As I'm sure many people do with ECUFlash, I've been logging and tinkering with fuel/ignition maps lately. I was doing some logging on the way to work this morning, and when I got to work and graphed the data I saw something odd...

This was my first 3rd gear pull from ~3k rpm to redline...



Looks OK to me... very little knock, smooth boost (I log boost with a JDM MAP sensor) and load curve. Here's my second 3rd gear pull that was done shortly after the first one...



There was obviously no changes in the tune between these two pulls. There's a little more knock in the second pull for some reason, but still under 3 counts. But what I'm wondering why I'm seeing these odd spikes in the 2-byte load value? Any thoughts? Thanks guys!

 

It's just a side effect of the logging speed and the fact that you are logging the two bytes of the load variable separately. The spike is because the MSB changes from one value to the next at the particular point and you logged the LSB after that change.

It can happen at increments of 80 load. So, if you look your spike is right at 240, which spikes to about 320.

Eric

 

I get those spikes and dips randomly with 2-byte load. I have seen no common RPM value or corresponding data that seems to trigger it. You'll notice that the spike/dip only occurs for a single sample. It's just a glitch and you don't need to worry about it.

 

I get those spikes also. No need to worry about it. I usually edit them out to get a better load value.

On a different note, you are knocking in the second log because your car got hotter. I would suggest that you log your car after driving for an extended period of time to make sure that it does not knock under normal temperatures.

 

Thanks guys, just the info I was looking for.

So, not trying to be one of those newbs who say 'look at my tune', but...





My mods are sorta custom tune, KN drop-in, and 3" catless TBE. This is a 2006 Evo IX. Here's what a 'virtual dyno' graph of the second run looks like...



Anybody who knows more than I do... any areas look too rich or too lean?

Thanks!

 

Anytime, you see a spike, the actual load is 240. If you want, you can edit it in the Excel file.

 

^Enlightening.

I always theorized that this occured as the datalogger logged one of the bytes twice in the same sample, then compiled the 3 values to the max value.

But handy to know.