trying to slope 160 # injectors twin turbo mustang GT 2V

I posted this on E85 forums as well as several others with no feedback so I am trying here.

I am trying to tune my twin turbo 01' GT with 160 # injectors. I am also running this on E85. Running a SCT BA2800 MAF with a diablo Mafia extender.

Tuning with an xcal 3 and advantage 3 software. my goal is to tune for 20 lbs of boost and plan to run as high as 30 lbs.

My tuner says we have to get the values for the low slope and high slope as well as the breakpoint.

Can someone please help me get started with this? I didn't get any help from the tuning SCT manual and SCT will not provide tech support on these injectors to me?

Is there a formula to get started? Also what else am I going to need to change? Any required reading? There are guys that will supposedly give this info but they want money (which is ok I am not expecting anything for free) but I want to learn how this works and how to do it myself rather than someone just giving me the numbers.

Here are the injectors I am using the Bosch 160s.
http://www.lightningforceperformance...oducts_id=1095

Thanks for any advice.

Sean

Sean, its Damon here. I was doing some googling on the same subject (i really want to run e85 as well) and ran across your post, which revealed something previoulsy unknown to me - the MAFia!!!

there was never an adjustment made to the MAF transfer function for hte MAFia.

PM me your phone number.

more info

Ok so a friend of mine has found some info and we have put a plan in place to get this done. Please let me know if all of this makes sense:


The site he found is here:

BjpRace.se

Statiskt flöde vid 2.7bar (39.15psi) = 1680 cc/min / 160.0 lb/hr OBS vid 2.7Bar
Statiskt flöde vid 3bar (43.5psi) = 1770 cc/min / 95.0 lb/hr / 12.0 gm/sec
Statiskt flöde vid 3bar (43.5psi) = 998 cc/min / 95.0 lb/hr / 12.0 gm/sec
Dynamiskt flöde vid 3bar (43.5psi) 2.5ms pulse width and 10ms repetition rate = 2.2 gm/pulse
Resistans: 5 Ohm .Många motorstyrning klarar köra dessa som både hög eller lågohm.
Kontakt: Standard Bosch EV1
Low injector slope: 0,044912 lb/s
High injector slope: 0,035086 lb/s
Minsta repiterbara pulsvidd vid 2.7bar: 1.642ms
Injector voltage offset as function of batt voltage:
6V=2.26ms 8V=1.32ms 10V=0.82ms 12V=0.5ms 13V=0.38ms 14V=0.46ms 15V=0.16ms

****There is no break point data though?

Here is the plan he put together please let me know if all this makes sense.

Most of the e85 stuff I have read says to change the injector data (slopes), rather than the MAF curve or speed density stuff when tuning for e85, this is to get the rest of the EEC calculators to work right. With gasoline we assume we have good flow data on injectors, and therefore assume that any errors in deliver A/F ratios are MAF errors. With e85 we know up front that Stoich is different (9.7 vs 14.7) so we might as well fixed what is "wrong" first, which is the "gas equivalent energy flow rate" of the injectors, which in theory is determined by dividing 9.7/14.7. The tuners that share info on the internet claim the difference is really more like 42%, and some even claim that it is only 20% at idle and closer to 40% at WOT so expect to do some tweaking.

after several hours of research, if it were me, here is what I would do;

"1. get the MAF curve right - this requires draining the e85 and filling up with 91. put 91 in the tank, 160 injectors in the rails. you will have to purge the lines and rails also possibly remove the MAFia, although if you anticipate pegging the BA2800 MAF, and you are tuning for gas with known injector slopes, you might be able to leave it in. load the stock tune in advantage, then the SCT 91 octane blower base tune, then put in the injector data above, MAF data from the BA2800 value file, crank fuel reduced to 25% of stock to avoid crank flooding. start the car and tune as normal, you need a 91 octane tune anyways, in case you run out of e85, so this is not a waste of time. it will give you a known good MAF curve so you can focus on injector slope changes when you go to tune e85 later.

2. assuming you got the car to run right on 91 with the 160's drain the rest of the 91 from the tank, fill up with e85, run the car for a minute (until you see it go lean or the CEL comes on) to get the 91 out of the rails/injectors. whack 34% off of the injector slope data (multiply by .66). This requires more research. Not sure if you also need to change the voltage tables and breakpoint too, but I would start with just the low and high slopes. Now that we know he car runs on 91 the trial and error on the injectors for e85 will be easier. don't mess with the MAF curve yet, it has already been verified on gas. start the car, tweak the injector slopes up or down for lamda=1 at ~2000-3000 rpm (midrange), at least one person reported that idle is not a good place to start with e85, you will be too lean up higher if you do.
if you can get good short term fuel trims at 2000-3000 rpm by adjusting the slopes, time to go back to MAF compensation, adjust the MAF curve for good trims at idle, higher rpm/load"


So my questions are as follows:
1. What are the target Fuel trims at idle as well as cruising, WOT, high RPM no load?
2. What is the target AF for E85 at WOT?
3. I also am running a diablo Mafia for when I peg the BA2800 (which has already been done before) for initial tuning?
4. What is the relationship between slope and lean/ rich conditions? For instance if I am lean at idle would I increase or decrease the low slope?

Thanks
Sean

1 - Target fuel trims are 1.0, just like gasoline. The stoich point IS different for E85, but the 02 sensors will read stoich as lambda 1.0 no matter what fuel is used because they are analyzing the exhaust gases for byproducts of combustion and the stoich point is the theoretical point where all the fuel is consumed. Lambda 1.0 will be the same, no matter the fuel used with a wideband. The A/F reading with a wideband is actually a calculation based on lambda and the stoich point of gasoline. But when you use Lambda, you are using the native language of the wideband. And like I said, the 02 sensor in a wideband doesn't care what fuel is used, it'll read lambda 1.0 as the stoich point of the fuel. Lambda 1.0 for idle and cruise and on a blower or turbo, just run the 'lambda equivalent' A/F, which if for gas you are targeting around 11.5 A/F, the the Lambda you are looking for would be 0.78ish.

2 - Don't use A/F for WOT unless you have a wideband that will allow adjustments for the fuel used - use Lambda. See #1 above.

3 - Not sure what uyou are asking, but you can run a MAF.ia with a BA 2800, no problem, just be sure to change the tune.

4 - Myself, I'd never mess with injector slopes because doing this will throw calculated load off. I'd get the MAF transfer correct, then go from there. For those huge injectors, I'd either contact the manufacturer or vendor and see what they suggest or just scale up form know good values, like from 60# injectors and take it from there.

I have tuned quite a number of cars with E85 and they are not a problem at all. I don't think you have to go through all that trouble of draining the tank, refilling etc. Just not needed. Tune it like I described above and you'll be fine.

The other reason to use Lambda and not make a separate tune for JUST E85 is E85 is rarely E85 - it varies seasonally. Around here -Ohio - You get E85 rarely, usually E60, but it just does vary. If you really want to know the ethanol content of a fuel, exactly, you need an alcohol content analyzer. Zeitronix sells one that works really well.

Hope this helps,

Don

The manufacturing isn't going to be able to give me anything I don't think. I have the slopes I found on a german website but I don't have the breakpoint? Don't know where to start for this?

I am tuning with a wideband.

Can you please tell me what the process is for scaling for known values? I have a value file for the ford racing 80 lb injetors I could use this? But the slopes aren't linear are they I can't just double them for 160s.

Thanks for your input.

If you have the slopes, you are halfway there. I'd just use Excel to plot the breakpoints of all the know injector values and go from there - you should be able to add an estimated point or do some 'what if' calculations and see what the curve looks like. I'd be inclined to use something like 0.00007973 and see how it works.

You will have to get your raw values established then scale slopes, breakpoint, manifold volume, MAF transfer and displacement in half. Then scale it all back for the MAF.ia too. If you don't do the initial scaling, you may not be able to enter values that large in the software.

Maybe an easier option is an AEM....

Ok still a little confused on the scaling. So the computer can only take values to a certain point correct? So effectively what you are doing is tricking the computer into thinking the motor is half the size with have the slopes, etc?

The mafia is already in right now? Should I tune it without it first and then when I start pegging the MAF put it back on?

Yes an AEM would probably be a better option but can't invest in that right now.

I found the injector slope straight from Ford so I used those which I have also attached for you to look at. The slopes and injector minimum pulse width were larger than the maximum values in Advantage III so I scaled by 50%. I scaled high slope, low slope, injector minimum pulse width, breakpoint, manifold volume, maf transfer function and engine displacement. The car would run but barely, it was extremely rich around 11.5 at idle. Short term fuel trims were between 5-15 and voltage at 980 through a snap on solus scanner. After the initial scaling we started pulling fuel out of the MAF transfer curve, we also tried to adjust the low and high slopes. Nothing worked we couldn’t get it to lean out.

Also, we were told to put all of the multiplier values to 1 as it’s a return fuel system.

Fuel injector values before scaling:
Low slope 0.044912
High slope 0.035086
Breakpoint 0.0002751948
Min. pulse width .00642

Fuel injector values after scaling by 50%:
Low slope 0.022456
High slope 0.017543
Breakpoint 0.0001375974
Min. pulse width 0.00321

After numerous attempts at pulling fuel through the MAF transfer we added 10% to the slope values (multiplying values by 1.1) again nothing worked.

Thanks for any input you may be able to give.

Can anyone please walk me through the scaling process just to make sure I'm doing it correctly?

I cut the slopes, breakpoints, and engine displacement in half. This is all that it should take to get it right isn't it?

I was so ridiculously fat I tried everything, pulled tons of fuel out of the MAF transfer, I tried to increase the slopes again to the max values even after the scaling and nothing would get it to lean out. It was below 10 on AFR.

Any thoughts? I was thinking that maybe the min. pulsewidth is too large or maybe the injector offset table needs to be scaled down as well because it's still adding the same amount of fuel to the scaled slopes from the offset table?

Sean, what you are trying to do won't work. The problem is the injectors. First of all, you must understand that the 160# Ford Racing injectors are low impedence. This means they are peak and hold. While the factory high impedence injectors get a nice even 12v from the ECU, they initially flow more than the rated amount of fuel. Then, pretty quickly (at the breakpoint), they return to the normal rated fuel flow. Imagine a garden hose that you've got your thumb over. You release your thumb, and the water shoots 10 feet, then pretty quickly settles back down to only shooting 5 feet once that initial pressure bled off. This is how factory injectors work. You already probably found the factory Ford specs for the injector...

http://www.fordracingparts.com/parts/ics/m-9593-e303.pdf

Normally, the high slope is # of fuel per second. So if you multiply this times 3600 (60 seconds x 60 minutes = 1 hour), you would get the fuel injector size in lbs/hr. Take a 19# injector for example. With a high slope of .005445, times 3600, you get 19.6 #/hr. That makes sense.

Now Ford still rates their 160# injector the same ways and all. But how they are actually applied in a car is different, because it's a low impedence injector. Ford basically nails the injector up front with a bunch of voltage, to open the injector up quickly, then immediately backs off the voltage bigtime, keeping just enough voltage to hold the injector open at the desired flow rate. Thus the peak and hold terminology. Now watch what happens when you try to use this injector as if it were a high impedence injector: Take the hi slope of .035086 and multiply it times 3600 = 126.3#/hr. Sounds lower right? But remember there's not nearly as much lag time, so that injector would start flowing faster. In the end, it's going to flow 160#/hr at 39.15 psi like Ford said. But how it got there is a different story.

Your REAL big concern here is injector drivers. If you run the stock ECU with these injectors, you WILL burn up the ECU's injector drivers, because they are not designed for the extra power that peak and hold injectors require. You need to look into an aftermarket injector driver box.

Next, you need to understand what Don was talking about with lambda and stoich values. Think of Lambda as the percentage of oxygen that's being used. For example, .8 lambda on gasoline (a 14.7 stoich) would = 11.76 air/fuel ratio. But a .8 lambda running E85 (a 9.85 stoich) would = 7.88 air/fuel ratio. Tune in lambda and don't worry about air/fuel ratios! The O2 sensor sees lambda, and then multiplies that lambda number times the preprogrammed stoich value to come up with an air/fuel number anyway. You should shoot for around a .8 lambda on a turbo car (maybe .78 at the richest), and about a .85 lambda on a n/a car. Regardless of fuel. Shoot for 1.00 lambda at idle, unless you have a lumpy cam. Then you might need to throw some extra fuel at it, like .96 lambda or something like that.

You can change the stoich value in most Ford PCMs from the 14.64 gas setting to whatever you want. In the case of E85, 9.85. The fuel tables are already listed in lambda, so that alone will effectively retune your car for a different fuel, assuming you want to keep all the same lambda settings in the base fuel tables. You would still need to retune spark tables. Since E85 is higher octane, you could and should move the borderline spark tables closer to the MBT tables, to take advantage of the higher octane gas. Some dyno testing would be advised here, to know exactly how far.

When you try to scale stuff, it should probably only be for one reason: you hit the maximum lbs/hour of air that ECU can allow. Otherwise, scaling is a bad idea. Certain things are scaled together also, but not all things. Like the lo and hi injector slopes, cubic inches (to keep load right), and maf xfer function. But not the other injector values! The flow is changing here, but not the time the injectors take to open or close, or the battery voltage effects on the injector.

Adding a MAFia to the airflow meter will just cut down on the size of the voltage coming out of the meter, making the ECU see less air. It may be a way of working around a pegged MAF by scaling it, but it also would lower load. Lower load = higher spark table values. Higher load = lower spark values. So as you increase airflow (increasing load), you would be increasing spark values because the ECU sees the lower mafia values, and assumes that lower airflow and load. On a turbo car, this is not a real good idea! Same thing goes for calibrated airflow meters. Fox bodies don't use load for the spark tables, so a calibrated airflow meter can work out "ok" on them, but SN95 5.0 cars and all mod motor cars do!

Unless you exceed the ECU's airflow limit of 63.9#/minute, leave calibrating, scaling, and mafia'ing out of the picture. Also realize that each injector has a minimum "on" time, as stated in their specs. If you have a 160# injector, at it's minimum on time is 1.642 ms, then that's the minimum squirt of fuel you can get at idle. If that's more than your motor requires or can handle, then it's gonna get rich quick. This is the problem with big injectors, especially on small cubic inch motors. They may be able to handle that fuel up high with a turbo setup or something, but at idle they might not want as much fuel as that injector forces upon them.

If you decrease the slopes of the injectors, you're telling the computer that the injector it's controlling is smaller than it really is. That will make the computer possibly command a higher pulsewidth value to get the same amount of fuel. So don't scale that unless you're scaling maf xfer and cubic inches as well.

In short, you need to model the fuel injector very accurately. Then you can model airflow correctly using the wideband. If the ECU knows the right amount of air, it will calculate the fuel right for you... assuming you told it the right injector values.

You might also want to read this article on injector specs...

http://calibratedsuccess.com/Assets/Documentation/Fuel Injector Article.pdf

It may help you understand injector modeling better. If you learned nothing else, learn this: Your stock ECU will burn up running 160# low impedence injectors. You need an injector driver box for sure. If you can do the math on how big your injectors need to be for your power level, and find that you can get away with 80# injectors on E85, then switch down to them. It will make everything a lot easier for you. Also, Don is right about stoich values changing depending on how much alcohol the fuel has in it. Bear that in mind, and consider running the worst lean stoich scenario, and that goes for timing values too.

Thank you for all of the info.

I am using the versafueler for an injector driver.

I did change the stoich values to 9.7 to match that of E85 so the EEC can make the necessary calculations.

The minimum pulsewidth for these injectors as per the Ford calibration sheet is 1.64 ms which I believe is about twice what it should be at idle. Can you walk me through how to do the required pulse width calculation at say 900 RPM.

As far as the scaling goes it is absolutely necessary to scale. The slopes are .044 and .035 respectively. SCT or Ford has a max slope value of .03 thus the necessity for scaling.

I think I have pretty much tried everything, I am now in the process of getting the injectors and Versafuler tested.

I know a guy in Minnesota named Dan Schoneck has a GT500 and his tuner Andy said it ran great with the 160s. Provided it is quite different with a 5.4 displacement and the Ford GT supercar cams and DOHC.

My car has VT custom grind stage 3 blower cams which I'm sure makes this an even bigger challenge to get a decent idle but again it's not just the idle it doesn't run well anywhere.

Thanks again

Sorry, I just saw your post. Apparently I'm not on the auto-email for replies thing. Did you get your problem solved?