Jump to content
SAU Community

Recommended Posts

Hi peoples,

I got a bit of a curly one for you all!

I just had my car (r33 S2) tuned with a safc2, and althoug i am very happy with the peak output the car achieved, my tuner told me the mid range is being held back a bit becuase of the knock sensors detecting some knock. Obviously the result is timming being pulled out and petrol put in. At about 5500rpm it comes good again and the power curve heads upwards dramatically.

My problem with this is that the tuner told me that it is common for knock sensors to be overly sensitive from factory, as he was unable to detect any knock using his own form of knock detection. I have read threads on how to block out the knock sensors by replacing them with a resitor, or by installing diodes between the ecu and the knock sensors, but am unsure of wether this is a wise move.

Has anyone got any experience with this problem, as i would like to improve my mid range tune without changing computers or anything like that. Any help would be great...

Link to comment
https://www.sau.com.au/forums/topic/306075-knock-sensor-question/
Share on other sites

Sounds to me like a poor excuse from a bad tune.

What it would sound like to me on face value is you are still hitting Rich/Retard on the factory ECU even with the S-AFC.

Now, correctly tuned/setup the S-AFC should bend the signal enough to avoid any factory ECU retarding elements.

The factory knock sensors are not over sensitive, you should not be adding restrictors and so on.

If you think he is right, get another one and replace your factory one. I wouldn't be surprised if the exact same thing happened - because of what i said about about poor S-AFC setup.

The factory knock sensors pick up other things besides knock and it registers it as knock, this is why tuners use there own knock sensing equipment, the proper way to fix the issue is a full ecu like a power fc then you wont have these issues.

^ umm the PFC will still pickup the exact same knock level - if the sensor is playing up.

They can pickup a starter motor being loose and banging around for instance.

However putting in a PFC won't magically reduce the sensors sensitivity, or what it reports, so you would still see "danger" on the PFC... IF the knock sensor is acting up.

In which case you would never know if the car is in danger or not...

I still go with what i said initially though, swap out the sensor, if the same thing happens it's the tuner or something unlucky like a loose starter motor.

^ umm the PFC will still pickup the exact same knock level - if the sensor is playing up.

They can pickup a starter motor being loose and banging around for instance.

However putting in a PFC won't magically reduce the sensors sensitivity, or what it reports, so you would still see "danger" on the PFC... IF the knock sensor is acting up.

In which case you would never know if the car is in danger or not...

I still go with what i said initially though, swap out the sensor, if the same thing happens it's the tuner or something unlucky like a loose starter motor.

Very true, i have seen lots of reports of people getting high knock figures but when the tuner is using the after market knock detection there is no or very little knock, replacing the knock sensor( of which i have 2 from a spare engine i bought if you want to try it) may cancel out some of what the factory ecu gets but they are well known for registering a lot more knock than is actually there which will always result in a lower quality tune when only using a piggy back.

Sounds to me like a poor excuse from a bad tune.

What it would sound like to me on face value is you are still hitting Rich/Retard on the factory ECU even with the S-AFC.

Now, correctly tuned/setup the S-AFC should bend the signal enough to avoid any factory ECU retarding elements.

The tuner i used is one of the most recomended in my area. His comment on the overley sensitive sensors was a generalisation about the sensitivity of factory sensors. Basically he meant they lean on the side of caution a little too much.

It's sounds like "word3d" has also heard of this, going by what he said about tuners not detecting knock even though the sensors are dectecting something.

Also safc only bends the singnal of the voltage from the air flow meter, so how could that influence R&R kicking in for knock? i know you can stop

R&R kicking in for exceeding the factory set airflow reading, but not for knock!

I never said the 'apparent' knock was related - just that it sounded like an excuse for R&R kicking in as a result of a job not being done 100% right.

As i said, swap out the sensor, you'll soon find out where the issue lies.

I have had a few GTR's and S14's that show high knock figures for no reason, one in particular was a SR20 S14, on the Kmon is was silent, ran a very safe map in it, yet the PFC D-jetro would show up to 50-60 on decel and even at idle would hover around 15-20, at wot it would often hit 95, I went over every squar millimeter of that engine bay, not a single thing loose or contributing. 6 months later I did an engine build for more power, same thing with knock, did not change.

Eventualy I replaced the knock sensor at a regular service and it dropped to about 3-4 at idle and 15 odd on decel with 50-60's at WOT (usualy see 25's to 30's on similar setups) I then replaced the complete knock harness from sensor to ECU, didnt make a difference. I even tried to switch out the PFC's with its sister car running identical setup and still the car im mention would always read close to double,

What I did notice though, if you run a hard core twin plate or tripple plate clutch it can be a very good contributer, SR20's knock sensors usualy go balistic right before/as they chow the teeth off 3rd gear as well. Not saying somthing is wrong with your driveline, just that it could be somthing a little futher down the line thats causing your odd readings.

Good luck, hope you find the problem.

Eventualy I replaced the knock sensor at a regular service

Was this a genuine article or aftermarket. I am presuming that given your circumstances they would be new, not 2nd hand.

Very expensive?

Where from?

Thanks for your replies guys, it is a bit of a grey area i know. Im sure someone else is learning from all this!

people forget that with the SAFC there are always 2 actions for every adjustment.

SO if you get the mixture nice (usually by leaning it out from the factory rich settings) the second action is to the timing table.

Reduce fuel = increse timing

Increase fuel = decrease timing

If you take out too much fuel you generally add too much timing especially around peak torque in the midrange.

Finally if for some reason the car switches over to the low octane map the tuning is done based on the wrong map. A good practice is to run a very aggressive ramp rate once finished then turn the car off for 30seconds and redo the run just to make sure both runs are identical. If it was in the low octane map for the first run the car will most likely knock its head off and be lean on the second run.

I hate SAFC's on anything mildy smart (r34's etc) as it is impossible to get what i consider good results... especially as i monitor the base tables with nistune and see just how average the base figures are to work with.

  • 3 weeks later...
Was this a genuine article or aftermarket. I am presuming that given your circumstances they would be new, not 2nd hand.

Very expensive?

Where from?

Thanks for your replies guys, it is a bit of a grey area i know. Im sure someone else is learning from all this!

Owned , I just had my knock sensors replaced , problem with sensors was found with a Apexi Fc hand controller , cost is $ 350 per sensor from Nissan not including install - big $$$$$$$, must remove exhaust manifold! The other option is to overide the fault with ECU upgrade .

Edited by NISSAN GTR
Owned , I just had my knock sensors replaced , problem with sensors was found with a Apexi Fc hand controller , cost is $ 350 per sensor from Nissan not including install - big $$$$$$$, The other option is to overide the fault with ECU upgrade .
how silly of me . it is the wrong side Edited by NISSAN GTR

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now


  • Similar Content

  • Latest Posts

    • Block bump. $400. As above cyl 6 needs bore or sleeve.
    • I would think making the argument that the travel is limited by a spring flexing against a spring perch as "the same method". Later on in the document they do state that the spring can't bind on full bump travel and cannot come loose in full rebound travel as well. (which is all very sensible). The laws are actually pretty sensible and reasonable. It's just that the people who enforce and check them don't actually read them or know them accurately. "Oh, coilovers? Instant fail mate. Don't even need to look at it." - Guy who will be instantly reported by me. There is probably merit to people who do get defected for height also get defected for the suspension in that state that allows it. I did never consider the people who are complaining about coilovers being picked on are also running around at 50mm off the floor.
    • I think given SAU's knowledge of E85 we can strongly conclude that 10% ethanol in almost any situation is entirely fine. Almost all of the myths against E85 were overblown, let alone E10.
    • From your link See bold text, is this referring to damper settings, if so that may a issue for "some" inspectors, I cannot see aftermarket coilovers having the evidence that "must be available that its functional performance is equivalent to the original" Maybe just remove the adjustment knows and hope for the best???? Meh 5.2 Suspension travel In all instances, modifications to a vehicle’s suspension must ensure the integrity of the system and not compromise the ride quality. At least two thirds of the original suspension travel should be maintained in both directions (rebound (i.e. extension) and bump (i.e. compression)), and rebound must be limited by the same method used by the vehicle manufacturer or if this is not practicable due to the nature of the modification, an equivalent method. If an alternative method is used, evidence must be available that its functional performance is equivalent to the original.
    • They actually don't - They adhere to VSB14 rules just like Victoria. The rules are against CABIN adjustable height, and it quite clearly states that the height has to be within parameters. I asked the VASS engineer to confirm this when I got my car engineered and they refused to engineer the coilovers because they didn't meet the requirements for requiring engineering. (mine are height adjustable.) People "Not wanting to bother" with "Actually reading/knowing/adhering to the rules" should result in fines and immediately losing the ability to issue blue slips and/or RWC's in Vic.
×
×
  • Create New...