Starting up - COBB Ecoboost Platform and Questions

Hello Everyone!

I am new around here, and will follow up with an intro in that section, but I wanted to toss out some specific questions that I have been struggling to get REAL answers on for a while - specific to COBB and the Ford Ecoboost platform (Focus ST in this case).

I took the COBB course in order to even be able to access their software, and while it was informative, it did not offer much in terms of specifics, this is obviously a hands on learn as you go type of intro course that you need to take before they let you loose with the software - so I stumbled across this amazing place!

Very excited to get to work here with these lessons and this community as my current tune is based on a COBB OTS map, with small modifications as I try to learn the ropes, in hopes of being able to fully tune myself when/if I upgrade to a larger turbo.

General Questions:

Fuel Scalar:

How I understand this is as follows, I would love to know if I am on the right track... This is basically telling the ECU what to expect 1 lambda to equate to on a AFR scale, referenced to 0% ethanol pump gas. So setting it lower when I run, say, E30, tells it that the same 1 lambda it is looking for at idle, etc is actually a bit more fuel, relative to having 0% eth content, and this goes the same for under load etc. This being a DI car there is no place for me to instead adjust fuel injector scaling or something of the like, so instead you tell it that it will need more fuel with a higher eth content using this.

If my long term fuel trims are elevated, is my only option to adjust this scalar? (perhaps meaning that while the value used is for E10, but maybe my pump gas only is E5, throwing the calculations off). I cant adjust a pulse width like I am used to seeing in this DI car :)

Load to TQ and TQ to Load tables

I was initially under the impression that I should be looking at fine tuning these tables, but when playing around with them all I got was a CEL and limp mode - so I stopped as I apparently did not understand how they worked. Looking at OTS maps I see they took the OE table and basically extended it out to higher load and TQ values - is that all there is to adjust on these tables?

I was looking to them to adjust the throttle blade angle vs the pedal and it looks like the better way to go about this would be to leave these alone and use the throttle requested TQ, which on this ECU seems to be where you adjust your pedal to actual throttle position requests. Am I on the right track with this thinking?

I am seeing on my tune the throttle plate move around a lot during WOT, likely to limit boost while I have it turned down right now, but referencing logs of a tune I actually paid for, the throttle will stay wide open regardless when the pedal is to the floor. Doing some reading on this, it sounds like its actually not the worst thing in the world for the throttle to be allowed to move - my instinct is telling me the WG should be doing this however, but I may just be thinking about this wrong.

Can you tune the injection window on this ECU? Or control pulse width at all like you can with the aftermarket options?

I watched the webinar last night on direct injection, and injection timing - first off, SUPER cool video. But on my ECU, poking around in the AccessTuner software, it looks like the only thing I can really define is how large the window is, and where to clip it at. Not really that important as I am guessing that since these are stock injectors, the OE timings are probably right about where I would want them.

Timing Advance Ramp Rate

COBB OTS maps ramp up in small increments, very quickly (when no knock is present), while the tune I have to reference that I paid for goes in larger (0.5 deg) increments and ramps more slowly. I would like to hear opinions on the pros and cons to these both, if anyone has something they would like to share :)

COBB Custom Features:

"Enable Switch (Allow CL Control In Adverse Conditions)

This setting will allow the closed loop fuel control (stft/ltft) to remain active if previously disallowed during cold and power demand states. If this setting is 0, the OEM logic will be used."

This setting I am having trouble understanding what it's intent is. In the off the shelf maps this is set to OFF. The car runs basically in 100% closed loop (as far as I can tell/know, except maybe cold start). As Andre has mentioned in a video I watched already he seems to not like the LTFT to be active under WOT conditions, only stft, which makes sense - but what is this switch doing? If its off does that mean there are no fuel trims in WOT conditions? Not sure how that would be true since its in closed loop under WOT (unlike the last platform I came from, Subaru). Leaving this off I see that trims appear to still be active under WOT conditions as well, so I am just confused.

Slightly Elevated Ethanol Content, and AFR:

Looking at the pump gas (E10) and E30 logs I have on a tune I paid for, the target AFRs under WOT are basically the same, 11.5:1, with only an E30 mix, is there benefit to moving this around as long as it seems to be working well?

Knock!!

Reading through blogs from the protuner I have maps from, knock is regarded as something you can never fully expect to be rid of, that keeping it at a low intensity, and random (not ALL cyl knocking etc) is 100% normal and fine, as long as overall, in a pull your timing advance stays trending positive. While I have not watched many of the course videos yet, a few of the intro ones that I watched before purchasing the classes were regarding knock as something you should really not being seeing. I would just like to clarify if my understanding is correct, that - especially on pump gas, it is normal to see, as long as the intensity is kept under control and its random, not consistent knock, that you are safe. I am thinking this is, because the stock tune knocks worse than even a COBB OTS map! :)

And lastly, just for fun... There is a well known (in the Ecoboost community) tuner called Stratified, out of Canada that does a 'crackles and pops' feature, while I am not actually wanting this myself, combing through the options I have in my software are the same as they have, and I can't think of an obvious way they are doing this - I am guessing they are adding fuel when you lift off the throttle to get that crackle/burbling noise, but i am more curious as to how that would even be done, given the set of tools we have to work with within AccessTuner.

Thanks!

The first thing you need to do is make an Excel (Google sheets, whatever) spreadsheet that compares the completely stock map (stage 0) with Cobb's stage 1 map. Look at what they did first. It sounds like you've already done this in a preliminary fashion but if you don't have an easy comparison you'll really confuse yourself. So in your Excel sheet you need:

1. Description of the table

2. Left side: Stage 0 setting (showing axis scaling)

3. Middle: Stage 1 setting (showing axis scaling)

3. Right side: Difference between the two using an Excel formula (Stage 1 - Stage 0). So if the spark timing increased by 1 degree, this table should show a 1.

Reflashing an OEM controller is so complicated that you have to be very organized in your versioning of your tune. You have to track everything carefully. Start with this exercise. Post your comparison spreadsheet and then we can go through your questions.

Hello Raymond,

I will get on that - I perhaps am a bit spoiled by the compare map function in the software which lets me flip back and forth between maps and also see differences. I notice that a lot of the scaling is very strange on the OE maps, I will post a screencap when I am back on that computer, but for example it will scale like (load in this example): 0.2, 0.4, 1.0, 2.0, 2.0, 2.0, 2.0, 2.0. Makes it a bit tough to direct compare when they scale in this fashion.

Regarding versioning, I note everything I do between saves in a versioning doc, and increment the version number on the files to keep track of what is what. I can see how easily you would get lost on what change did what action if you did not!

Thanks

Here is a good example of what I am talking about on the OE tables and their weird scaling

Hey Matt, I haven't personally tuned on the Ecoboost platform as yet but I've asked one of our friends from Cobb to chime in so hopefully he will have time to do so soon.

Awesome! Thanks Andre!

Raymond, I have been comparing to OEM tables, I am going to see if I can snag a Stg 0 map to compare to, perhaps that would have the OEM value but scaled correctly. If not then I am not sure that the excel method will yield good data.

You've got to get it all out in 1 place first. You can rescale by Excel on your own so you want to have it in there.

Thanks, in situations like those tables I posted above, where their scaling just goes right to one value and then repeats itself, do you just assume that you can make it scale linear? See OE-LTT.jpg

When I see a scaling like that it makes me think a couple possibilities:

1. The function isn't even used in the stock tune. It's a copy and paste job from another engine that's just turned off. So if the load is 2.0, and there's another scalar that only turns on that table for below 1.4 load, well it's not even used. Cobb may be turning on the function by changing the scaling because the stock scaling effectively disables it.

2. The function is there but there a bunch of modifiers applied to it so that the final calculated value actually makes sense.

You can make it linear but it helps to understand or at least speculate why the original one was scaled like that. The usual answer like I said are modifiers tables or deactivation. But to answer your question, I would start by copying Cobb's table and scaling.

Thanks Raymond, maybe this will become more clear once I move that all to Excel, I plan on working on that this week.

And @Andre, if you have not tuned an Ecoboost yet - I have an idea for a worked example module *wink wink!!* :)

In case others stumble across this at some point, I wanted to add a few answers I have gotten through chatting with Kyle over at COBB who walked through some of my questions with me - Thanks Kyle if you are on here too! :)

After some back and forth I summarized the points of our convo and pasted it below here as well - he confirmed that my understanding of these was correct. Slowly but surely picking this stuff up :)

Now...to get to work on my VE tables as well, I am sitting at +10% trims at cruise and idle....

- Enable Switch (Allow CL Control In Adverse Conditions) is not used on the current Ford Ecoboost applications, this is leftover from older revisions. Its purpose was to allow closed loop fuel trims to remain active if the wideband O2 sensor was in fault.

- Stoich Setpoint tells the ECU what lambda = 1.0 is of the fuel in the tank, in relation to E0 pump fuel

Monitors:

- Actual AFR = Lambda(Raw) * 14.7

- Actual AFR (SP) = Lambda(Raw) * Stoich Setpoint

All AFR (or Lambda) values entered into other target tables (Power demand/cold AFR desired, etc) are calculated with via lambda, since in the case of E10, lamba 1.0 = 14.02, when you put in a value of 14.7 into a desired AFR table it would still call for a REAL target of 14.02 since it is really requesting a target of 1.0 lambda.

How did you get on with this Matt?

I do a fair bit of Ecoboost tuning but dont want to spend a lot of time answering questions from an old post if your all done and dusted now.

If you still have any questions such as the above about LTT and TTL or knock control I may be able to help.

Hello Stu!

Things are going well! Regarding the TTL and LTT I am still unsure if I understand how to adjust those tables in any way other than scaling them up to allow actual loads you are achieving with higher power levels to be represented. If that is all there is to it, then fine, I did try and adjust them (together, with the changes in one being inverted on the other) but ended up getting a CEL and limp mode so obviously I didn't understand. I ended up using the throttle requested TQ tables to get what I wanted which was to adjust the load request based on throttle input. I was only attempting to adjust the LTT and TTL tables to more mirror reality but I do not fully understand them.

I am now pretty well dialed in (for road tuning at least) on 91oct and E30. Putting down (according to vdyno, flat road, getting consistent results in either direction) 270WHP/380WTQ on E30 which I am quite happy with.

I did recently upgrade to a catted downpipe and cat back exhaust, and was surprised to see that my WGDC didnt need any adjustment...I was expecting it to, but I still hit 100% WGDC by ~4100 RPM...smalllllll turbo...

Right now I continue to fiddle with the Speed Density / Map Vs Air Charge HDFX tables to dial in fuel trims. Mostly trying to smooth out the oscillations and get trims down to between 0 and -5% and making steady progress. I am +3% to -5% now and really should probably call it good but hey..

On knock control I think I have a good understanding, I have experimented with larger steps at slower intervals and smaller steps and quicker intervals of advance and currently am sticking with the smaller steps/faster intervals approach. I allow 2 knock events before swapping from the fast rate to the slow rate of spark advance, and have my knock response table (at least what I think) is pretty aggressive to protect the engine (pic below)

I run E30 most of the time now however and the only real downside I am experiencing is when I sent my oil in for analysis (Blackstone) they noted my viscosity was a bit low (4k miles on oil) so I am going to go with a 3k change interval and see how that goes.

Thanks for checkin in!

Morning Matt,

Typically I am eating breakfast and just about to head out of town to pick my little girl up so I cant fire up a huge response but it sounds to me like you all but done now anyway!

TTL and LTT (Torque To Load & Load To Torque for those unfamiliar with / Ford Bosch control logic) is a crazy concept to get your head around as its OE only, and it doesnt help that you have 32 of the damn things to do (16 sets of tables for each matched camshaft position). Think yourself lucky though, this years ECU's have 64! In essense. the only way to get it right is to run the vehicle up on the dyno, choose your RPM and load and then dial in the correct NM figure at that RPM / Load site for the extra load you have input.

For example:

Lets say you want 700NM at 4000 rpm and the torque axis only goes to 600NM with a load of 2.5. That means if you request 700NM in your limiter tables, the ecu will refer to the table and ask "Hey, ive got a torque request here, what piston crown load is needed for 700NM at 4000rpm?"

When it sees the highest figure is 600NM, thats the load you get, and no more.

So, you have two options here:

1) Lie about the load required to make 600NM and poke in more load on that cell so it really makes 700NM. (Sadly - most do this)

2) Rescale the table leaving the factory numbers intact up to 600 NM so all the factory model is accurate still, and add another row to cover 700NM.

You can now add some higher load figures along that row and see what happens on they dyno.

Thos load numbers you input will need to be refined over tens of runs, but you will eventually get a load curve that gives you the torque you need.

Now doing that for every cam point is a real headache as you can imagine.

What you can do though, is configure the engine to use the OP table and do just that one for starters, and then you can copy and paste what you have learnt into the main tables and adjust them as you go to get them right when your back using all 16 mapped points for the VVT.

This is all assuming you have use of a dyno.

Alternatively, with no dyno you are going to have to guess the load required to make X NM and see how it goes on the road.

Please Note:

If you havent exceeded factory LTT scaling yet you may not ever have to touch those tables at all.

I will try and remember to pop back later and explain how I dial in the inverse TTL to match the LTT.

I have written some EcoBoost blogs that may help you understand the OE logic a little better, but i have no idea how weblink policies work here yet and I really dont want anyone to think I am here to advertise, I am not.

Speak later mate.

The whole concept is based on using models to control the engine rather than dumb look up tables. At the OEM level that is all tuned on an engine dyno running 2 or 3 shifts a day with multiple engineers and technicians tuning those maps with special OEM tuning tools.

Thanks guys, makes sense. I will not be exceeding the scaling already in there for some time, would need a larger turbo and thats not in the cards at the moment.

Stu,

Shoot me an email with those articles please, would love to take a look: mwilliams252@gmail.com

Another thing I finally found some information on was the 'blowthrough' tables in the speed density tables - apparently this is to account for valve overlap introducing fresh air through the cyl and into the exh stream. There is no information on how to adjust these and most of them are flat values across the board. I notice that some etuners (based on logs) have it disabled. I have tried both to see what changes and I cant notice a difference in leaving the logic enabled or turning it off. When I log and show when these are referenced its over a very short stretch of RPMs right about Pk TQ so I have left this on and the tables default, doing my adjustments based on fuel trims in the 'standard' MAP vs Air Charge HDFX tables. I assume the ECU will reference these tables when it knows there is overlap going on introducing enough air through the cyl to throw off the measured AFR.

Any further information you have to add on why you may want to disable that logic as I see some tuners doing?

Thanks

Morning Matt,

I have dropped you a mail linking to my Ecoboost strategy Blogs.

Good to hear you dont need to get involved in recalibrating the LTT and TTL systems, thats going to save some tears! :)

Ref Blowthrough.

I have certainly noticed COBB disable Blowthrough strategy but I cant advise why they have done it.

That said, I recall COBB have chosen to move over to their new simplified SD CCF2 feature for all their FRS3 tunes now which gives us a nice big VE table to tune the fuelling on, so that may well be why blowthrough is now disabled if they are doing the same on your USA ST's?

I personally leave as much of the OE logic as possible intact. I really try to learn about, understand and work with these modern systems, not dumb them down and head backwards in time to dump tables if I can at all help it.

Thanks - I did actually swap over to the CCF SD system to test it out, but Raymond actually had similar comments, why go backwards in time? Learn the new stuff - so I swapped back to the OE system and work with it :)

The Etuners logs I have looked at still use the OE SD strategy, but some do disable that BT strategy - no idea why. I ended up leaving it on...I would need to log actual fuel injector/fuel mass deliveries to see if it actually changes anything I would thing. Watching just target vs actual AFR I see no meaningful change.

I replied to your message, very nice stuff :)

The OE system works best IMO. I find the COBB strategies just dont drive as nice.

They make decent gains, sure, but the driver experience just isnt right, its like stepping backwards everywhere bar WOT.

This is GOOD news though, as we would never get commissioned to do custom tunes if teh free ones were exceptional! LOL