closed loop lambda failsafe

[mapper]

Hi 

Just wondering what happens if the internal Lambda controller detects a error. On analogue gauge i use closed loop lambda control just on the low load part of the map, because a fault sensor can force the ecu to take too much fuel out. Consequently the engine could run dangerous lean. I'm aware that there is a setting for the alowable control range for closed loop lambda.  

I'm wondering what the closed loop lambda control will do in case of a error: 

- if we use the internal lambda controller on the furry or Thunder

- if we use a external CAN-Lambda controller? for example does a ALM-CAN transmit all error codes to the ECU? does the closed loop lambda control shut off in this case or are there a chance that the mixture can go lean as on a analogue wideband?  

 

Edited April 18, 2016 by mapper

[Simon]

Fury and Thunder keep the last good lambda value but the error status is set. 

Have suggested to engineering that maybe faulting to a high Lambda is a better idea. So engine will go rich.

 

For an external CAN lambda its going to depend a lot on what the CAN transmits from the unit. Currently we are not geared to read the fault codes from an external device.

[Adturb]

I don't know about the Fury and Thunder, but in the Extreme's Closed Loop settings there is 'FTrim Limit +/-' so the ECU won't compensate past that level.

[Ducie54]

Thunder has the same feature.

[mapper]

Yes I know, thtats what i've mentioned above.

[mapper]

Thank Simon for clarify. Yes would be great to get a possibility that closed loop lambda shouts off or goes to a high lambda. Maybe the engineers can integrate a plsusibility check of the signal for error detection.

For example. Input value must change at least once a second while engine is running. 

Regarding fault code. Can i setup the fault codes manually  from the ALM-CAN. Because it's just a controller without a own display. So would be good to have this infos. 

Otherwise i would really like to see a external wideband controller with all the functionality and diagnostics from Link. What i'm aware there is something in development, which is great! 

 

[Scott]

Hi mapper, 

We do have some fault codes built into the firmware for the internal lambda controllers on the Fury and Thunder ECUs, these are in a particular order. The problem is that the fault code numbering from a third party CAN-Lambda device would have to line up with the order that we use, otherwise the wrong fault code would come up.

[warmup]

I'm wondering if you can separate +/- to two parameters from Closed Loop FTrim Limit,

then we can limit the negative fuel trim in a very small range that won't cause damage.

A fail safe can't prevent a half bad o2 sensor from sending back wrong signals.

Some stock ECUs do their closed loop f-trim this way.

 

[Scott]

Hi warmup, this is certainly a good idea, and is currently on our request list.

[mapper]

If we already separate the +- parameter, why not do a Table?

Around idle, there is often more than 5% correction needed to cover all runing conditions (e.g after a hot restart, heat soaking etc.), under WOT and high RPM just a small correction should be applied to prevent lean running as we spoke about. On overrun situations there is often a lean spike which forces the system to go on to the rich limit. This correction are than applied for a small moment at idle, that scholdn't happen and could be limited with a 3d ftrim limit. (set less ftrim limt at load load and high rpm) 

Edited April 22, 2016 by mapper