G4+ Fury didn't active RPM limit when it should

[warmup]

The ECU is G4+ Fury, FW 5.5.7.2969, PC-Link 5.5.6.2942

The car is a 2012 STi with full built EJ257 motor.

 

I was tuning a 2012 STi's launch control yesterday, the car has been tuned on dyno. 

 

During the first pull on the road, I found that the engine can rev to 10000 RPM!

 

At first I thought that was related to LC or Shift Cut, then I turned LC and Shift Cut off, it can still rev over RPM limit I set.

So I turned off the "Advanced Mode" of RPM limit and did a stationary full throttle test, it over rev again !!!!

Then I turned on the "Advanced Mode" of RPM limit back, changed some parameters, better now.

(GP RPM limit is always off during the whole test)

 

I went to road test again, at 1st gear, full throttle, if I pushed the clutch down right before the rpm limit I set, the engine will randomly over rev to 8500 even 10000. 

 

Then I added two params to log, RPM limit and RPM limit(Status) and tried again,

this time I found that when engine over rev(logged engine speed > RPM limit), the RPM limit(Status) was keeping OFF!!

 

Sometimes the engine won't over rev during a test, when I check the log the RPM limit(Status) is "Active" when engine speed >= RPM limit.

 

If I wasn't very careful to turn off boost control during the whole test or I didn't install the best valve spring I can find to this engine, I would have lost the whole long block.

 

I attached the latest map, some logs and screenshots.

RPM_Limit_Actived.llg

AChang_PTE6266_V19_Copy.pclr

RPM_Limit_Not_Actived2(9000RPM).llg

RPM_Limit_Actived_Car_Is_Stationary.llg

[halltech]

Hi, I've had this problem when I first had my car up running.

Its not rev-limiting because your rev limit mode is set to fuel cut and your injector duty cycle is going over 100% which prevents the rev-limit from working.

You should set your rev-limit mode to ignition cut.

Is the onboard lambda sensor installed on the car? In the logs the lambda trace is pretty lean around 5000rpm and super rich up top which is why the injector duty cycle is going over 100%

-Cameron

Edited April 16, 2016 by halltech

[warmup]

Thank you halltech! That's really helpful and explained everything.

A previous set of ID 2000cc injectors were damaged by running E85, I'm waiting for new injectors, temporarily installed a set of 550cc to test launch control.

Fuel was only roughly tuned. That's why injector duty cycle was that high.

But what is designer's logic behind this? The basic rev limit should unconditionally active or what is the meaning of the limit? It should be a baseline of engine protection. 

A stock ecu will cut fuel at red line no matter whatever other conditions are.

[Simon]

Hi Warmup

It is an issue in the firmware however it is so deep in the code that making a fix could open up many more new issues.

We are looking at ways to at least make the tuner more aware the duty cycle is maxed out.

 

 

 

[warmup]

Hi Simon,

Thank you for the reply, I can understand what are you saying as a programmer. Looking forward to revises.

It will be appreciated if you note this feature in PC-Link's help docs.

 

[mapper]

Hi Warmup

It is an issue in the firmware however it is so deep in the code that making a fix could open up many more new issues.

We are looking at ways to at least make the tuner more aware the duty cycle is maxed out.

 

 

 

Sorry Simon that's no excuse!!

It's a error that can destroy easily a very expensive race engine! It's not seldom that we have to go up to 90-95% of Injector DC during tuning (because of to small injectors). It's absolutely not acceptable that the revlimiter doesn't work anymore if we have a overboost situation, colder air or what ever that forced the INJ DC to go above 100%. On most engines with hydraulic lash adjuster, ignition cut isn't a option, because that can lead to some mechanical problems (hydraulic adjuster pump up etc.). 

Edited April 18, 2016 by mapper

[Simon]

Hi Mapper

Sorry Simon that's no excuse!!

It's a error that can destroy easily a very expensive race engine! It's not seldom that we have to go up to 90-95% of Injector DC during tuning (because of to small injectors). It's absolutely not acceptable that the revlimiter doesn't work anymore if we have a overboost situation, colder air or what ever that forced the INJ DC to go above 100%. On most engines with hydraulic lash adjuster, ignition cut isn't a option, because that can lead to some mechanical problems (hydraulic adjuster pump up etc.). 

If you would like us to stop selling units and start the firmware from scratch to remove this issue and then have any number of other unknown issues sure.

It is unfortunate that it exists.  We will do our best to come up with a work around. This will include documenting it in the help file and if we can get extra warnings into PC link we will.

A Map limit should be in place to stop an over boost situation. And if you wish you can add a GP limit based off INJ DC so that if duty goes above say 90% limit engine.

 

[mapper]

Sorry Simon if my question was a bit harsh. The engines we tune cost sometimes a fortune and i would be a very bad and costly mistake if we crash an engine because of a bug I wasn't aware. 

I can understand after your explanatian that it isn't as easy as i hope. 

Anyhow, I like to make some suggestion that could maybe work for you.

- Programm a additional GPLimit in the background that cuts cylinders at 99% DC, or doesn't work no other limiter anymore after DC has gone over 100%?  Add a fault code, that users are aware what happened. 

 

- Span the y-axis of RPM 3D table with INJ DC

- output the DC to another class in the code, setup another RPM limiter that goes active in case DC>100%  i don't know how the code looks like, most likely much to complex for me, but maybe it helps to get an idee for a solution. 

 

 

 

 

 

 

[Simon]

Yep now you are thinking along the same lines.

I certainly think an alterative limit might work well. Will leave that up to the engineers to figure out.