COBB AP/ATR Beta Fuel Scaling

[Alpha]

Originally Posted by phate We are already running our injectors >100% with e85 and seeing no ill effect. I regularly see 110%, and as high as 120% in my car. Bozo's car has seen as high as ~145% without issue.

Doesn't this negatively impact reliablility and increase the likelyhood of dangerous fuel dips?

Would the new DW intank pump help alleviate those issues past 100% IDC?

[phate]

I don't think anyone really knows the answer to that question. 100% is a calculated figure...but what does 100% represent?

My car runs fine seeing that high of IDC, so I'm not terribly concerned about it unless someone can show me evidence of it being harmful.

[cld12pk2go]

Originally Posted by David@COBB Doing some research on Injector startup dwell and Injector phasing (Branching off of djuosnteisn's findings) with the oscilloscope on the dyno. I want to make sure we are utilizing the full potential of the injectors, as 100% E85 on our beta scalars ran us out of injector. If phasing and startup point can get us a bit more time, we might be able to run 100% with stock injectors. -David@COBB

So is the fuel scaling not included in the ATR update that was released to marketing late last week?

[Enki]

The IDC % doesn't seem to change with variances in VVT and/or timing, so we aren't even sure it's an accurate number to go off of in the first place.

[Erich]

Originally Posted by Dano Step 1: Clone yourself [rinse repeat as necessary] Step 2: Expose all tables and fix everything in one week Glad you're on our side @David@COBB

We should probably steer away from clones and stick with the original where ATR is concerned.

[Cobb Tuning]

Okay guys, it's time to round up everything, as I'm pretty much done with researching the fuel scalar and what it takes to run E85.

Looking at the data djuosnteisn compiled, along with other datalogs and tests, I'm confident we have the controls to run E85 mixes in the MS3 with minimal Map changes. If you look at the data djuosnteisn has provided, we can see that the injector opening is getting dangerously close to the allowed injection window before the intake cycle.

In essence, we've found that Mazda has put in relatively small injectors to get the resolution they need for low load and RPM, and then use the DI system to ramp up to the level needed during high load and boost. Changing the engine's way that it calculates how much fuel it needs at the fundamental level, as well as injector size, allow us to run the system with alternative fuels without too much effort. Running 50% or 75% mixes are easy to do with the current controls I've been using.

That being said, I'd like to hear from the people who were in the beta to chime in on if the changes I made to the maps, and if they were happy with the results - and if not, what other things would they like to see.

-David@COBB

[rfinkle2]

Originally Posted by David@COBB Okay guys, it's time to round up everything, as I'm pretty much done with researching the fuel scalar and what it takes to run E85. Looking at the data djuosnteisn compiled, along with other datalogs and tests, I'm confident we have the controls to run E85 mixes in the MS3 with minimal Map changes. If you look at the data djuosnteisn has provided, we can see that the injector opening is getting dangerously close to the allowed injection window before the intake cycle. In essence, we've found that Mazda has put in relatively small injectors to get the resolution they need for low load and RPM, and then use the DI system to ramp up to the level needed during high load and boost. Changing the engine's way that it calculates how much fuel it needs at the fundamental level, as well as injector size, allow us to run the system with alternative fuels without too much effort. Running 50% or 75% mixes are easy to do with the current controls I've been using. That being said, I'd like to hear from the people who were in the beta to chime in on if the changes I made to the maps, and if they were happy with the results - and if not, what other things would they like to see. -David@COBB

My 50/50 map ran very well.

It is infinitely easier to create a good working map using the correct calculated loads and other variables than fudging the maf curve.

My apologies for not being able to provide more data, but I drive < 3000 miles a year.

[superskaterxes]

Originally Posted by David@COBB Okay guys, it's time to round up everything, as I'm pretty much done with researching the fuel scalar and what it takes to run E85. Looking at the data djuosnteisn compiled, along with other datalogs and tests, I'm confident we have the controls to run E85 mixes in the MS3 with minimal Map changes. If you look at the data djuosnteisn has provided, we can see that the injector opening is getting dangerously close to the allowed injection window before the intake cycle. In essence, we've found that Mazda has put in relatively small injectors to get the resolution they need for low load and RPM, and then use the DI system to ramp up to the level needed during high load and boost. Changing the engine's way that it calculates how much fuel it needs at the fundamental level, as well as injector size, allow us to run the system with alternative fuels without too much effort. Running 50% or 75% mixes are easy to do with the current controls I've been using. That being said, I'd like to hear from the people who were in the beta to chime in on if the changes I made to the maps, and if they were happy with the results - and if not, what other things would they like to see. -David@COBB

wait, are you saying that your new controls let you run 50% without touching 100% IDC? is this on BT setups or just K04 guys?

basically have you found a way to stretch the stock fuel system even more?

[Cobb Tuning]

Originally Posted by superskaterxes wait, are you saying that your new controls let you run 50% without touching 100% IDC? is this on BT setups or just K04 guys? basically have you found a way to stretch the stock fuel system even more?

IDC is a calculated value based upon a lot of parameters. When you are forcefully changing MAF, you'll see crazy high IDC's as load is part of that calculation.

Based upon our research, you can run up to 75% E85 mixed with 25% 93 Octane, and not touch 100% IDC. This was tested on a stage2 car with all the bolt-ons and a FMIC.

-David@COBB

[Enki]

So, new maps in the works?

Edit: New maps as in new beta maps; or are you going to release the new modifiers with information so we can just update our existing maps?

[Dano]

Originally Posted by David@COBB IDC is a calculated value based upon a lot of parameters. When you are forcefully changing MAF, you'll see crazy high IDC's as load is part of that calculation. Based upon our research, you can run up to 75% E85 mixed with 25% 93 Octane, and not touch 100% IDC. This was tested on a stage2 car with all the bolt-ons and a FMIC. -David@COBB

sounds like you need a BT car for testing...I can test but I no haz BT car...not really Anthony? do you have corn in your neck of the woods?

edit: does it take 100% IDC to reach the max optimal fueling window or much less like 80%? @David@COBB

[superskaterxes]

Originally Posted by Dano sounds like you need a BT car for testing...I can test but I no haz BT car...not really Anthony? do you have corn in your neck of the woods?

im running ~30% E85 and 13.0 AFR with 1200cc of meth to stay under 100% IDC. my maf curve also is not adjusted at all for E85 other then in CL regions. i use the meth to compensate for the increase in the E85.

when my car hits 100% its like a monsoon in my cylinder and i get extreme blowout and the car pops and fires and goes nuts. its only when i hit 100% exactly on the dot. i have millions of logs if u need to see anything.


this is the whole reason i went 5th injector with dustin so that i wasent relying so heavily on meth as extra fuel. turns out the 1000cc injector i bought was useless as its worse then the meth. just picked up a 2200 that i will hopefully have by saturday =D

[Cobb Tuning]

Originally Posted by Enki So, new maps in the works? Edit: New maps as in new beta maps; or are you going to release the new modifiers with information so we can just update our existing maps?

No more maps, we will be release the controls instead.

Originally Posted by Dano sounds like you need a BT car for testing...I can test but I no haz BT car...not really Anthony? do you have corn in your neck of the woods? edit: does it take 100% IDC to reach the max optimal fueling window or much less like 80%? @David@COBB

It doesn't matter how your car is boost/load tuned....

I'm not sure what you mean by "optimal fueling window". Can you clarify a bit more?

-David@COBB

[Enki]

He's talking about in reference to port injection, where you don't normally want to exceed 80% IDC.

Edit:

Originally Posted by David@COBB No more maps, we will be release the controls instead.

2 weeks?

[cld12pk2go]

Soooo ready to get this control as I am still having the throttle close up top with the latest ATR release. I suspect that having my loads shift down 10% will greatly help with this on my higher gear pulls.

It should also allow me to actually tune 3rd gear for decent power with traction. I am currently using DBW tables to limit load, but I cannot go over the 1.94 load row without screwing up WOT in higher gears that target > 2.00.



Let me know if you need another BT Beta tester...

[Dano]

Originally Posted by David@COBB No more maps, we will be release the controls instead. It doesn't matter how your car is boost/load tuned.... I'm not sure what you mean by "optimal fueling window". Can you clarify a bit more? -David@COBB

yes, so if the injectors are firing longer than optimal to keep fueling correct, atomization suffers and or blowout is more likely to occur. Some have speculated that somewhere around 85% is the limit of IDC before things degrade.

oh and by BT I mean big turbo...lol more air needs more fuel

[rfinkle2]

Originally Posted by Dano yes, so if the injectors are firing longer than optimal to keep fueling correct, atomization suffers and or blowout is more likely to occur. Some have speculated that somewhere around 85% is the limit of IDC before things degrade. oh and by BT I mean big turbo...lol more air needs more fuel

This is interesting because I hear both sides of the story.

I agree some people feel that ~85% is optimal, but Phate was running >100 IDC's and reported the car ran well.

[Dano]

Originally Posted by Enki He's talking about in reference to port injection, where you don't normally want to exceed 80% IDC.

yes and no...lol I thought it was actually worse with DI.

Originally Posted by rfinkle2 This is interesting because I hear both sides of the story. I agree some people feel that ~85% is optimal, but Phate was running >100 IDC's and reported the car ran well.

yes same here thus the reason I am attempting to get confirmation from The Man.

the engineers here @Ziggo have calculated the time window needed to complete the injection event...IDK what it is so I asked

I think Dustin also worked with an Oscope to calculate said window...

[Cobb Tuning]

Originally Posted by Enki He's talking about in reference to port injection, where you don't normally want to exceed 80% IDC. Edit: 2 weeks?

I'm bundling some other useful features up with it

As for the "fueling window" with IDC - unless all your calibration is correct for MAF/Fuel, etc, it'll be hard to trust the ECU's calculation of IDC.

Looking at the ECU's output with an oscope is a bit more telling. With Direct injection, if you continue the spray longer than normal (closer to the spark), it won't necessarily atomize correctly, and you'd have diminishing returns.

So there will be less gain by running over 75% mix E85, as you'll be pushing the limits of the stock fuel system and injectors. So staying below that is the key here (unless you get bigger injectors).

So to answer your question, optimal fueling is less than 75%-80% E85 mixed with 93. That rides the line between atomizing well, as well as measurable power gains vs. other mixes. You'd be surprised how a 50/50 mix compares to race gas

-David@COBB

[Cobb Tuning]

Originally Posted by rfinkle2 This is interesting because I hear both sides of the story. I agree some people feel that ~85% is optimal, but Phate was running >100 IDC's and reported the car ran well.

As an extension of my post above, as long as your fuel scalars, load, MAF, and other things are calibrated correctly, we've seen the car run up to 100-110% IDC with no odd effects.

-David@COBB

[Dano]

thanks for the clarification David.

[superskaterxes]

Originally Posted by David@COBB I'm bundling some other useful features up with it As for the "fueling window" with IDC - unless all your calibration is correct for MAF/Fuel, etc, it'll be hard to trust the ECU's calculation of IDC. Looking at the ECU's output with an oscope is a bit more telling. With Direct injection, if you continue the spray longer than normal (closer to the spark), it won't necessarily atomize correctly, and you'd have diminishing returns. So there will be less gain by running over 75% mix E85, as you'll be pushing the limits of the stock fuel system and injectors. So staying below that is the key here (unless you get bigger injectors). So to answer your question, optimal fueling is less than 75%-80% E85 mixed with 93. That rides the line between atomizing well, as well as measurable power gains vs. other mixes. You'd be surprised how a 50/50 mix compares to race gas -David@COBB

so wait, your telling me with this new scalar we have enough fuel for 450-500whp?

[djuosnteisn]

Originally Posted by David@COBB Looking at the ECU's output with an oscope is a bit more telling.

David, we're you guys able to determine if, indeed, the injector start time was opening earlier and earlier (even before the intake valves open)?

Did the experiments shed any light on the possibility of inj phasing?

[mrmonk7663]

When will it be released??? Is it possible to also make a table in excel or something that can calculate the load necessary to have dialed into the map in conjunction with the fuel. The Throttle requested load ABC tables affect the map a large deal, even when boost tuning, and with e85, they have to be adjusted a great deal, and it is hard to properly adjust it because there is nothing to really go off of in the lower load areas.

[Ziggo]

See this thread for more info:

Logged Injector pulsewidth is incorrect

I found that IDC as calculated by the ECU\logged by the AP is exactly the logged pulsewidth (which is incorrect) divided by the time to complete 360* of rotation. The real IDC has to be compensated by for fuel pressure and cylinder pressure. That I would expect to be a very complicated calculation.

I also have a chart there in the first post that addresses an estimation of the IDC we log vs the effective IDC based on spark advance, VVT, and the SAE specified 3ms atomization time (which might be a conservative number when considering the higher temperatures and densities of a turbo'ed charge) the real key is that because the atomization time is a fixed time interval, the desired maximum logged IDC will probably go down as RPM goes up.

Thats one of the thing I asked DJ to pin down using supplimental fueling.

[Cobb Tuning]

Originally Posted by superskaterxes so wait, your telling me with this new scalar we have enough fuel for 450-500whp?

Depends on your modification level and fuel being ran. If some of the companies will hurry up and make injectors, I'm sure we could match a very large turbo's fuel requirements now that there will be a control for it.

Originally Posted by djuosnteisn David, we're you guys able to determine if, indeed, the injector start time was opening earlier and earlier (even before the intake valves open)? Did the experiments shed any light on the possibility of inj phasing?

We'll have to return to it on a later date.

-David@COBB

[Enki]

Originally Posted by David@COBB Depends on your modification level and fuel being ran. If some of the companies will hurry up and make injectors, I'm sure we could match a very large turbo's fuel requirements now that there will be a control for it.

Assuming, of course, that aftermarket HPFP can keep up with the demand as well. From what I saw in Phate's logs, he was at the edge of what the pump could handle on stock turbo.

[rfinkle2]

Originally Posted by mrmonk7663 When will it be released??? Is it possible to also make a table in excel or something that can calculate the load necessary to have dialed into the map in conjunction with the fuel. The Throttle requested load ABC tables affect the map a large deal, even when boost tuning, and with e85, they have to be adjusted a great deal, and it is hard to properly adjust it because there is nothing to really go off of in the lower load areas.

That is the beauty of the scaling.

Calculated loads will be accurate vs. shifting everything up from maf curve manipulation.

[silvapain]

My beta map runs great on 100% E85 with IDCs < 100% at WOT.


Tapadatass

[mrmonk7663]

Yes, I understand that for sure. But how will you effectively set the throttle load abc tables...there are alot of cells to populate. Its easy to log load at full throttle, but logging load at all throttle positions and rpms to dial in the table is almost impossible. You would think a formula could be made that based off of max load, load at idle, and some other factor that the cells coud be autopopulated via an excel speadsheet or something.

Originally Posted by rfinkle2 That is the beauty of the scaling. Calculated loads will be accurate vs. shifting everything up from maf curve manipulation.

[Enki]

You could probably start by multiplying the table by the same scalar you used for the MAF table.

[mrmonk7663]

Hmm, I might try that when the new revisions are released. I worked a shitload on my load tables with E85, and it was a pain in the ass to get to where I wanted it. It was good, I thought I could make it better...and in some regards I did, but at the sacrifice of driveability. I saw large mpg changes depending on how the tables were set up, and I saw massive changes in throttle respons and when/how boost came on, by solely adjusting the 3 throttle req load abc tables.

[rfinkle2]

Originally Posted by mrmonk7663 Hmm, I might try that when the new revisions are released. I worked a shitload on my load tables with E85, and it was a pain in the ass to get to where I wanted it. It was good, I thought I could make it better...and in some regards I did, but at the sacrifice of driveability. I saw large mpg changes depending on how the tables were set up, and I saw massive changes in throttle respons and when/how boost came on, by solely adjusting the 3 throttle req load abc tables.

Enki has a great idea, and I have some ideas as well. When the controls are released, I'm sure there will be a thread about it and I'll let you know what I'm going to try. I'm sure the other guys who have some e85 experience will chime in as well.

[silvapain]

My beta map is just my 93 map with the scaling, plus a slight MAF adjustment in the high voltage range because E85 adjustment isn't linear.

Oh, and increased timing and load targets because E85 is the shit.


Tapadatass

[RichieRichness]

When tuning @Schmitty5's car I had to add 5% to the entire curve for a 3/9 gallon e85 mix. When we went 50/50 it was about additinal 5%.

[mrmonk7663]

Yes, to the maf curve I had to do about the same.

Originally Posted by RichieRichness When tuning @Schmitty5's car I had to add 5% to the entire curve for a 3/9 gallon e85 mix. When we went 50/50 it was about additinal 5%.

[rfinkle2]

Originally Posted by RichieRichness When tuning @Schmitty5's car I had to add 5% to the entire curve for a 3/9 gallon e85 mix. When we went 50/50 it was about additinal 5%.

Originally Posted by mrmonk7663 Yes, to the maf curve I had to do about the same.

I think the e85 control is going to be set up so if you have a 93 octane map that is on point, you will basically be able to input your mix, and the fueling will be scaled for you. (the back office stuff David is working on will change the fueling so you don't have to shift the curve by artificially calling for higher maf readings).

[Enki]

On the topic of throttle/load adjustment:
Last month I had to go home from the track and dial the tune WAY back because the car was aggressive/part throttle un-drivable; coming out of a turn at mid to high rpms, tipping in the throttle a bit would just blow the tires off at full boost. I guessed at a rescale for the TRL tables and drove around the rest (note that it drove fine either WOT or regular city driving).

Today for lunch I took a stage appropriate OTS map and dropped a scaled up version of it in the table, and the result is MUCH easier to drive. Not perfect, but much, much better.

[mrmonk7663]

Enki, I did the same thing 2 nights ago. I used the stage 2 map, and scaled the load tables, which yielded much better driveability than I previously had. The e85 release will be cool but it will force all of us on e85 to retune the car over again grr.

[silvapain]

Originally Posted by mrmonk7663 Enki, I did the same thing 2 nights ago. I used the stage 2 map, and scaled the load tables, which yielded much better driveability than I previously had. The e85 release will be cool but it will force all of us on e85 to retune the car over again grr.

It's not that bad.

Like I said, I have a beta map with the scaling. I sent @David@COBB my last 93 map, and he put in the scaling. Then I took my MAF-scaled tune and scaled down the MAF calibration by the amount David adjusted my beta map. I also reverse-load-adjusted my load-based tables from my E85 tune via an Excel file I created, then threw those into the beta map. That way my increased timing and load targets would be in the new map and correct for the MAF.

Took me an hour or so, and the tune worked great. Matter of fact, when I went back on 93 gas, I took that beta MAF calibration and put it on my pump gas tune. Worked great with no adjustment necessary.


Tapadatass