Hi,
Hoping you can help me to understand something I am seeing in my logs please.
I am seeing some spikes in the ignFinalPri1 values and I am not sure what is causing them. At first I though it might be the per cylinder compensations, but they seem to happen too infrequently and there are sections in the rev range these don't seem to occur. I have attached a screenshot of a 4th gear pull; ignBase1 in blue and you can see the downward spikes of the ignFinalPri1 values in green. There are no knock related or any other ign timing compensations I can see in my logs that could be causing this. Any opinions or suggestions welcome.
Cheers
Richard
ignFinalPri1
ignFinalPri1
- Attachments
-
- S6_4thGear.JPG (76.64 KiB) Viewed 9133 times
-
pat
- Syvecs Staff - Cleaner
- Posts: 356
- Joined: Fri May 23, 2008 10:23 am
- Location: Out there... somewhere
- Contact:
Re: ignFinalPri1
Richard,
The most likely cause is indeed per cylinder correction. You cannot see what the other cylinders' knock corrections are on that datalog, only the instantaneous knock energy readings, ergo you can't see what the per cylinder correction is doing. The reason that you are seeing the effect more or less with RPM is simple aliasing. In order to effectively see each cylinder's final value you'de need to be logging at least twice as fast as the power stroke frequency; at 3000 RPM there are 50 revolutions per second and therefore 100 power strokes per second, You'de need to sample at least at 200 Hz to guarantee to see each final value. Anything less will be subject to aliasing. Of course that would need to go up to at least 400 Hz to make it reliably record every event up to 6000 RPM....
Hope this helps,
Pat.
The most likely cause is indeed per cylinder correction. You cannot see what the other cylinders' knock corrections are on that datalog, only the instantaneous knock energy readings, ergo you can't see what the per cylinder correction is doing. The reason that you are seeing the effect more or less with RPM is simple aliasing. In order to effectively see each cylinder's final value you'de need to be logging at least twice as fast as the power stroke frequency; at 3000 RPM there are 50 revolutions per second and therefore 100 power strokes per second, You'de need to sample at least at 200 Hz to guarantee to see each final value. Anything less will be subject to aliasing. Of course that would need to go up to at least 400 Hz to make it reliably record every event up to 6000 RPM....
Hope this helps,
Pat.
Re: ignFinalPri1
Thanks Pat. I know you can only see the cyl4KnockIgnRetard in that screenshot, but I can assure you the other three cylinders are at 0 as well.
Actual per cylinder compensations in the cal are -1 deg on cyl2 and -1.5 deg on cyl4.
What you have said regarding the sample rate makes perfect sense and I am slightly annoyed I didn't realise that myself.
Thanks for your explanation.
Actual per cylinder compensations in the cal are -1 deg on cyl2 and -1.5 deg on cyl4.
What you have said regarding the sample rate makes perfect sense and I am slightly annoyed I didn't realise that myself.
Thanks for your explanation.
-
pat
- Syvecs Staff - Cleaner
- Posts: 356
- Joined: Fri May 23, 2008 10:23 am
- Location: Out there... somewhere
- Contact:
Re: ignFinalPri1
Richard,
The per cylinder corrections aren't shown in the log per se, inFinalPri1 is just the last ignition value that was seen at the time that the log entry is written; since there's no guarantee which cylinder was last you'll get this sort of "scatter".... the S8 can log each cylinder's leading and trailing spark angles, perhaps that's something that we should add to the S6... may avoid confusion!
Cheers,
Pat.
The per cylinder corrections aren't shown in the log per se, inFinalPri1 is just the last ignition value that was seen at the time that the log entry is written; since there's no guarantee which cylinder was last you'll get this sort of "scatter".... the S8 can log each cylinder's leading and trailing spark angles, perhaps that's something that we should add to the S6... may avoid confusion!
Cheers,
Pat.