Forum » Practical Reflash Tuning » Mazdaspeed3 knock sensor calibration

Mazdaspeed3 knock sensor calibration

Practical Reflash Tuning

Discussion and questions related to the course Reflashing


Page 1
Author
40 Views

If anyone is familiar with tuning the Mazda MZR DISI engine platform via Cobb Accesstuner, I'm hoping you can help me out. I've done some heavy modifications to the engine e.g. forged short block (Manley), balance shaft delete, ARP studs, solid motor mounts, and full bolt ons. And as of recently I switched to an E70 mix. However, even after switching to the new fuel I'm still seeing excessive knock retard, mainly in the mid rpm range, but is still very unlikely to occur when using ethanol. I suspect phantom knock is occurring due to the drastic increase in overall engine noise. I'm wondering if there is a way to calibrate the knock sensor sensitivity with the Accesstuner software? I did notice there are knock offset and multiplier tables but nothing that seems to directly relate to the knock sensitivity. Any help would be greatly appreciated

Can you post screenshots of all the knock tables you have available in Access Tuner?

The first image is the knock offset and the second is the knock multiplier

Attached Files

I wouldn't be surprised if those are not very precise descriptions of what the tables actually do. My advice is this. As a learning exercise (NOT a long term setting) See if you can figure out whether you can deactivate or desensitize the knock sensor by setting those tables to max or 0.

So, take both tables and set to 0. Does the knock sensor behavior change in the questionable areas? If that doesn't work, set the tables to 99999 or whatever. If that actually deactivates the knock sensor (or eliminates the retard), you can then narrow down what you need to do. So for example, if setting both tables to 0 deactivates the knock sensor, try picking a more narrowly defined speed and load area for both tables. Set that area to 0, OR alternately just cut the values in half for both tables.

USE COMMON SENSE when testing this. If you get crazy knock retard (it may clip to a minimum spark retard value) you could overheat components, and likewise you could suddenly knock without protection. So be careful how you drive.

You can also experiment to see which is more useful, the multiplier or the offset table. For simplicity's sake I would adjust both tables first. It's also entirely possible that both tables don't actually do anything that would be useful to you. In that case you can use the RPM and load lower limits, which is a less precise way to limit knock sensor operation. For example, you would set the minimum rpm to 3000 if you're not worried about actual engine damaging knock below 3000rpm.