Phantom Knock on #2 cylinder?

FlashPro questions & answers specific to the 2006-2009 (US, UK, Asia) S2000
zanter
Posts: 4
Joined: Tue Jun 16, 2015 7:32 pm

Re: Phantom Knock on #2 cylinder?

Post by zanter »

Thanks for your guidance. I changed the o2 sensor before so I didnt expect it to fail so quickly. I was almost going for bigger injectors to solve my fueling issues and I can't figure out why my AFR is stuck at 14.5 on WOT( despite the high lambda WOT afr settings) but the partial throttle fueling is ok. Anyway just ordered a denso replacement lambda sensor from amazon and hopefully it resolves.

Perhaps the "knocks" come from resonance from the J's racing headers I am using but its mainly concentrated on cylinder 2. I have been reducing the timing on the knock areas resulting in power dips but doesnt help much, so its annoying. In the past I used to have knocks which I could hear as pings lower down the rev range on high load and can tune it out but I can't hear any obvious knocks at 6k over the engine revs. I will try reducing the high knock sensitivity even more by increasing the values. I reckon the knocks will cause the ECU to pull timing out so my last resort is to turn off the knock sensor :)
EFICU
Posts: 3295
Joined: Mon Feb 25, 2019 10:14 pm

Re: Phantom Knock on #2 cylinder?

Post by EFICU »

What brand is the sensor in there now? Do you have any datalogs from before you started adding a lot of fuel, or maybe three or four back while you were adding fuel? If you're adding fuel and the AFR isn't changing, watch the injector duty cycle to see if that is going up. If the duty cycle is rising, but the AFR is staying the same, then something is wrong, most likely with the o2 sensor.

When raising the knock sensitivity tables, only raise the cells that are showing false knock. Don't go in and make global changes in that table which raises all the cells. Look through the datalogs and make note of the cells that are showing false knock, then raise those cells only. That way you're only adjusting cells that are causing you issues in the hopes that in the future any real knock the ECU will record and pull timing if needed.

If you want, I can input the knock sensitivity tables from our final files with Futureal33 and see if they help you remove your false knock. Otherwise, I would really study the cells that are showing false knock and raise those in small increments to keep retesting.

Feel free to make you own thread too, just in case the other people in this thread are getting notifications they don't want from our posts.
Post Reply