Reducing K while idle, no effect?

[Torque]

My car is running a bit rich at idle, 12 AFR.
I decreased the injector multiplier while monitoring AFR, it had no effect!

To see if it had any impact at all I halved the K value ... no change in idle or AFR.

Then I increased K, that worked (engine starting to stumble at some point)

My K factor is 190
600CC injectors.

[RJBijsterbosch]

I''m not trying to give the right answer but i just want to think about it so that i understand nistune better.

Could it have something to do with the injector latency value. Because this is a fixed value i am assuming that Changing K factor and injection multiplier has no effect.

What i'm trying to say. When the total injector opening time @ idle is already close to the injector latency, then it will have less or no effect when decreasing the values you mention. It will only have effect when increasing. When the total injection time is 0.7 ms @ idle and the latency is set @ 0.7 ms than you can't get any lower than 0.7 ms of total injector opening time.

Did you watch what happened with the total injector time when decreasing/increasing the mentioned values.

Maybe i'm having it completely wrong.

[Torque]

Hello there,

I used K and Injector Multiplier interchangeably, it is the same thing.


The smaller the multiplier,the larger the ECU thinks the injector is.
So when I'm decreasing K (multiplier) the ECU should reduce injector opening times to achieve target AFR.

I have not logged injection times while reducing K since the car got really warm, I'll do this log next time around.


Have you measured your AFR at idle? (no gear/no speed)

Cheers

[Matt]

Your ECU has a minimum load (TP min) table

TP = K * MAFQ/RPM

If you lower K too much then TP may be clamped by TP min table. You will need to lower values in this table for lower K values to have further effect

[Torque]

I'm running AFR 11.9, I sit a lot in traffic and would like to run somewhat leaner

So let's do the calculation ..at idle:

1 Volt MAF (raw lookup 715)
750 RPM
K=190

How is that calculated please?

I get 181.13dec

That's an 8 bit number ... so I guess I did something wrong?

My TP Min is set to 8, here's the idle log:

idletp.JPG

(124.07 KiB) Downloaded 3857 times

[Matt]

ECU works with 16 bit numbers for the TP value (the TP MSB index is /256 which makes it 8 bit)

TTPmin would use the index (8 bit value).

The RPM value is also an adjusted number something like 750/12.5 = 60 RPM adjusted index for example

I don't have the exact details without going through the ECU code in finer detail, the formula is what I have worked out from reading it

Assuming 715 * 190 / 60 = 2264 / 256 = 8 TP value seems to be what you see there, and matches your log

[Torque]

Hi Matt,

Thanks for that.

I basically wanted to calculate what my TP would be with K lowered from (currently) 190 to 170 (the result is 7.9 down from 8.8 )

So TPMin does lock in the K and MAF values then?

[Matt]

TP min is a lower clamp on the resulting TP value (which comes from MAFVQ x K)

So if you have a lower K value or MAFVQ (lower MAF voltage) it will not go lower than TP min will let it

[Torque]

That's what I meant

On that note it might be worthwhile to put this into the notes pertaining to the injector resizing or MAF change procedures.
When you install bigger injectors (resulting in lower K) you might get a richer idle due to TPMin.

I will report back if it worked out for me, 11.9 is noticeably too rich

[Matt]

We now use TIM (feature pack firmware) for injectors. All newer boards are loaded with FP firmware (where available)

For the MAF resize, if one sticks within the factory TP range there should be no need to change TP min (in theory)

[Torque]

I reduced TPmin by 1 (from 8 to 7) allowing me to reduce K.

Consequently I could lean out idle mixture by reducing K (I was idling at 11.9ish) but for some reason the car didn't like
the leaner mixture and started to miss beats as soon as I moved leaner than 12.2.

I don't know why that is, my injection time was about 1.5mS, I have 600cc Nismo injectors (sidefeed yellow 555)
Possibly they don't fire reliably at these short durations? I honestly don't know.

My latency is 600us, I could lean it out there ... haven't tried yet.

As for TIM, what happens if you change the MAF?
Doesn't that still affect the TP scales?

[Matt]

Depends where you leaned out the mixture in TTPmin, the entire bottom end of just a particular RPM. Lowering the whole table could let it go leaner than what you would like, since it is a minimum load clamp table.

You could temporarily drop the latency and see how that goes

If you change the MAF, it will still adjust K constant (for the maximum capacity of the MAF). Usually people dont use the full MAF capacity so it is necessary to adjust K constant until you use the full map range again, and offset fueling adjustments with TIM.

This way there is no need to change the TP scales. I've written the procedure for this in our manuals and here forums if you search my previous posts

[Torque]

OK, thanks ... interesting.

Well my MAF (Q45) goes up to 4.99V in the logs at full boost that is, so I do use the full scale.

I'm not sure that the feature pack would be right for me, I'm not interested in launch control and flex fuel would be nice but here in the UK E85 is rare.

I've been honing my TP scales and warm up / cranking tables for years and don't really want a retune from scratch

[Matt]

Yeah probably not much point doing the upgrade then. The TIM adjustment was added because with people using large injectors (1000CC-2000CC) or MAFs (HPX in 4") the TP load readings were thrown way out after moving K

[Torque]

Flex fuel would be interesting, but I've never seen any gas station providing it .. otherwise I'd be more tempted.

My mate just blew his engine with launch control on a LinkG4 ECU.
The oil pump didn't like the stop and go @ higher rpm and broke, the whole engine went kaput.