Timing dip with varying high load

[Puffwagon]

I have my vehicle running the custom os, extended tables, 4 bar boost sensor etc.

My timing tables axis max load is 2.6 whereas I will hit 6.0 load in the log. This works fine for the most part however I've noticed that if my boost control isn't spot on it will go into other load ranges, 5.7 etc etc.

This is all very normal except I've noticed that my timing will have a dip or vary if I don't have a near perfect boost curve.

There must be some sort of extrapolation going on, the only other thing I can see causing the variation/dip in timing is the wg pressure trending downward.

 

[Puffwagon]

Here is the example;

The only difference in the tune from these log pics is me smoothing out the boost curve via the wgdc table and a small sd table change to fix the lean spot just after 161 seconds.

The timing maps are unchanged between the logs, in the first pic the line is still 19 degrees and the dip is 14 degrees.

I didn't rev it out as much after I fixed it but you can see that the timing hadn't dropped off while at WOT. I lined up the blue line for reference.

 

 

 

[Roland@pcmtec]

You are logging the wrong DMR. This was in our default layout files about a year ago in abets build at one stage by mistake. It has since been removed but if you load old config files with it in there it will still log the cell index dmr. 

The value you are seeing is the cell index of the load axis, not the actual load value. If you remove it from your layout and look in the tree for load the will show you the correct value. 

We updated the description so it should be fairly clear now that the wrong DMR is not load but in fact cell index. We've also hidden it from search so people don't select it by accident as there is no real good reason to log it. 

Once you add the curve load dmr map trace it against your spark map and you'll see which cells are getting hit. It will also draw a torque curve across your timing map making it much easier to tune. 

[Puffwagon]

Sorry I should have clarified that too, I am logging the actual load, it just isn't in the chart.

I'll trace it and see what happens.

[dat111]

@Roland@pcmtec are you able to log injector duty cycle with pro or is it workshop only? ive looked though it all an cannot see injector duty cycle 

[Puffwagon]

This is with load added into the charts, same pics

 

Edited November 7, 2022 by Puffwagon

[Puffwagon]

I couldn't trace the good chart, the log was too long. I traced the timing dip and it looked as expected, flat across the bottom of the timing map. While I am curious as to why the timing is moving around when it is showing it referencing the same timing cells, I figured I can make it stay flat and do what I want by adjusting the axis like this.

 

[Roland@pcmtec]

8 hours ago, Puffwagon said:

I couldn't trace the good chart, the log was too long. I traced the timing dip and it looked as expected, flat across the bottom of the timing map. While I am curious as to why the timing is moving around when it is showing it referencing the same timing cells, I figured I can make it stay flat and do what I want by adjusting the axis like this.

Use the control key and you can select a subset of the chart. (shows in blue) then you can show the line plot/map trace over that section only, eg the dyno run only. Have a look in the datalogger video where they do the blue drag select (press ctrl once to initiate the cursor, then again to end the selection).

Your datalog is showing a very low resolution for load. I've highlighted the actual log values you are getting. Right click and make sure load is set to high priority, I'd also consider removing anything else from the log you don't need so you can have a very high resolution log 

 

9 hours ago, dat111 said:

@Roland@pcmtec are you able to log injector duty cycle with pro or is it workshop only? ive looked though it all an cannot see injector duty cycle 

It isn't in pro at the moment for some reason but it will be added shortly!

[Puffwagon]

Cheers, that's handy. I had a look at the picture and there are frig all samples in the histogram. I'll go ahead and remove the unnecessary items and make load high priority.

So do you think the reason it is showing varying numbers is because of the logging rate? Perhaps it is hitting a different number occasionally and because the logging rate is low, the average is highly variable which shows up as a variance in the chart?

That might explain why I get some weird numbers in Virtual Dyno lately.

[Puffwagon]

I've got the bare minimum on there now to safely adjust the fuel and boost and it's gone from 18 samples per second for the final spark number to about 55 samples per second per channel. I'll take it for a drive tomorrow and see how it goes.