Nistune Forums

Okay. New plugs installed, created a new tune on the correct base. The car is running alright now, but i need to tweak it to get it happy idling and such.

Now I have a question. I resized the injectors with the built in tool. Doing so adjusts the k-value to from 288 (stock 370cc injectors) down to 117 (for my 910cc injectors). As you know, this also adjusts the load scales, TTP max/min, and the cranking tables.

Now, say I need to adjust the K value to get the car to actually run happy. I have to adjust it to approx 138 k value to do so. Since that's 21 points higher than the resize tool set it to. Would I be better off making the adjustment through the resize tool by 'lying' to it to get the k value to 138 (as doing so would also make the TTP, load scales, and cranking table adjustments), or should I make the adjustment just to the k-value in the "injection multiplier" window?

I know fine tweaking is needed, I'm just trying to figure out what the best practice is for this type of adjustment.

Thanks for the help guys! I miss my car

Good question. I usually use the Injector Resize tool and then tweak to suit myself. Often you'll find that K will only need a small tweak anyway - so the other parameters don't need major work.

Load scales should be adjusted during final tune as they are based on both K constant and maximum boost - so they'll rarely be correct after using the Inj Resize tool as std load scales are designed for std boost.

Bottom line is that the tool will give you a good starting point for any of these parameters. But final tuning is still up to you.

PL

That's what I figured. The only reason I ask was because I had to make sure large change to the k constant and I wasn't sure if it made more sense to have to adjust everything else. I know about tweaking the load scales to get the map best utilized, but I don't know about the cranking table or TTP (info on these would be appreciated!)

Anyway I actually had to bring the K value up to nearly 150 to get the car to run nice and smooth. Seems weird to me to have to have it so high considering the injectors are 910cc... wouldn't a K value of 150 seem very high for 910cc's? or am I worrying about this for no reason?

Cranking table can be hard to get right - the Inj Resize tool just multiplies values by old inj size vs new inj size. From there it's trial and error I'm afraid. Not usually hard to sort. And usually only an issue if using big injectors. Yeah - that'd be you...

TTPmin/max can probably be left alone. If TTPmax is too low then you'll see inj pulsewidth being limited and it'll lean out - pretty easy to pick.

You may need to juggle K constant and Inj Latency to get your idle right. Once again everything becomes much more critical once you have big injectors. I suspect you'll need to pull K back a bit and increase latency.

Of course the fuel table values will also come into it once you load it up. The key is to find a good combination of K, Inj Latency and fuel table figures to get AFR's sweet everywhere.

PL

If TTP max was too low, you mention I'd be leaning out. Do you mean leaning out as RPM/load increases? (like hard to keep AFR down?)

I've got latency at about 980 uS and it's nice with the 150 K. I 'can' drop the K down to around 120 and up the latency to say 1100 uS and it'll idle great, but it's definitely running rich from there on up. My wideband is giving me some issues right now so I'm waiting to toy with the car will I get that back in order to pull some logs.

Yeah, you usually try to load them up and they'll just hit a wall where you keep adding fuel (via K or fuel map values) but they won't go rich enough. That's when you often find that TTPmax is too low.

PL

Understood. I don't think I have any issues with TTP.

This has been a very helpfup thread for me! Thank you guys.

I'll report back when I can gather some datalogs with timing and AFR. The engine is already running much smoother so I'm excited for the results.

Did you want your board reprogrammed?

No sir! thanks for the offer though. I believe the issue has been resolved. I switched to the correct base file, completely re-tuned the car, and its running better than ever! Logged some runs and the timing is reasonable again.

What is the best way to effectively lean out the fuel through the entire power band? I am running a bit too rich throughout. Would decreasing the K-value be better than dropping the filter values on the whole chart?

Theoretically you should leave K where it is if you've got it running nicely (idle, low load, full load all good).

But in reality I'll sometimes give K a tweak if I just wanna bump the overall fuelling up/down a bit...

PL

First, 48P01 is strange, it has DTC EGT flag enabled ....

You should decrease K down to 120 and adjust latency accordingly to get stable idle, and then tune all the values inside your maps. 1.11ms latency is not a crazy value.
On my Z32, I run K 200 and latency 1.12ms on 740cc on E85 @3bar (higher K) with all the targetted AFR in all ranges, that's equivalent to 550cc in my case. You will notice that 910/550 = 1.65 and 200/1.65 = 120 ....
Regarding TP, you reached 45-50 that corresponds to 75-80 range in my case where I reach at full boost 14.5, 15.5psi peak with 10.5 AFR peak (11-11.5 stable).
I don't understand why you told you were running rich with such values.
You should tune idle and partial load with O2 sensors disabled first, otherwise everything is jeopardized ...

Btw, your timing @ full boost is a bit too high on normal unleaded. I run your values 22-24° in the 4000-5500 range BUT I run Ethanol. And I know I'm on the edge regarding timing particularly in the 3000 range on full boost i.e. TP=80 (I still have knock voluntarily at mid boost). I can be retarded down to 11° when fully boosting from 2500rpm ....

Is this a Z32?

Torque wrote:Is this a Z32?

Obviously, yes ...

Doing so adjusts the k-value to from 288 (stock 370cc injectors) ...

Since you have posted several threads with these issues, I'm offering an exchange over board. Please email me and I'll send one with the correct base image for your ECU and it may resolve these problems

What is the connection to the base image then?