When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
Originally Posted by kane.s2k,Oct 25 2004, 12:36 AM
The factory ECU is always in closed loop. There is more to tunning than just fuel. The factory ECU doesn't make fuel adjustments very quickly, but timing adjustments are made on the fly.
From the standpoint of fuel corrections, the ECU is definitely NOT in closed loop at WOT and / or high RPMs. Timing doesn't have anything to do with the discussion of how Inline Pro tuned his VAFC.
Originally Posted by kane.s2k,Oct 25 2004, 12:36 AM
For the first 10-20 minutes of the engine started every restart.
When I was using a VAFC for tuning, and had my throttle settings set to the default 10 / 50, I had to pull my backup fuse to keep the ECU from untuning itself. A few weeks later, my plugs started fouling out and my gas mileage dropped significantly (about 20 miles per tank). The ECU was running its default fuel map all the time, and it was too rich. It also had such a poor idle that it would often stall at stop lights. It has always taken at least 3 driving cycles after resetting my ECU for it to drive like normal.
After getting tired of the car running so crappy, I decided to change my throttle settings to 80% / 81% and put the backup fuse back in. The car ran perfect after that.
You can see where this is all leading to. Using a VAFC to tune is always a compromise. Fair enough if you just have a cold air intake but we're talking many thousands of dollars worth of custom built engine here. If you had a Porsche would you put retreads on it?
Even the E-manage isn't really up to this job because (as previously stated) it can add fuel nicely (great for FI) but all the changes used on the VAFC were to remove fuel. The E-manage does this exactly the same way as the VAFC so no better.
Originally Posted by white_S2000@16,Oct 24 2004, 04:24 PM
quyksyrt- good luck with the 05 model you are going to need to ditch the 8k redline its going to hurt you pretty bad cause you get vtec the same time as the older models adn dont have the same top end interested to see what it does with just 8k redline, becuasse you do still make the same peak ponies stock as the 2.0L with a thousand less rpms. I think you may make the same peak power as the 2.0l but the old ones will pull top end in a race but you may have a better tourque advantadge lower in the revs.
Gernby-the AFC sttigngs are as follows: Thr. points Lvt Ne points Nar thr.
25% Hi 1000rpm -10
45% Low 2000rpm -15
3000rpm -15
4000rpm -15
5000rpm -10
The Hvt setting are untuned so they dont matter the injectors are just dumpin fuel. The stumbling doesnt happen anymore the trick is all in what you tell the car to do as long as you go about the right way you can keep the AFC from screwing up.
Kwando- i agree there coming, these things take time no one wants those graphs mor ethen me beleive me ive been waiting for them.
Highrpmek- dont know your experiences with them,but im sure your reasons for disliking inline are sound ones to you. Your about the second person ive ran across who has this vision of inline so im gonna half to say you guys are in the few here. My experiences with them have been quite good between me and my freinds car tehy do alot fro me and my freind andthey have been a pleasure to work with on this project. I have seen inline engines last for well over twenty thousand miles. The types of motors and cars that come out of inline are very high performance thus over driving them miss-using them changing, tuning and other parts all disrupt what inline puts out, im not accusing you of this all im saying is ive heard and seen of more then one circumstance where these cause the faluires of their work. Basicly things like that are going to happen its part of the nature of the busness they run and in the end its typically the customers fault and of course there are exceptions.
Quicksyrt- hes not dissing me personaly, infact he congratulated me clearly ryan dislikes inlinepro which he is very well entitled to his opinion no matter his reasoning.
my 2.2l makes a lot more power than a 2.0l. 218rwhp pretty much bone stock. most of the 00-03s around here dyno between 190-200 whp. And, I've done roll-ons with a couple of the guys around me, and walk them everytime.
anyway, that's not the point, though. The VAFC tune is really scary, unless you have bigger injectors. In fact, given those numbers, I think they are a bandaid more than anything else. What is gonna happen is that most of your driving is in closed loop mode as many people said and they are applying those corrections in closed loop... even if you have larger injectors, (440s or so) you are gonna be running way rich at part throttle, which the ECU will correct out... that will lean out your open loop settings with the long term fuel trims. With those uncorrected and dumping fuel...
band aid, at best. They might've built a nice motor, but you really should get a custom tune done, a VAFC is just not a great solution. What you are doing is telling the ECU that there is less air coming in than really is... the problem is that this screws up the load calculation and can cause the car to run in closed rool deeper into the RPM range, which is obviously bad. anyway, it's scary that they didn't need to change the timing, either... I'm sure there is some power (or a lot of reliability if not) in doing some careful fully programmable ECU tuning.
When I was using a VAFC for tuning, and had my throttle settings set to the default 10 / 50, I had to pull my backup fuse to keep the ECU from untuning itself. A few weeks later, my plugs started fouling out and my gas mileage dropped significantly (about 20 miles per tank). The ECU was running its default fuel map all the time, and it was too rich. It also had such a poor idle that it would often stall at stop lights. It has always taken at least 3 driving cycles after resetting my ECU for it to drive like normal.
After getting tired of the car running so crappy, I decided to change my throttle settings to 80% / 81% and put the backup fuse back in. The car ran perfect after that.
My car has never taken more than 20 minutes to adjust idle and part throttle. It gets the idle down pretty good in the first 5-7 minutes I let it idle.
The car is in closed loop in all conditions. When you pull the back-up fuse and it loads the base map again, it runs extremely rich in ALL engine loads and TP. It leans out the mix slowly. If you dont believe me, go stick a wideband in your car and pull the fuse. It's running pig rich all the time with the base map. Timing changes affect a/f also.
If the throttle settings work pretty well at 80/81 then great. Just saying you should watch it. I'm not saying it doesn't work, i was just fixing a few things that were incorrect.
closed rool deeper into the RPM range, which is obviously bad
I'm only gonna pick on this part. Please explain how running in the 'closed loop' high load range is bad.
The only reason other than to get a timing map to suit the engine perfectly and for performance. The stock ECU in my guess would possibly pull a couple degrees of timing from the piston rubbing up and down the sleeves scrapping the two sides. It might see that as knock. Other than that should be normal. I'd probably get a e-manage instead just to get the fuel down on a larger map. Timing controls for your car with the e-manage would be useless because of its large increments and the +/-1 degree error it has.
Originally Posted by kane.s2k,Oct 25 2004, 07:01 AM
My car has never taken more than 20 minutes to adjust idle and part throttle. It gets the idle down pretty good in the first 5-7 minutes I let it idle.
The car is in closed loop in all conditions. When you pull the back-up fuse and it loads the base map again, it runs extremely rich in ALL engine loads and TP. It leans out the mix slowly. If you dont believe me, go stick a wideband in your car and pull the fuse. It's running pig rich all the time with the base map. Timing changes affect a/f also.
If the throttle settings work pretty well at 80/81 then great. Just saying you should watch it. I'm not saying it doesn't work, i was just fixing a few things that were incorrect.
Forgive me if I come across as rude, but I just completed a 24 hour work day, and am settling down after a few beers...
Your car must be an exception to the norm on the idle quality after an ECU reset. You are the 1st person I've ever heard say that their idle was okay after running for less than a few driving cycles. There are several knowledgeable people on this board that will avoid resetting their ECU at all if there is a way around it.
I do have a wideband in my car, and I no longer have a VAFC. I replaced the VAFC with a self-tuning piggy-back that I built using an off-the-shelf micro controller. I have no insider information about the design of the OEM ECU, but I spent months developing my fuel controller, and many more months logging the long term effect of using it (WBO2 data logs). I don't know of anyone on this forum that could be more knowledgeable about fuel control using a piggy-back than myself.
I don't have any evidence to support my statement that the OEM ECU is going to be in open loop in ALL respects at WOT or high RPMs. I would hope that it would always make adjustments for timing if the knock sensor detects detonation. However, I know that there is NO WAY the ECU adjusts the fuel maps during WOT or high RPMs. Not only would it show up in my long term tests, but it just wouldn't work with the setup the S2000 has. The narrow band O2 sensor only indicates that the AFR is above or below stoich. Since the AFR should not be anywhere near stoich at WOT or high RPMs, the data from that sensor is meaningless. The only thing that I suspect the ECU probably does is to calibrate the MAP sensor voltage to give an apropriate adjustment to the default WOT / high RPM maps. If that is the case, then that is even more reason to not do any fuel tuning during partial throttle since the learning that the ECU does would then effect the narrow AND wide throttle tuning.
Here is a comparison of dyno plots from my car on the same dyno on different days using no fuel tuning (red), tuned VAFC (green), and my custom fuel controller (blue). Take note of the AFR scale. This is the flattest curve I could do with the VAFC. I was surprised my fuel controller didn't achieve a flatter curve, but it is only as acurate as the WB that it is self tuning from (see below). I don't know if the shop's WB is better than mine, but one of them is definitely better.
Here is a log of what my fuel controller reads (and tunes from). I coded a stair step at 6500 RPMs (the x-axis is time, not RPMs) to go from 13.7 to 12.7 to see how well the controller could do it. The results were WAY better than I could have hoped for.
The theory on just MAP adjustments could be what it's doing actually.
That's pretty neat on the self-learning ECU you made. Very very impressive. I read about other people mass-producing the same type of thing for other cars. Would be great to make one that can control timing for the S/C and turbo guys.
The first thing I notice on that dyno plot is there is very little difference between the dyno pulls while the a/f is widely different, that and they didn't plot the torque curve for ya..
You should think of a/f as thermal managment. To a degree the engine is not going to care if you are running 13.5:1 or 10:1. The power is going to be made through ignition timing adjustments.
The VAFC tune is really scary, unless you have bigger injectors. In fact, given those numbers, I think they are a bandaid more than anything else. What is gonna happen is that most of your driving is in closed loop mode as many people said and they are applying those corrections in closed loop... even if you have larger injectors, (440s or so) you are gonna be running way rich at part throttle, which the ECU will correct out... that will lean out your open loop settings with the long term fuel trims. With those uncorrected and dumping fuel... 
