Jump to content
SAU Community

Recommended Posts

Hi all,

I have been tuning my 20det and have into a bit of confusion. When I log the knock sensor values, sensor 2 is has quite different values than sensor one, even at areas where there is little knock possibility. Is this because they could be wired in series? Although wouldn’t make much sense, I haven’t looked into it much. I will get large knock spikes at about 5500rpm -6000rpm, we have taken a lot of timing out down to about 11-13* and ran different types of fuel and have had no luck. I would think that the voltages of the two sensors would be more similar but I could be wrong. I’m not sure if this is false knock or not but I would like to bump the timing back up to 16* ish as the afr’s were spot on and it felt good or if anyone has any other suggestions that would be much appreciated. Unfortunately don’t have access to knock ears. Cheers

 

Set up is a rb20det

gt2860rs

440cc injectors

platinum pro

image.jpg

image.jpg

Edited by smart_garrett
Link to comment
https://www.sau.com.au/forums/topic/481270-strange-knock-sensor-values/
Share on other sites

Stock knock sensors are not really worth a knob of goatshit. They will miss real knock and detect non-knock. If there's a little more mechanical noise at one end of the engine than the other, ie valve train clatter or rod noise, then one will/may register knock that the other one doesn't.

You CANNOT tune based on what the knock sensors are telling you. You can only tune using knock ears or similar.

2 hours ago, Duncan said:

I'd start by swapping the front and rear sensors to see if the problem moves with the sensor (so replace it), or if it is the same (ie, there is a wiring or actual problem)

Good idea mate, will give it a go

2 minutes ago, GTSBoy said:

Stock knock sensors are not really worth a knob of goatshit. They will miss real knock and detect non-knock. If there's a little more mechanical noise at one end of the engine than the other, ie valve train clatter or rod noise, then one will/may register knock that the other one doesn't.

You CANNOT tune based on what the knock sensors are telling you. You can only tune using knock ears or similar.

Yep, makes sense could be picking up drivetrain “noise” as well. I had a great time helping tune our e46 race car here in the us for a series called lucky dog, had knock ears there and was able to hear it audibly, really helped make it perfect. Just my first dab dealing with rb’s, good to know they’re not reputable

Update, I have played with timing and fuel bit more and retarded the timing 5 degrees and it still made little to no difference. It seems to detect the least knock competitively (if it even is real knock) at about 16* timing at high rpm/wot, which is conservative from what I've seen people online have their 20's at. I also put a tank of fresh 100 AKI (104 RON) and it seemed to make little to no difference of the seemingly high knock values that are spiking to 200+. This is leading me more to think one or more of the sensors is faulty.

I'm surprised no one has mentioned this yet. The platinum pro has terrible 'knock detection' it's closer to 'random noise detection'. I'm not saying the ecu is rubbish, it just that it can't do proper knock detection.

The platinum pro is a very old ecu by todays standards. If you want better knock detection, get a haltech elite ECU or better yet, by a plex knock monitor. 

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

    • I have no hard data to report, but I have to say, having driven it to work and back all week, mostly on wet roads (and therefore mostly not able to contemplate anything too outrageous anywhere)..... it is real good. I turned the boost controller on, with duty cycle set to 10% (which may not be enough to actually increase the boost), and the start boost set to 15 psi. That should keep the gate unpressurised until at least 15 psi. And rolling at 80 in 5th, which is <2k rpm, going to WOT sees the MAP go +ve even before it crosses 2k and it has >5 psi by the time it hits 90 km/h. That's still <<2.5k rpm, so I think it's actually doing really well. Because of all the not-quite-ideal things that have been in place since the turbo first went on, it felt laggy. It's actually not. The response appears to be as good as you could hope for with a highflow.
    • Or just put in a 1JZ, and sell me the NEO head 😎
    • Oh, it's been done. You just run a wire out there and back. But they have been known to do coolant temp sensors, MAP sensors, etc. They're not silly (at Regency Park) and know what's what with all the different cars.
    • Please ignore I found the right way of installing it thanks
    • There are advantages, and disadvantages to remapping the factory.   The factory runs billions of different maps, to account for sooooo many variables, especially when you bring in things like constantly variable cams etc. By remapping all those maps appropriately, you can get the car to drive so damn nicely, and very much so like it does from the factory. This means it can utilise a LOT of weird things in the maps, to alter how it drives in situations like cruise on a freeway, and how that will get your fuel economy right down.   I haven't seen an aftermarket ECU that truly has THAT MANY adjustable parameters. EG, the VAG ECUs are somewhere around 2,000 different tables for it to work out what to do at any one point in time. So for a vehicle being daily driven etc, I see this as a great advantage, but it does mean spending a bit more time, and with a tuner who really knows that ECU.   On the flip side, an aftermarket ECU, in something like a weekender, or a proper race car, torque based tuning IMO doesn't make that much sense. In those scenarios you're not out there hunting down stuff like "the best way to minimise fuel usage at minor power so that we can go from 8L/100km to 7.3L/100km. You're more worried about it being ready to make as much freaking power as possible when you step back on the loud pedal as you come out of turn 2, not waiting the extra 100ms for all the cams to adjust etc. So in this scenario, realistically you tune the motor to make power, based on the load. People will then play with things like throttle response, and drive by wire mapping to get it more "driveable".   Funnily enough, I was watching something Finnegans Garage, and he has a huge blown Hemi in a 9 second 1955 Chev that is road registered. To make it more driveable on the road recently, they started testing blocking up the intake with kids footballs, to effectively reduce air flow when they're on the road, and make the throttle less touchy and more driveable. Plus some other weird shit the yankee aftermarket ECUs do. Made me think of Kinks R34...
×
×
  • Create New...