Nope. I can't tell the desired OEM A/F ratio at WOT (it may not even be static). You've got 8052 assembly programming code and fuel/ignition mapping data on the chip. The code has no comments. Disassembled it's 800k of code! The data is/are just numbers that tell the code to signal the fuel injectors and spark plugs (oh, and the ecu manages to calculate and send a speed signal, monitor sensors for errors, turn on fans, perform pre-emptive and realtime system and self diagnosis, talk to an SD1/2 serial device, etc.).
Exactly. Get it into the ballpark, then fine tune it. That's why you've got to do the fpr vacuum test. Better still would be to know the specs on the 348 and 355 fprs to know what fuel pressure each maxes out at. Use the highest!
Here's where it gets interesting. Notice the bottom graph and the bottom data points. Now, this picture oversimplifies things (lots of dynamic activities going on), but still you can tell that the Ferrari OEM chip is leaning back the fuel mixture around 5000 RPMs...right where Plugzit ran out of gas! Image Unavailable, Please Login
Well, here's what's happening on the project. ND provided some chips that added 10% to the fuel above 4500rpm and I put them in on Thursday. I sent him my originals to read and compare to the Mondial chips he used for the 10%. The motor now runs good to 7000, but is still quite lean as evidenced by an occasional pop on shifting and a high idle, kinda like when you have a vacuum leak. Horsepower is coming up. Originally as started with the 355 tbs and 348 injectors, it would make just over 145rwhp levelling off or dropping after 4500rpm. Not bad at that range and a good indication that we're on the right track. Anyway, with the 10% solution, it showed 205rwhp and still way lean. We're gettin' there! We're gonna add another 10% fuel throughout the range this time and see what we get. The plan is to keep adding 10% at a time til it drops off, then decrease 5%. I think at that point, we'll have gotten 75% or so of the potential additional horsepower. The current livery remains the MAFs with pantyhose filters ziptied on the open end. I ran it for a day with nothing. It was surprising the difference the pantyhose made on the plus side. The mafs must be very sensitive as the air turbulence in the engine compartment made the engine flutter somewhat in driving-or maybe it was my imagination. Certainly air management is something I'll address if I continue to run it in this iteration, but the real goal is velocity stacks. So, right now I just want to get it very close, then do the changeover to stacks and do final tuning with them. Today at the VCR I saw a GM airflow sensor used beside the stack (that's right, just beside the bell) on a GT40 continuation car with TBs. He uses an aftermarket tuning device, but it was cause for thought! Anyway, that's where we are. 205rwhp, still very lean but now close to the 225rwhp of the old system. Gonna make lots more power!
I'm ready to burn the 2nd set of chips. I've added another 10% to the maps that I modified the first time around that bumped Plugzit from 5000 rpms up to 7000 rpms. Bruce sent me his OEM 2.5 chips, so I scanned them in to compare them to everything else that I have on hand. It's interesting. His OEM 348 chips use **identical** fuel maps to my 1989 Mondial T OEM chips. Likewise, they both share a 7200 RPM rev limiter if I'm reading the code/data correctly. That similarity struck me as odd, as I was under the impression that the early 348's would rev to 7500. But...and here's where it gets really bizarre...the timing maps are different. Now, you'd think that the 348 would have the timing advanced as much as Ferrari could push...and that the Mondial T would have the timing just a bit retarded from the 348... Not so! The timing maps on the chips that I sent to Bruce, which were based on the only 2.5 chips that I possessed (the 1989 Mondial T), are slightly more advanced than the timing maps on the OEM 348 2.5 chips that Plugzit mailed to me! Which is to say, the 1989 Mondial T was sold with more advanced Timing maps than the 348. Oh, and about that rev limit. The OEM code/data for my 348 Spider says that I've got a 7560 RPM threshold (i.e. the rev limit delay of 200 rpms starts kicking in here). This is slightly different than the advertised 7700 base to 7900 rpm max cutoff for the 348.
Verrrrry interesting! Be sure to put in 10% throughout the whole rev range from idle on up this time. Is there no fuel map past 7200? Whatarewegonnadoooooo?
No worries. The fuel maps go up to a 7720 RPM map point. After that, the value for 7720 is used. So Ferrari/Bosch built maps way, way above the rev limit for our cars. Are you sure about adding 10% to the low RPM ranges? Hang on a second and I'll post what's been done as well as what I'm proposing for your 2nd set of chips.
OK, well, if this poor old chat site can stay up a little longer then perhaps you'll be able to view these pics. 1st is your old OEM 2.5... Image Unavailable, Please Login
Here's the 1st set of chips that I burned for you (that took you up to 7000): Image Unavailable, Please Login
Dang it Plugz!!!!! Will you "plug" in the Gtech chubby gave you, and give us a sneek peek at what it did up to the 5k rpm. Cummon it's only fair. After all Goth did give us the numbers from his run.
You know if you want to be cheap and minimize your guessing you could wire in this. You could drive at specific rpms and someone else can write down the A/F ratio. Then you would know how far you are awy from stoichiometirc 14.7:1 or 13:1 or whatever you want. http://www.jcwhitney.com/autoparts/Product/Pr-p_Product.CATENTRY_ID:2008855/c-10101/Nty-1/p-2008855/Ntx-mode+matchallpartial/N-10101/tf-Browse/s-10101/Ntk-AllTextSearchGroup?Ntt=o2+sensor
Yes, need some more fuel at lower end, runs lean there too. Big temptation to raise redline, but I'd better wait til it's tuned a little better. Nice of Ferrari to put in the maps! Wish I knew more about what I was looking at on the maps!
I may do a Motronic 2.5/2.7 chip reading/building tutorial one day. It's a pretty wild system. Whereas I've heard some guys talk about aftermarket ECUs that would have 8x8 or 12x12 maps, Motronic 2.5 and 2.7 can have 255x255 maps, and as many such maps as you want. Motronic units also have a wicked watchdog timer...basicly each subroutine has to end by telling the watchdog that all is right with the world. Well, a software/data error would typically prevent a subroutine from giving that OK...and if the watchdog sees a few miliseconds go by without that OK, then it takes over and starts the ecu at some very stable, known code (basicly a "reset" on the fly). There's almost no way that a 2.5 or 2.7 system could go as long as a tenth of a second with an error as we're driving...which is pretty important because these units are learning as we drive. Motronic 2.5 and 2.7 build their own drive-time maps over time. Get an error in that data and it could cause all sorts of havoc...save for the watchdog timer. This is also why the occasional ECU reset probably does some serious good.
Is that when you wear the cape and the pointy hat with the crescent on it and use a wand as a pointer? Dark science....sounds stoichiometric to me....
Well, OK. I'll add fuel down low, but I'm concerned that you won't rev up as fast. As for what you are looking at, the bigger the number in the tables, the more fuel that gets delivered. Those numbers impact a length of time that the injectors are pulsed open (injectors pulsed longer will deliver more fuel). For instance, look at the very bottom of any of those 3 pictures above where it says "Full Throttle." Below Full Throttle the stupid software chart says "0%" on the left side, with RPM numbers in gray going from left to right. Below the gray RPM points are the black text on white background injector times. Ignore the first few values (66, 4, 8, 8, 10) because I don't trust the software that I'm using for those points. After that, say, on the bottom picture (above), you'll see a number for the injector under each RPM point. Those can each be raised to at least 158. Move "138" up to "158" under the 3000 RPM point and you've just bumped up the fuel at 3000 RPMs. Motronic 2.5 and 2.7 use at least 3 different fuel maps (the others aren't shown on this thread yet), adding the values together at each RPM point to give a final answer to the fuel injectors. My proposed 2nd set of chips is pretty well going up in fuel about as far as I feel comfortable for this set of maps. If we have to add still more fuel later on then I'll work on the other 2 maps for each set that aren't yet shown.
Okay wait a minute. I just read #204 and you are making less power? I though it went like a scalded dog? 300hp - 15% drive train loss = 255rwhp, and you are at just 205rwhp? Please tell me that those numbers were only up to 4,500rpm.
Yes! That would help, especially if general throttle position was also known (about half throttle, 3/4 throttle, accelerator pedal smashed into the floor like stomping on a bug, etc.).
I've got to go with Samy on this one. The first set of chips that I sent to Plugzit show that the car needs more fuel. Bruce needs to raise the fuel pressure. Then add the wideband O2. Then we tune with more of my chips.
It moves really good on the low end with a lean motor. Made 145ish at 4000-4500 rpm. Started at 225hp with the stock motor, but the stock motor doesn't make full horsepower til higher in the rev range. Keep in mind that those aren't dyno figures, just figures from the obese one's g-meter, so they're more relative than accurate. Down now about 20hp from stock with a motor that's way lean is pretty good..when we get the right amount of fuel to it, it'll scream. We're still using the 348 injectors which were set up for the 54mm tbs. It would have been interesting to see what the 355 injectors would have done off the bat, but as you know, we could not use them since the brain would not fire them.
1. How to increase fuel pressure? 2. Why increase fuel pressure when injector time of fuel delivery can be increased? 3. How to get realtime feedback-and more importantly adjustment-on Motronic system? 4. Will these instruments tell us "how much" fuel to add? i.e., will we be able to go directly to the optimal chip design? 5. We know it's lean because we've added more air. 6. How much more $ will it cost, and what will it tell us we don't already know? While I'm open to any avenue, I really don't want to just throw dollars at a solution to a problem that can be solved simply and cheaply. We've made one change-added air-and need just to add more fuel. So, the question is, how much? I'm very excited about what we've done so far and the results are pretty much exactly in line with what needs to happen to get what we want. More air allows more fuel (the needed ratio is pretty constant 14.7/1) which makes more horsepower. If it weren't lean now, then we'd be in trouble. Lean means we can put in more fuel. Heck, we're getting 205rwhp with a motor that's not even getting close to the right amount of fuel!
