I neverlog throttle position because of logging space. Once you save you're .csv file, use excel (or whatever) and add a column, name it throttle position, and where the ETC goes to 82%, make that 100 in the TP column. Works great!

 

Made some pretty good progress with the waste gate DC. Pushed the torque curve peak much later than normal for a k03. Also to my surprise it looks like it opens up room for more agressive timing in the same areas. I'm starting to really see the need for a downpipe

Sent from my SM-N900P using Tapatalk

 

So you're on the GTX now?



Sent from my iPhone using Tapatalk

 

I'll start the install tonight

Sent from my SM-N900P using Tapatalk

 

datazap.me | unorthatrox | gtx 104 4th stop short, HW, 4th

 

If you haven't corrected it yet, i had a quick scan of your datalog and from 5.5k rpm your wgdc is too aggressive and that's why you had such a major throttle closure.

 

yeah this was a pretty early log meant to develop my WGDC. At first I thought the closure was causing the lean condition but I realized I hadnt ramped the fuel yet.
for some reason that log Vdynos VERY high (345 vs the average 300 at this stage) and from what I can see the only change from previous was that I reduced spark a bit for safety, but I looks like its all put back in by the computer in the next version of the tune.

I've done most of the important VE logs now and running at the 1/8th mile tonight. Not expecting anything crazy since im on street tires. Taking some logs now shooting for 24psi. Everything looks clean from the 22.5 runs.

 

tweaked a few things and this is going to the track tonight.

 

Stock tires aren't helping. Crap 60ft times but I was trapping 85mph or better

Sent from my SM-N900P using Tapatalk

 

Any time slips?

 

Yeah but I didn't top 9.4s.

Sent from my SM-N900P using Tapatalk

 

Yeah that mph is impressive but the 60 is killing you

 

Sent from my SM-N900P using Tapatalk

 

So I decided even though VE still is giving me issues and I knew Id run into a little knock... to push for 25psi
4th Gear pull on 91
what I got was quite a surprise. Is it possible Im out of flow? or should I be hunting a boost leak?

I've also noticed sometime in the past week my air mass logs have gone crazy. I was hitting 40. Now they wave around under 30.

 

Yeah you got some MAJOR issues! Do not run your car like this!

Your airflow numbers are way off for that Boost @ that RPM.

Because Boost comes from the final MAP sensor, it should be indicative of your manifold pressure. Your STFT's are through the roof. This combined with your airflow estimates tells me your VE is WAAAAY off.

You're going to scorch your pistons awful quick here.
You are flowing high airflow amounts, but the computer thinks otherwise.

I'd also think there is a leak as well, because 100% WGDC should have that turbo blowing up.

You got some big issues. I'd recommend undoing any VE changes you may have made, as it is completely off.

 

Ditto on everything Bugasu said! No more WOT on this one!

Airflow and load are waaay off and you're pulling almost 5* of timing in cylinder 2. I'm seeing 25 psi around 68%dc down low and 75% up top, so there is something off .

Don't want to see you added to that "other" thread. Start over bud.

 

yup. I don't know if I got it backwards in my head last round of VE corrections or what. This one has absolutely imploded on me tho. I may go ahead and try to build from stage 0 and see what that path is like.

 

Going back to my previous tune. I still have indications that I'm out of flow for this big turbine housing. But most things are back on track. I was at a point where I wasnt sure where to go with VE because im walled up against the 100% limit of Act. Corr. I suppose the only option he is to follow the STFT percentages and bring those back in? I also had realized the slower spool didnt reach the threshold for optimal power on the VCT charts. Too many changes at once had gotten me in trouble after the tune below.
Old Tune

 

Even that guy is pretty poor.

I don't know why the ST caps at 100% VE. Perhaps Braden @COBB can fill us in? Perhaps the VE corr isn't the actual "VE" of the engine.

It looks like you're kind of doing it backwards.
If VE% goes up, that means Airflow goes up. If STFT is positive, you need to ADD VE. If STFT is negative, you need to REMOVE VE.
From your graph, it looks like Down low you're peaking at 100% VE and your STFT goes up. I would think you would need to increase your VE, but it looks like you can't.
Up top though, you're at 100% VE, but your STFT drops, meaning we're pulling fuel. So we're over estimating airflow. You need to subtract VE% here I would think.


The thing is, VE is really hard to tune with just datalogs of WOT. You really need some kind of steady state values. Another thing is your FRP keeps rising as your STFT goes negative. That could be because your injector mapping is inaccurate for those PSIs, so you're actually injecting more fuel than the ECU models.

There's a lot in play for VE tuning, and it's of critical importance that you can trust your fueling system is already accurate.

VE on the ST quite frankly isn't as straightforward as the tuning guide makes it out to be. My Optimum Power VE tables show 0 for everything. How is VE calculated then? Cobb hasn't released the full info on that yet.

I'd start over with a fresh tune, and work your way up. Try not to modify the VE tables much. The changes even for switching the turbo should actually be pretty minor except for when you're under boost.

It's really kind of unclear why Modelled can go over 100, but actual cant to me.

 

braden and i have discussed the 100% clipping. You can read the short exchange here.
http://www.focusst.org/forum/cobb-atr-software/24929-est-ve-actual-vs-corrected-clipping.html
I suspected zero charge tables being the culprit but I went in his suggested direction and my attempts trainwrecked. About half a dozen VE tables from the OTR stage1 91oct tune clipped severely like this when running my initial logs at spring pressure. It just seems the characteristics of the larger turbine are very different for some reason.

I need to track my steps and review logs to see what I've done because I had increased rail pressure to combat a high rpm lean at one point. My AFR has stayed very consistent since that change and I had planned on doing some detailed tweaking after VE's were dialed in. Hadn't considered the relationship between the 2.

The non-wot portion of my tune is working very nicely. I have a few logs of stable idle and cruise and mild tip ins. Everything is pretty accurate there with the exception of decel when the injectors shut off to save fuel. But thats really the last of my worries as its a no load condition.

The optimal power for VE is determined by 2 tables at the very bottom of the VCT section. Those 2 tables instruct the ECU on which HDFX intake and exhaust table to use (and transition) so long as all the thesholds are met. Those being minimumn loads and throttle position, maybe one more that i cant remember as Im at work right now. I at one point thought my slow spool up was not meeting conditions for the load and was not following the OP tables but after changing it (screwing up the tune) and now going back and verifying, the ECU was still directing use of the OP table even though I was not meeting mimimal threshold. (I dont understand how that is possible for for simplicity sake, I am going to leave that alone for now and keep the factory values.)
To be more detailed about which table is being used. During OP, the ECU "weights" each table as it transitions through the RPM. I assumed that weight would be a % of total VE, like 40% table 6 and 60% table 7 but the logs show it to be far more complicated than that. If you ever log all 15 HDFX weights on one log and show them all on datazap you may compare it to someone with an etch-a-sketch having a seizure.

So put all that together and you'll se how much i was changing in the span of 2 revisions. I would love to find a correct way to fix my clipping issue as simply guessing at Max Load at WOT did not help at all. Maybe if I make VERY slight adjustments to it to get the 100% down around 90 so I can see tangible numbers to make my VE corrections? I have a strong suspision now that altering that table is what made my load and air mass numbers go completely crazy.

 

more adventures. it was not max load at wot throwing my airmass numbers all over. Im playing with only a portion of my rpm which is fully weighted for table 06 right now and making adjustments only to it. This section is running at -11% trim and decreasing my VE by 11% only changed the trim by 1% and reduced the airmass of that section by approx 10%. Not at all the desired effect. It did bring the respective model vs act VE numbers about 5% closer and moved the act. VE significantly down from 96 to 90. Still I feel that something about using slope as a direct VE tweak is somehow incorrect by all the other calculations it alters. I think im going to break away from this and experiment only on table 06 for a while. its a pretty easy going cam lock up position.

 

Give the new alpha fw a shot, I think you'll find the HDFX monitors to be much more useful now!

Cheers,


Braden @ COBB
Sent from my iPad using Tapatalk

 

and Im going back out for another log. haha. THANKS!

 

2 log or not 2 log....lol

tizzapin

 

Ok I need some help here. Here is an HDFX log at spring pressure on table 06. This is a pretty mellow table for me as now extreme adjustments are required to get the STFT close. You also get a good look at my "cranky" cyl 2 in the corrections. I get about a 50/50 shot of cyl 2 deciding it doesnt wanna play nice with the others. My question is, how do I get VE modeled and actual to sync up? Ive experimented with tweaking offset, quadratic term, max load at wot, none of them have the desired effect during my testing.
HDFX 06

(4) (C) Est VE Actual – The estimation of Volumetric Efficiency using current measured aircharge.
(4) (C) Est VE Model – The estimation of Volumetric Efficiency based on the ECU calibration settings for Slope, Offset, and the current MAP.

Im considering loading up the factory torque to load and load to torque tables to see how those run. I have a haunting feeling the answer lies somewhere in the load as it technically aircharge. I'd love someone to chime in with the magical answer.

 

So I think Im seeing another correlation. TTL-LTT tables directly impact the VE as well correct? IE... if logged est. torque and load is lower than the TTL model. then the VE act will be lower than VE model by the same factor. Its hard to do the math without clear breakpoints but it looks pretty close. I had a margin of error around 4% and I think thats due to using the non corrected VE logs.