One of the big disadvantages of the the knock sensors we use is that all they can do is pick up sounds. There is no way for the sesnor itself to differentiate between acual pre ignition / detonation or say the increased engine nose from a modified vehicle. The manufactuerer tunes the ecu to react only to cerain frequencies which the engineers have found correspond to actual detonation. Often modifications to the car can effect the harmonics of the engine and resul in normal sounds which the ecu interprets as knock.

Often it is difficult for the tuner to dinstinquish between the two without removing the plugs and reading them and or by listening to the block with a chasis ear.

Its important to check for any metalic objects contacting such as down pipes on the sub frame - fmic pipes against the air intake pipe, etc etc.

Also, a great way to test fo knock is to pur in some race gas. If the sound is not effected then its usually a mechanical sound. If pouring in race gas eliminates the sound then its mos likely actual detonation.

In the future these primative devices will be obsolite as the new cutting edge method to sense knock is by throwing an arc across the spark plug to measure the conditions in the combustion chamber after the combustion. This is a much more accurate method to asertain if the clyinder is knocking and has the added benefit of narrowing down the detonation to a specific clyinder so that the timing or enrichment adjustments can pin point the individual cylinder which is knocking.

Of course these new methods mean nothing for us.

One solution to the problem is to isolate the knock sensor from the block to reduce its sensitivity - I have seen people fabricate spacers and buffers to desensitize the knock sensor. I want to mention that I would NEVER suggest this to a customer of mine and I would never do it on a customer's car. Its just something I have seen done.

The more heavily modified your car becomes the more of a trade off the stock ecu becomes. At some point a stand alone ecu becomes more and more useful.

Al

 

Why is that >?

 

Great additude and a realistic approach.

All too often this is the reality of the situation when a car comming to be tuned, in fact does not need JUST a tune but rather may have outstanding mechanical issues which need to be diagosed and corrected before a proper tune can be completed.

Al

 

Ok sorry I see your explanation.

Its odd I have never seen any such problem.

Al

 

Of course this is the main issue with the stock ecu.

With the ecu unable to distinquish between one sound or another there is no way to correct for the false knock without leaving the engine exposed to actual detonation. Since the customer needs one map to drive all the time, the knock protection must remain active for protection.

Those engines which create harmonics that trigger knock counts will frequently have erratric timing when the ecu is picking up the false knock.

In those examples that have this condition - as in the case of the OP - the sound comes and goes from one pull to another often without any ryhme or reason.

At the end of the day, while the stock ecu may prove dicey to tune with some individual modified cars that have harmonic sound issues, in most cases the stock ecu works very well and is also very good at protecting the engine from actual detonation.

These kind of false knock situations are not that common (thankfully). Although that is no comfort to the OP.

Al

 

Ok, just went to the track on this setup to see what it can do before I get the AEM EMS. I went to Brainerd International Raceway.

60ft: 1.948
330: 5.310
1/8: 8.117
MPH: 90.32
1000: 10.522
1/4: 12.554
MPH: 109.28

I'm not a good driver. It's still got some juice left in it and probably could have gotten low 12's if I improve on my 60ft times. I am running the ACT 6-puck sprung and am a disgrace to the HTA. Hell, my stock turbo ran a 12.7 @ 109.

 

Yeah, your car has SERIOUS problems. That is terrible times. I would not even drive the car as it seems you are having huge problems with knock/power loss.

 

Thanks for keeping us posted and good luck with the AEM.

 

It actually felt like it pulled pretty good. It's just not fast. I didn't do any logs which I should have done.

 

With all that said just take it to another tuner to look over the map and see if there is something that can do about it.

 

I think you will love the AEM, I know I love mine. Just make sure your tuner takes the time to dial in the knock control (voltage vs rpm table) and the knock retard setting. By default the table is not where it needs to be.

They can do the tune with the knock control off, but make sure you get an internal log (uses the internal buffer on the aem to log, very very fast in comparison to laptop logging) of the final pull. You can then use the log to build a profile for the voltage vs rpm table. Then back out a half degree or so from the tune, turn on knock control, and enjoy your new ride!

Shoot me a pm if you need any help with any of that

 

Well, actually I think a better driver for sure would have gotten higher traps and into the 11's. A Subie was as fast as me on the track but not on the open road if you obviously know what I mean.

 

While i hope you get the issues figured out with your setup, i can tell you that your car is unable to run faster or even come near 11's with how your car is tuned now. I ran 12.4's @109 on the stock snail with minimal upgrades at the same track with horrid 60ft times. I'm not trying to bag on you, as i hope you get everything worked out and start running 10's on that bad boy, but you obviously have some issues that need addressing. This is probably a stupid question but i didn't notice any kind of intercooler upgrade in your sig, and if your running a stocker(i'm sure your not) the ait's would be scary high.

 

wow 37r = alot of lag + only 350hp???

get beat by civics much?

 

Yeah I forgot to add it in my sig. I have a 3" ETS intercooler. It was originally for running the FPGreen. I might need a bigger core.