rusefi.com

Can you change the load axis of the afr target table to something else then what the main load is using?

Middle-left says "Use TPS instead of Load"

AndreyB wrote: ↑

Sun Aug 30, 2020 10:52 pm

image.png

Middle-left says "Use TPS instead of Load"

I did see that in there. But that says for the ve table, not the AFR table.

My bad. No, that we do not have.

It would be a trivial checkbox to add but is that a useful checkbox?

OK, you got me curious, what are you planning to do that would need the afr target decoupled from the ve table and other fueling?

OrchardPerformance wrote: ↑

Mon Aug 31, 2020 1:34 am

OK, you got me curious, what are you planning to do that would need the afr target decoupled from the ve table and other fueling?

I never said decoupled from the ve table. The AFR table still sets the fueling you want. Just because I am in boost doesn't mean I want to add fuel. Maybe I'm at 7psi of boost at 3000 revs but want Lambda 1 and not .8. likewise if I am at 100% throttle I want my lambda at .8.

OK I see, so its more a case of being able to overide the afr table to do power enrichment above a certain tps point.
That makes sense.

For clarity, using the afr table to tune the labmda for normal running conditions and run lamdba one everywhere you don't have to use rich mixture for cooling (like an oem does) but still be able to run 0.8 lambda at full throttle.

I can see this being possible with one of the fsio tables as a lambda traget modifier but I do wonder if this also brings a requirement modify the ignition timing too

I think this way of handling fuel (driver demand based, essentially) is a great idea. I'd do a similar thing on my skyline as it's entirely possible to get a couple pounds of boost at 3000RPM at 15% throttle going up a hill - and I have no desire to be at .8 there. Being able to modify or otherwise change ignition timing based on AFR target would be required to maximize the benefit from having different lambda targets. You could also use this same workflow for lean-cruise, i.e. pump in a bunch of timing at low loads when targeting 1.1-1.2 lam.

I would suggest:
-Maximum lambda vs. load (to make sure that you get some enrichment when you get high loads at part throttle)
-Target lambda vs driver demand (to get more power from engine when driver asks for it, to get lean cruise when driver asks for it)
-Modify timing based on load and lambda target (to get timing and fuel on same page in terms of desired behavior)

blundar wrote: ↑

Mon Aug 31, 2020 1:38 pm

I'd do a similar thing on my skyline as it's entirely possible to get a couple pounds of boost at 3000RPM at 15% throttle going up a hill - and I have no desire to be at .8 there.

So why not just leave the AFR target table at stoich up as far as you want? You won't be at 120kpa anyway when wide open, so why gate AFR on throttle instead of MAP?

mck1117 wrote: ↑

Mon Aug 31, 2020 9:56 pm

blundar wrote: ↑

Mon Aug 31, 2020 1:38 pm

I'd do a similar thing on my skyline as it's entirely possible to get a couple pounds of boost at 3000RPM at 15% throttle going up a hill - and I have no desire to be at .8 there.

So why not just leave the AFR target table at stoich up as far as you want? You won't be at 120kpa anyway when wide open, so why gate AFR on throttle instead of MAP?

how do you know you won't be at 120kpa at WOT?

OrchardPerformance wrote: ↑

Mon Aug 31, 2020 9:52 am

OK I see, so its more a case of being able to overide the afr table to do power enrichment above a certain tps point.
That makes sense.

For clarity, using the afr table to tune the labmda for normal running conditions and run lamdba one everywhere you don't have to use rich mixture for cooling (like an oem does) but still be able to run 0.8 lambda at full throttle.

I can see this being possible with one of the fsio tables as a lambda traget modifier but I do wonder if this also brings a requirement modify the ignition timing too

Its more of the case of being able to use a VE based ecu correctly. The VE map shouldn't be used to deal with fueling. It should be used to map the airmass though the engine and any fueling changes should be allowed to be done however the user wants. The whole point of VE tuning becomes moot if the commanded fuel table is locked in terms of load and rpm.

hentai wrote: ↑

Fri Sep 04, 2020 4:59 am

OrchardPerformance wrote: ↑

Mon Aug 31, 2020 9:52 am

OK I see, so its more a case of being able to overide the afr table to do power enrichment above a certain tps point.
That makes sense.

For clarity, using the afr table to tune the labmda for normal running conditions and run lamdba one everywhere you don't have to use rich mixture for cooling (like an oem does) but still be able to run 0.8 lambda at full throttle.

I can see this being possible with one of the fsio tables as a lambda traget modifier but I do wonder if this also brings a requirement modify the ignition timing too

Its more of the case of being able to use a VE based ecu correctly. The VE map shouldn't be used to deal with fueling. It should be used to map the airmass though the engine and any fueling changes should be allowed to be done however the user wants. The whole point of VE tuning becomes moot if the commanded fuel table is locked in terms of load and rpm.

Can you elaborate on how you would do what you are requesting?
What would be your suggestion of the way to do it correctly?

For cable throttle:
1. VE table for airflow and just airflow.
2. Pressure vs. maximum lambda (leanest allowable commanded) Why? So you don't end up at 1.1 lam @ 220kpa @ 3000 RPM on a small-turbo car
3. TPS vs RPM for power enrichment mode enable
4a. Power enrichment: RPM vs target lambda table, pressure compensation for target lambda in PE, possibly minimum lambda setting to avoid over-richening?
OR
4b. Power enrichment: RPM vs. Load (outright pressure, grams/cyl, etc. Pick your poison) with table values = target lambda

For DBW, a torque based system makes sense.
1. VE table for airflow and just airflow
2. Pressure vs. maximum lambda to limit how lean you can be
3. TPS vs. RPM vs. desired torque
4. Torque calculation from airflow, modify for lambda, modify for distance from MBT timing, modify for vcam.

I would completely make use of the PE vs. non-PE modes on my RB20DET skyline because it's so easy to make boost at low throttle angles going up a hill.

now tracked on github: https://github.com/rusefi/rusefi/issues/1761

mck1117 wrote: ↑

Sun Sep 06, 2020 8:25 am

now tracked on github: https://github.com/rusefi/rusefi/issues/1761

this is now done for VE and AFR, i.e. the ask of this thread