WGDC Drops. Default to WGDC Base. No PID added

[WOT808]

Hey all,
I've been having this weird issue for about a month now.

• Symptom: Spool will slow and boost strays far from target. For some reason, the car defaults to the WGDC base value and PID doesn't intervene. Seems like some sort of limp mode (no codes or CEL).
• Cause: This usually happens after a good multi-gear pull. I caught it happening during a pull in the log below. I can't pinpoint the trigger.
• Temporary Fix: Restarting the car fixes the issue.

Next step is to switch to one of my older maps to see if that solves it.
Do you have any experience with this?

Car:
N54
Stock Turbos
6 speed MT
FBO + Inlets/Outlets
MHD flash (based off of an SPX file)
N20 Sensor
E50 Fuel
WMI

https://datazap.me/u/m6keharr/5-and-6-injector-swap?log=0&data=3-7-21-29-30-31&zoom=2207-2460

 

[Jake@MHD]

I think this happens when PID takes too long to reach target. There's some timers in there for it I believe, then PID is turned off.

Try to set your base table as close to what it should be as possible and see if it happens then

 

[WOT808]

I think this happens when PID takes too long to reach target. There's some timers in there for it I believe, then PID is turned off.

Try to set your base table as close to what it should be as possible and see if it happens then

That makes sense. I'll try that and report the result. Thanks!

 

[doublespaces]

That makes sense. I'll try that and report the result. Thanks!

Curious if that solves your problem, let us know

 

[WOT808]

I lowered boost ceiling a little and added some wgdc, not enough apparently.
https://datazap.me/u/m6keharr/log-1507203232?log=2&data=3-4-22-30&solo=30

I'm going to lower target after ~5.5k rpm and add a little more wgdc tonight.

 

[Twisted Tuning]

I think this happens when PID takes too long to reach target. There's some timers in there for it I believe, then PID is turned off.

Try to set your base table as close to what it should be as possible and see if it happens then

this is EXACTLY what the issue is. Once you exceed the timer, it bases out. there are timers for over and underboost.

OP, if tuning yourself, Zero out adders. get boost within 1-2psi of target. then adjust PID.

 

[tony@codewerx]

Along the way ... taper the boost targets.

These are stock turbos ...

 

[WOT808]

Messed with the BLM. Still need to get actual boost to better match target. Log below is after several multi gear pulls.
I'd say problem is solved. Thanks for the suggestions!

https://datazap.me/u/m6keharr/bpc-2k17?log=0&data=3-5-6-7-8-9-10-11-22-29-30-31&zoom=3948-4121

 

[WOT808]

Along the way ... taper the boost targets.

These are stock turbos ...

Whats target makes the best power top?

 

[tony@codewerx]

Whats target makes the best power top?

Usually I see 14 psi @ 7000 in 3rd in totally stock rides (no dp, no inlets, no outlets).

A lot depends on how the Vanos is tuned.

There is a proportion of effort that needs to be worked out between the engine and the turbos in hitting the VE that is targeted. Especially so if it involves stock turbos.

You have turbo outlets installed so your logs will show a better result.

 

[WOT808]

Let's bring this discussion back to life.

I'm still trying to find the right BLM.
Seems like 2.5 BLM was doing the job until it 'adapted' the next day. Logs below

Day 1 log
https://datazap.me/u/m6keharr/log-1509226834?log=0&data=3-5-22-29-31

Day 2 log
https://datazap.me/u/m6keharr/log-1509364320?log=2&data=3-5-22

Doesn't look like PID drop was an issue before the N20 and 3 column scaling.
The log below has a 20 psi target >6k rpm and never dropped PID.
https://datazap.me/u/m6keharr/v108-morrisville2k17?log=0&data=3-5-22-29-30-31&zoom=10754-11079

Are there any other timers to consider?

What are your thoughts?

 

[RSL]

Doesn't look like PID drop was an issue before the N20 and 3 column scaling.

What are your thoughts?

MAF change with scaling. MAF is axis on WGDC, BLM, etc. tables.

 

[WOT808]

MAF change with scaling. MAF is axis on WGDC, BLM, etc. tables.

The MAF breakpoints in the tables I've looked at fit my scaled maf values well.
Is there a way to turn this PID timer off?

 

[RSL]

The MAF breakpoints in the tables I've looked at fit my scaled maf values well.
Is there a way to turn this PID timer off?

That means almost nothing, it's the values in the cells for the breaks that matter, including adders/ceiling. PID shouldn't have to do much work at all if base is dialed in.

 

[tony@codewerx]

Let's bring this discussion back to life.

I'm still trying to find the right BLM.
Seems like 2.5 BLM was doing the job until it 'adapted' the next day. Logs below

Day 1 log
https://datazap.me/u/m6keharr/log-1509226834?log=0&data=3-5-22-29-31

Day 2 log
https://datazap.me/u/m6keharr/log-1509364320?log=2&data=3-5-22

Doesn't look like PID drop was an issue before the N20 and 3 column scaling.
The log below has a 20 psi target >6k rpm and never dropped PID.
https://datazap.me/u/m6keharr/v108-morrisville2k17?log=0&data=3-5-22-29-30-31&zoom=10754-11079

Are there any other timers to consider?

What are your thoughts?

Try reducing the BLM more at the high end ...

 

[chrisdfv]

I'm struggling with this issue right now. If I set base too low, pid will match target right on, but then will time out on shift or at the top of the gear. If I raise base closer to after pid, I overshoot target/get throttle closure, but PID will continue to add wgdc instead of staying close to base. If I tune with target way out of the way and rely heavily on PID, it does not time out. I'm not sure how to approach this issue. Raise D-factor to give it a longer leash?
Boostbox
Scaled N20 flash


https://datazap.me/u/cd919/v4?log=0&data=3-6-23-30-31

https://datazap.me/u/cd919/v9?log=0&data=3-6-8-25-28-30-31

 

[RSL]

After PID is lower than base in the ones with the closures, so it's trying to pull, just not doing enough. If you want to run boost psi that close/on target, I'd give it more room to pull harder. Not sure the PID off was a timeout. It was on target with low WGDC and happened during shift, so I'd hazard a guess that maybe an overboost trigger kicked in. No 30FE?

You could certainly give it more room to float over target without reacting if you're not too worried about it, give it a bit more space from the target, amplify the reduction responses or some combo of all.

 

[tony@codewerx]

I'm struggling with this issue right now. If I set base too low, pid will match target right on, but then will time out on shift or at the top of the gear. If I raise base closer to after pid, I overshoot target/get throttle closure, but PID will continue to add wgdc instead of staying close to base. If I tune with target way out of the way and rely heavily on PID, it does not time out. I'm not sure how to approach this issue. Raise D-factor to give it a longer leash?
Boostbox
Scaled N20 flash




https://datazap.me/u/cd919/v4?log=0&data=3-6-23-30-31
View attachment 13342


https://datazap.me/u/cd919/v9?log=0&data=3-6-8-25-28-30-31
View attachment 13344


View attachment 13343

I'm struggling with this issue right now. If I set base too low, pid will match target right on, but then will time out on shift or at the top of the gear. If I raise base closer to after pid, I overshoot target/get throttle closure, but PID will continue to add wgdc instead of staying close to base. If I tune with target way out of the way and rely heavily on PID, it does not time out. I'm not sure how to approach this issue. Raise D-factor to give it a longer leash?
Boostbox
Scaled N20 flash


https://datazap.me/u/cd919/v4?log=0&data=3-6-23-30-31
View attachment 13342


https://datazap.me/u/cd919/v9?log=0&data=3-6-8-25-28-30-31
View attachment 13344


View attachment 13343

Load request and actual looks awkwardly close.

What did you set in the BLM table ?

 

[chrisdfv]

After PID is lower than base in the ones with the closures, so it's trying to pull, just not doing enough. If you want to run boost psi that close/on target, I'd give it more room to pull harder. Not sure the PID off was a timeout. It was on target with low WGDC and happened during shift, so I'd hazard a guess that maybe an overboost trigger kicked in. No 30FE?

You could certainly give it more room to float over target without reacting if you're not too worried about it, give it a bit more space from the target, amplify the reduction responses or some combo of all.

I’d love to have Boost just below target but I have to ride out PID and it will usually end up reverting to base at some point.
No overboost codes. It will do this every time on back to back tries.

Load request and actual looks awkwardly close.

What did you set in the BLM table ?

It’s set to 3.0

 

[chrisdfv]

I’ve inclreased my PID values on the negative(left side), bumped up D-factor mult and lowered base about 2%

No real improvement and wgdc reverts back to base before the end of the log.

https://datazap.me/u/cd919/v10?log=0&data=3-25