goes away when the engine warms up. He did say that they might be able to reflash my ECU to get rid of the idle hesitation. I’m wondering if anyone knows if this reflash makes the ECU less sensitive to the knock sensor so that it doesn’t respond to piston slap (and also possibly less sensitive to preignition knock as well). If this is the case, I’m reluctant to let them reflash the ECU, which seems like a Band-Aid to cover the more serious problem of piston slap. Comments?

 

Pacey

 

http://www.ultimatesubaru.org/forum/showthread.php?t=43489

 

Go to that link, there are a couple of recalls on the ECU for air/fuel mix. Look up TSB's.

There is a TSB on knock sensors (same site) but its for an earlier year then yours.

 

nipper