Self-Tuning to ease the boredom.

[dyn085]

is boost psi reaching max TIP? that may be causing the closer, if not the duty cycle for that spot is probably low.
oh.. are you resetting KAM for each octane to allow OAR to re-learn? I don't like how we can't reset OAR to '0' through the AP, the only way is to uninstall and re-install the AP.

It shouldn't be. Max TIP is currently set to 2.65 bar, and I only ever briefly touch 23 psi. Admittedly though, I know little about the turbo system as a whole-which is why I leave it set to the OEM or Cobb values. I think I'm just throwing too much fuel at it down low. I think I'm going to lean it out down low and enrichen it up high compared to .8 lambda across the rpm range. If I'm right that should both correct my throttle closures and also give me some more room for timing up top.

OAR is a continually-learning process (under the assumption that the criteria is met of course) so there should theoretically not be a need to ever have to try and intentionally reset it. With that being said, my car has been flashed back to stock more times since I've started than I can describe because it has to keep going back to the dealership to get the stupid hatch fixed. Originally I would load a map, force OAR, then tune. After not being able to force OAR on the 91 tune/89 fuel I gave up and just started tuning 'around' it. That's when I ended up softening up the map to the point that suddenly OAR made a jump, and that's also when I suddenly understood how the whole system came together.

If that makes any sense.

So now I've decided to force OAR with my current fuel and have reloaded my timing tables from before making the octane switches, and coupled with the fueling changes I'm curious to see how it turns out. Good, I hope. The car is not due back into the shop until the 29th as the mechanic working my case is going to some related schooling, so I have ten days to tweak.

And for clarification for those that are following, I don't flash back to stock in some attempt to scam out work or hide anything, I have to do it because they need access to the OBDII port and have had to flash systems or whatever it is they are doing. When I sent it in the first time the AP3 was still plugged in and they lightly complained about it inconveniencing them, so I just flash to stock and remove everything so that they don't inadvertently screw me and get the extra convenience. I have a very good relationship with my dealership and they try to stay in the loop about mods to my cars because they know it helps draw in business and that I'll eventually be back again for another trade, and I have no problem with accepting the consequences if something goes amiss.

Edit: Hahaha, that went horribly.

 

[koozy]

Your tunes must be closer to the edge or the difference in fuel quality between octane points are enough to effect your OAR. My OAR is always at -1, hence why I wish I could reset it to 0. From Cobb's documentation OAR is persistent between flashes, changes occur through the learning process that begins when coolant temps reach 176 deg F. with the OTS maps.

 

[dyn085]

Your tunes must be closer to the edge or the difference in fuel quality between octane points are enough to effect your OAR. My OAR is always at -1, hence why I wish I could reset it to 0. From Cobb's documentation OAR is persistent between flashes, changes occur through the learning process that begins when coolant temps reach 176 deg F. with the OTS maps.

They absolutely are, but they also don't have the KS freedom to wander more than a degree or two. Basically, I raise my timing base as much as I can based on cylinder 4, then I don't allow the group to wander. Cylinders one and three could usually go farther, two a little bit less, and four almost none.

You stated that you were tuning based on the OAR being at -1, so just do the math to figure out what compensation it's allowing at each rpm/load and then subtract it from your total timing. It's more complicated than tuning at OAR 0, but if you are able to take your corrections down then you'll know that you have the added safety of it adjusting down for worse fuel.

OAR is persistent through flashes, but once it reaches its listening thresholds it will adjust more if necessary, regardless of how long is sat at whatever reading. I found with the FiST that it definitely continues to work and adjust when I intentionally filled with 87 while running the 93 OTS tune. I still had to push the car pretty hard to get enough actual knock and have it show up long enough to have the OAR drop, but it does continue to monitor and work. Just know that before it does the KS's have to be lighting up consistently or aggressively.

Part of my curiosity came from a thread in the FoST community that showed a bunch of FoST owners are too cheap to run anything but 87. That's a whole other conversation, but it definitely brings to light the fact that the FoST runs perfectly fine on 87 and that the majority of those in the blown engine thread were using something higher along with a pro tune. My underlying thought was that if Ford can tune the car with OAR to run multiple fields on a single tune that there's no reason any of us couldn't either-so long as we were aware of how OAR works.

The way I'm approaching it now (as of today) is that I either tune on a mid-grade fuel and OAR 0 or a high-grade fuel on OAR -1 and a lot of math. So long as you shut off the system while doing the tuning and turn it back on at the end, that should really in a tune that works all around, regardless of fuel. I'm over-simplifying, but that's my theory at least.

 

[koozy]

I'm currently tuning and tweaking an e50 tune. My approach has been getting the tune solid for e10 and then rely on the use of global timing to get me in the ball park for other tunes mainly because I want the tune to work for e10 - 50 through the switchable maps without having to load a map each time for different blends. Much easier to change maps on the fly through the cruise control buttons. I'm really hoping Cobb will soon release flex fuel capability through the AccessPort.

With some insight into how the ECU compensates for various octane levels I'm not worried about running 87, as long as the map is tuned to take that into account. Since running a tune I've used nothing less than 91, the commonly highest available here. Before the running a tune all I ran was 87, because it was cheaper and I felt anything higher was a waste with the type of driving I was doing. I also didn't care about squeezing every little HP/Torque from it. I'm now running blends of e85, rarely ever on straight 91 anymore. I started with e30 and now at e50, because well.... it's cheaper LOL, but more importantly e85 packs a punch to the tuning envelope in being able to extrapolate power from it.

 

[dyn085]

I'm currently tuning and tweaking an e50 tune. My approach has been getting the tune solid for e10 and then rely on the use of global timing to get me in the ball park for other tunes mainly because I want the tune to work for e10 - 50 through the switchable maps without having to load a map each time for different blends. Much easier to change maps on the fly through the cruise control buttons. I'm really hoping Cobb will soon release flex fuel capability through the AccessPort.

With some insight into how the ECU compensates for various octane levels I'm not worried about running 87, as long as the map is tuned to take that into account. Since running a tune I've used nothing less than 91, the commonly highest available here. Before the running a tune all I ran was 87, because it was cheaper and I felt anything higher was a waste with the type of driving I was doing. I also didn't care about squeezing every little HP/Torque from it. I'm now running blends of e85, rarely ever on straight 91 anymore. I started with e30 and now at e50, because well.... it's cheaper LOL, but more importantly e85 packs a punch to the tuning envelope in being able to extrapolate power from it.

That's exactly my plan as well, as soon as I can get E85 within a reasonable distance. I figure that I can have one map to handle my regular octanes and use the other four for various ethanol blends. I have the feeling that if I ever get near it I probably will solely operate on blended fuel.

Can you grab me a shot of how the OAR Learning Modifier is set up in the FiST? My FiST learns nearly instantaneously (within a couple miles of normal acceleration) while my FoST seems to take days or weeks without being forced, and I'm curious as to what the difference is. Idk if the base-map in the FiST is just that soft or if the modifier is that much more drastic in the FoST, but I would like to see if that's something I should tweak for better response. Driving around with tons of negative correction for long periods of time bothers me...

It's located in the Learning (Octane Adjust) folder in the Octane Adjust Ratio folder in the Ignition Timing Tables folder. This is how it's set up on the FoST-

 

[koozy]

OAR learns fast on my car. When I last removed and re-installed the AP I could see the learning happen as the numbers trended towards -1 every 40 seconds or so, by 15 minutes -1 was reached. You will love how the car responds with e85. I'm lucky to have a station within 5 miles.

 

[dyn085]

Awesome, thanks for posting that. It's pretty clear now why the learning is so drastically different between the cars-

FiST-

FoST-

 

[koozy]

Just checked the rest out of curiosity, they're all the same as posted above except for Stage 0 "stock".

Stage 0

 

[dyn085]

Lol, looks familiar...

I went and checked the stage 0 map on the FoST and there's no changes outside of the max advance/retard tables and one or two of the listening range disablers. Now I guess I'll go try some random things.

 

[dyn085]

Ok, so I have my 'normal driving' timing set to basically where I want it for now. To test my theory on OAR, I've adjusted my learning table to look like this-

My KS max timing advance allowed is only 2 degrees, so I rescaled the table to reflect that. I still have retard set to -6, but I want maximum correction at -4 to hopefully keep the car from getting into that range as quickly as possible. I also lowered the threshold to .65 load (as opposed to .8) so that corrections should occur sooner. The way that I'm thinking is that this overall setup should allow for relatively rapid negative corrections and settle out relatively quickly without allowing it to advance too rapidly. For now, this is how I'm leaving my 93 tune before putting 87 (maybe 89 because 87 sucks) in it.

If I'm right, I'll see negative corrections relatively quickly and the OAR should pick it up faster and adjust in the positive direction. Depending on how much knock occurs for however much time, the timing should drop incrementally relatively quickly until the negative corrections dissipate. After the corrections dissipate, I shouldn't see and movement on the OAR in the negative direction until I'm seeing repetitive KS additions over 1 degree cumulative. That's my hypothesis, so I guess it's time to put some 87 in and find out.

 

[dyn085]

Just for fun, here's the Fusion OAR setup-

 

[dyn085]

And the Mustang OAR (looks kinda familiar)-

 

[dyn085]

I guess I haven't updated this, so I'll paste in a couple of posts.

-Ok, so I just got back from a run to the grocery store and the OAR is definitely picking things up much faster. Within 12 miles of getting fuel the OAR had made a full transition from -1 to 1, after having added 5 gallons of 87. Corrections went from full-blown crazy to not so bad and sometimes even with KS additions, but I noticed that once negatives started to happen under acceleration it just started compounding. I logged part of my trip home, so I'll take a look at that in a bit to see if it gives me any insight.

I also noticed that I have it set to learn a bit too aggressively, so I think I'm going to turn down the learning from positive KS compensations and scale it out a touch differently. I don't mind it taking a little bit to learn when fuel is good, but I went from .78 to .7 back to .74 then .6...until I accelerated for a long enough period to get a smooth transition to 1.

Ultimately it showed me what I want to see, so now I have to see if it means what I think and, if so, make it do what I want a bit better.

-I think I figured out what I was doing wrong regarding my use of the OAR, and it's dumb. After looking at my log and seeing extremely similar numbers I began digging around a bit more through the folders. As it turns out, I was basing my thought process that the Oct Timing Adjustment Max (Advance/Retard) tables were where the OAR was doing its math-not the BL Timing Comp. (OAR) table. I guess it helps to not second-guess the tuning tips...

As a check on learning, I have a new plan. I'm going to shut my OAR off at 0 again and map out new timing with the 87 octane. Once I have that established, I'll refill with 93 and do the same and that should give me the delta necessary to math out the adjustments to put into the BL Timing Comp (OAR) table. Once that's done, it should theoretically work. At least, that's what I think should make it work regarding the way I want it to be used. I noticed that at 1 load and above there is no OAR adjustment, and I would think that that would be one of the most critical areas to have it operational on a setup designed to run multiple fuels. Once I have that logged and math'ed out I should be good (I think).

So that's essentially where I'm at with OAR right now. I'm taking the FoST back to the dealership tomorrow to have them work on the hatch for the fourth time so it'll be a few days (or longer) before I have it back again. On the bright side, I should have ATR for the FiST in the next day or two so I'll probably start messing around with it instead.

 

[dyn085]

Well, I finally got ATR for the FiST loaded. I perused a few things and made random observations. I'll take a different path with the FiST because it's used less, but ultimately I'll still be shooting for a tune that runs a few different E10 octanes first. I took the 93 stage 3 and threw a ton of fuel at it for a start point and will take a few random logs to see where it stands.

This will easily be more fun than the FoST.

 

[dyn085]

Well, I've picked up about 9hp/9tq from timing based on the 93 stage 3 OTS with added fuel. It still wanted more timing but I ran out of gas and don't need it to be 'perfect'. I didn't get an actual baseline on the V201 map because I'm usually not logging accelerator pedal position, but this is a comparison from my first revision to last. What is most impressive to me is that I'm basically running MBT almost everywhere at 1 load and below, so my fuel mileage has skyrocketed. Part of that is due to having a leaner cruise, but the car has been wanting timing in heaps and spades. I went ahead and filled up with 87 and will turn my OAR back on to force it back down to 1 before shutting it off, and then I'll figure out what makes it happy with that fuel so that I can math out a different OAR setup for my BL tables. Once that's done my 'everyday' maps will be good for what I need and I'll probably start with a stock map and work up from there.

 

[dyn085]

The OAR is getting close to what I want.

In the shortest explanation, I am using an 8-degree separation between my low and high octanes. I raised all of the base tables by 3 degrees, set the OAR BL Timing Comp multiplier to -4 degrees everywhere, then adjusted my OAR Modifier to be aggressive in reduction and slow to advance. This gives my movement in both directions that keeps me less-aggressive than the numbers I arrived at through my logging. My KS is still set to only allow 3 degrees of timing addition, so mathematically I still end up where the OEM maps would with higher octane, it just starts off higher to begin with and won't remove as much timing for lower octanes.

While closer, it definitely still needs tweaking. I still have 87 in and once I re-activated the system I knew it would move towards the negative but I assumed it would still stop moving somewhere on the positive side of the scale. Not so much, as I'm currently sitting around -.31. The main problem is that in low-load driving the car doesn't really care what octane is in it, so it keeps wanting to move timing towards MBT. So I've gone back into the OAR adjustment and raised the minimum load for learning to see how much of an effect that has on the OAR movement. I didn't really want to do that given my experience with the FoST's OAR learning, but I'll see how much of an impact it makes with the current fuel.

 

[dyn085]

Has anybody else done any VE adjustments? I locked my cams for each HDFX pair yesterday on the FiST and made adjustments accordingly, and that took far longer than I expected. My FiST didn't really need hardly anything, it was more an exercise for practice than anything else. I'm moderately curious as to what changes may/may not be needed for someone with an intake.

The FoST finally came out of the shop today and I was getting ready to go take it through the process but my boy busted his chin open so I ended up at the hospital instead. I'm kinda glad because a system blew through and dropped the temps into the 70's, so hopefully it's back into the '90s tomorrow.

 

[dyn085]

I guess not, but I'll go ahead and list my pro-tips for those that might consider doing it. It's not 'necessary' from any of the logs I've seen, but anyone that considers doing their own tuning for a BT upgrade will definitely need to.

After doing this twice now, the main thing that I wish I would have done was created 15 separate tunes based on each HDFX pair before I even started. I was out on the side of the road with my laptop plugging in all of the changes during the process and that led to the data-collection taking a little more than two hours each time. Take your base tune, enable the cam locking, punch in the associated pair cam angles, then save it with whatever helps you remember-I simply put the logged pair HDFX number at the end.

Once you've created the 15 separate tunes (16 if you enable the OP tables), set up your log list with whatever you want, but the applicable HDFX weighted percentage and LTFT/STFT at a minimum. Nothing different than what the tuning guide says, but this is something you have to adjust through every datalog. Do a run, return to your starting point, load up the next tune, and change your HDFX weighted percentage accordingly. Repeat the process and don't forget to annotate the barometric pressure with the AP3.

The nice thing about logging directly to the laptop is that you can log all 15 pairs in the same session, but if you're logging to the AP3 you'll want to break it into two sessions because the AP3 can't hold more than ten logs from what I've seen. Once you've logged everything necessary you can now head back to the house and do the math in the air conditioning. Excel does the math for you, but it's ridiculously time-consuming so you'll want a comfortable chair.

On my computer I added a 'VE' file to each cars folders, and inside are 15 separate numbered folders to represent each pair. In those files I transfer the applicable HDFX weighted datalog(s) and keep a copy of the Cobb VE tuning workbook. On my desktop I keep a master copy of the VE workbook that is solely used for making changes to the HDFX References-but it's an extra step that most won't need/want. I open the datalog, highlight the areas that inform me of the changes necessary, open the master list from the desktop, make the appropriate changes, then I copy the changes from the master list and apply them to the tuning guide in its own folder to view how it maps out.

Don't forget to get a datalog with all 15 weighted percentages logged simultaneously as it will not only show you areas that need work but will also show you how far to take your pulls. I think you can also datalog the lower rpm areas by simply changing your idle rpm, but I haven't actually done that yet as it would be even more time-consuming...but I think it would work.

Criticisms/suggestions welcome.

 

[Dylanc]

I've been reading all of your posts, new to the whole tuning thing but not afraid to try. I've been wondering if it's possible to make a daily tune for better mpg, reliable.... but still have stage 3 bolt ons. Just a thought I've had

 

[dyn085]

For me the entire learning aspect of it is fun. I like datalogging and testing, evaluating, and then seeing what effect my adjustments have compared to the original data. I'm weird like that, but it feels like a good hobby to me because I'm not trying to shoot for overnight perfection.

The underlying principle of tuning is correcting and maximizing the engines overall efficiency. In doing that you actually improve it across the board-fuel mileage, power, and safety/reliability. You can take it beyond that point by sacrificing one or more aspects for one another, but that's not 'necessary' and in reality should be avoided.

Even with bolt-ons there is a lot of room for improvement, so if it's possibly something you're interested in you should consider investing in some tuning literature to help give you a general idea of the overall process. It's pretty intimidating at first, but if you break it down into the four main areas (VE, fuel, ignition, boost) it becomes a lot more manageable. Cobb's OTS tunes and tuning guide/materials are a pretty good start point for both tuning and grasping some of the main components/monitors/adjustments of each area of the tune.