Jump to content

Overboosting / spiking between gear shifts


Recommended Posts

HI all,

Just wanting to pick peoples brains over an issue i've been trying to sort for awhile.  I have an fg mk1, 12psi actuator on a gt3576 with 37mm port, 4inch dump and high flow cat with a full exhaust.  For a long time now i've been trying to tune out 'overboosting or boost spikes' on gear changes , mainly 1-2 and 2-3, with varying success.  I'm running 16psi to 4750 which tapers down to 10psi at 6000rpm.  By doing this i've managed to keep the spike between gears to about 18.5-20psi. Any higher than that causes boost cut on gear change.   But to allow this to happen i have set auF0296 ( Overboost required for open loop duty cycle ) to   -22hg  and all the cells in the WG DC table the gear change may fall into to 0.1DC(completely open).  I've set au0296, to -22hg as it seems on gear change it uses the cell in the previous when calculating an error.  eg. 2nd gear @ 6000rpm, has commanded 10psi. So on change to 3rd if it hits 10psi +11psi error then it will boost cut.  It wont calculate it on the cell it is changing into. ie 3rd 3500rpm (16psi + 11psi).

So to the not so good.  If using manual mode and changing gears earlier at higher than 10psi. eg 15-16psi @ 4000-5000rpm. will cause the boost to spike higher on changing gears than just running through the gears in perf mode and letting the boost taper off.  Boost cut happens and it all gets a bit nasty.  This occurs regardless of how open the wastegate is.  I command it to open all the way but it seems like its not fast enough to vent between gears.  With my port job i don't suffer any boost creep or inability to control boost other than on gear change.

Should i be worried about how high boost gets on gear change?  Is there something i'm not changing in the tune i should be looking at?  I feel like if just up the error and overboost values to -30hg etc than i wont get boost cut on gear changes but then i would also not have any protection in gear(able to hit ~30psi).  Although this is unlikely to happen.

Any pointers / suggestions welcome. Any one experiment with this 'issue'?

Cheers guys.

 

 

Link to post
Share on other sites
1 hour ago, finnigan001 said:

Has anyone played with auF2714- Calibrateable wastegate duty cycle when gear change is in progess ?

I'm going to see what it does, and i'll report back any progress.

Yes this will limit boost on a gearchange, the problem with this is it will hold it for too long, so you'll get the opposite, a boost drop on gear change.

I would recommend playing with this table instead:

auF3002

Change in spark adder to waste gate duty cycle.

image.png.53adad378efc4981e2889ba4cc17bd7b.png

During a gear shift you should be getting spark retard (what causes the boost spike). Log the spark during your gear shift and work out the spark delta.

Eg if you should be getting 12 degrees spark at 5000 rpm 2.2 load, and on a gear shift you see 2 degrees your spark delta is 12-2 = 10 degrees. Hence the PCM will be multiplying the waste gate duty cycle by -0.0135. Eg a 1.35% drop. So if you were commanding 20% duty you would see 20 + (20 * -0.0135) = 19.73% duty cycle.

You could then try blowing these numbers out drastically, eg change the cell at 10 degrees to -0.2 (20%) so you get a 20% reduction in duty cycle. Eg you would see 16% duty cycle instead of the commanded 20%

  • Like 1
Link to post
Share on other sites

I've also heard of someone leaving the car in closed loop and setting the base duty cycle to be 10-20% lower than what is required. Eg command 40% duty cycle when 55-60% is required for the boost in the desired boost table. Then during a gear shift (open loop duty cycle will be forced) it will drop to 40% stopping the overboost. This is probably the wrong method, however it is very quick and easy if you are limited on time.

Link to post
Share on other sites
1 hour ago, Roland@pcmtec said:

I've also heard of someone leaving the car in closed loop and setting the base duty cycle to be 10-20% lower than what is required.

This has been how i've been trying to control this thus far, with limited success.  Definitely effects drivability  when manually accelerating through a gear. eg 2nd 1700 - 5000rpm.

 

1 hour ago, Roland@pcmtec said:

uring a gear shift you should be getting spark retard (what causes the boost spike). Log the spark during your gear shift and work out the spark delta.

Definitely going to try this out. I do get a spark retard on gear change. IIRC even in to -ve numbers during the change.  This maybe the final piece in the puzzle to controlling this.

Cheers,

Link to post
Share on other sites
1 hour ago, finnigan001 said:

Definitely going to try this out. I do get a spark retard on gear change. IIRC even in to -ve numbers during the change.  This maybe the final piece in the puzzle to controlling this.

Cheers,

Negative numbers are going to spike boost quite dramatically. I think this will be your solution. This will also activate during knock retard, so if you had a lean out during cornering etc this will stop a boost spike.

Let us know your results. I would be very curious to see a before and after data log as well.

Link to post
Share on other sites

Rolls i could kiss you :)

As you suggested auF2714- Calibrateable wastegate duty cycle when gear change is in progress did not help, boost spike was still prevalent and boost cut kicked in same as before.

However change the auF3002 - Change in spark adder to wastegate duty cycle, which i read as doing the opposite (changes spark based on WG),  FIXED EVERYTHING :).  I modified the the larger changes( 10-20deg delta) to -70% wastegate and now it actually changes the wastegate during the gear change.  As i cant read the WGDC with my logging, i assume that whatever the wastegate DC is at the time of gear change start is what the pcm uses during the change until it is in gear and then uses the lower rpm cell when in gear. eg. 0.6 @ 4500rpm -> holds 0.6 during shift -> 0.1 @ 3000rpm.  Which doesn't allow the gasses to release fast enough causing the large uncontrolled spike.

With this change i assume now it is doing this: 0.6 @ 4500rpm -> cuts it by 70% for every large spark cut(0.2-0.15 so basically wide open) -> 0.1 @ 3000rpm.  I still need to do further test on how much reduction i actually need. I choose 70% as i wanted the wastegate to open as close to the amount that it would be open at after the gear change.  I did this as i noticed i didn't have the spike issue revving through the gear when i set the 6000rpm cell similarly.

And now i don't have to artificially have the wastegate open wider than what is commanded in the mid range, the car will be smoother from lower revs through the gear.

Thank you so much!

Data Log - boost cut on short shift.xlsx

Data Log - no boost cut or spike.xlsx

  • Like 1
Link to post
Share on other sites
  • 1 year later...

That makes sense as to why it wasn't on my mates car, I found it on mine though.

I've just been playing around with both of those scalars and have found auF2714 'Calibrateable wastegate duty cycle when gear change is in progress' to do nothing that I can notice or see in the data logs.

On the other hand when I adjust auF3002 'Change in spark adder to waste gate duty cycle', it makes my car bounce between 0 and 1 WGDC (Images Below).

I'm not too sure if its because my car uses an injector cut mainly instead of spark retard, but that doesn't explain why it oscillates while on boost.

Screen Shot 2020-11-19 at 1.05.46 pm.png

17.5PSI Gear Change Clip - Spark Delta.csv

Link to post
Share on other sites

You need to enable auF0304 "Wastegate allow Max Duty Cycle on Shift" and set it to 1 for auF2714 to be used.

If your car is oscillating whilst on boost when auF3002 is used, then it means you are likely getting knock with a constant spark adder (subtractor) being applied.  Looking at your log I'm seeing the average at ~-3 degrees which should cause no duty change, however the actual spark adder for each cylinder could well exceed this explaining it. Log your spark adder for all 6 cylinders and see what it says.

  • Like 1
Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...