[rohan_fonseca]

@Shan2U: There won't be too many who'd dyno their car before and after an ECU reset!

[Shan2nu]

The Ignition learning and MAPS are patented documents, that how its a knowhow. Please search.

Beyond this, I am unable to add any further value to your queries.

@Shan2U: There won't be too many who'd dyno their car before and after an ECU reset!

No prob guys. The people who tested their cars prob felt the diff was not drastic enough to be mentioned.

Shan2nu

[memo45_m]

Originally Posted by Shan2nu

Coz as far as i know, performance cannot be gauged by feel. Even i feel my car performs much better when it's been ripped for a while, so whos to say whos right?

I am sorry to say this but they have test drivers for a reason, secondly I am one myself, backing with number is a good thing but there is nothing like the feel while you drive, and being a pro-dyno technician I can also tell you that the dyno does not provide all real world conditions and its still a machine. Also claims of any front wheel drive car doing & secs is still some thing to see, the fastest FWD car is a 9.5XX. And dude claims of 7 secs on little engines is a little to much.

Originally Posted by Shan2nu

PS : Theres one point you haven't answered yet. Which "base map" will the two OHCs use on the 4th run?

Shan2nu

Please read my post again, I have said when the ecu see the same conditions in consistincy over 2 drive cycles. Which means that the parameters have to match in exact voltage and amprege for the ecu to adapt the change and then it moves the values from the short-term fuel trim to the long-term fuel trim. Then based on the values in the long-term the ecu will againg make changes in the short-term. But these are still generic conditions each manufacturer has there way of running.
To begin with please porvide me with the answers to these conditions:
1. Is the car being driven by the same person?

2. Under the same real world conditions at the same time?

And a generalized both the OHC's will run different base maps as
1. OHC1 will be calculating fuel trims based on the previously stored data in it.
2. OHC2 will be calculating fuel trims from the manufacture base trims to begin with.
3. OHC2 will have a performance edge over OHC1 as its base maps, now to begin with see harder acc and higher timing curves. As thats what you have asked the adaptive system to add in.
To really prove this point you will have to ride a Automatic as it is more consistant in shifts as controlled by a computer. Yes Indian cars too have Adaptive learn. Well try RE-SETTING your ecu and consistantly driving you OHC in its ideal power-band zone for about 10 kms, then drive for about a day or two, then RE-SET the ecu again and drive the car in every gear without crossing 2500rpm on the tach for about the same distance, then try flooring the pedal to the floor and feel the accleation you get. Then see the difference in the way the car behaves.

[Shan2nu]

I am sorry to say this but they have test drivers for a reason, secondly I am one myself, backing it with number is a good thing but there is nothing like the feel while you drive, and being a pro-dyno technician I can also tell you that the dyno does not provide all real world conditions and its still a machine.

Im not looking for dyno figures, all i wanted to know was, how much of an improvement there was in the acc and roll on figures after the ECU was reset.

To really prove this point you will have to ride a Automatic as it is more consistant in shifts as controlled by a computer. Yes Indian cars too have Adaptive learn. Well try RE-SETTING your ecu and consistantly driving you OHC in its ideal power-band zone for about 10 kms, then drive for about a day or two, then RE-SET the ecu again and drive the car in every gear without crossing 2500rpm on the tach for about the same distance, then try flooring the pedal to the floor and feel the accleation you get. Then see the difference in the way the car behaves.

Thats what i'v been saying from my very first post (#74). I'm able to feel this diff without having to reset my ECU.

It was just this afternoon that my car felt lazy on the throttle, took it for a 13km hard run and it bcame very responsive. If i hadn't felt this on my car, i wouldn't have even started this topic.

If i had a Vbox or some other reliable performance equipment, i would have tested the acc/roll-on of my car in these 2 modes.

Shan2nu

[rks]

Originally Posted by Shan2nu

Thats what i'v been saying from my very first post (#74). I'm able to feel this diff without having to reset my ECU.

It was just this afternoon that my car felt lazy on the throttle, took it for a 13km hard run and it bcame very responsive. If i hadn't felt this on my car, i wouldn't have even started this topic.
Shan2nu

I don't think it is the case that you will *always* get a noticeable improvement in performance when you reset the ECU in a stock car. You might get a noticeable improvement in the following scenario.

Consider a car whose ECU has not been reset for a long time. Obviously you cannot possibly have been ripping this car for such a lengthy period of time and the ECU would have probably factored in many long-term considerations into its code. It is possible that the ECU would be moving on the wrong track and get you further away from the kind of performance you want. Of course if you take such a car and keep ripping it, eventually the ECU should learn what you want it to learn and correct itself. But it would be simplest to reset the ECU and then rip the car, for then the learning process will be that much quicker (it doesn't need to "unlearn" all that it learnt over a lengthy period of time). Sometimes the reset could even be essential to get the ECU on the right track.

The stock ECU is supposed to be fairly sophisticated and does factor in long-term considerations. That is why I believe in sticking to the stock ECU, for the manufacturer knows the car best. I have no doubt in my mind that my Santro's ECU has adjusted the ignition timing to extract noticeably better performance out of Speed 97. I certainly wouldn't want to reset it in the near future, because that would almost certainly bring down the performance of the car.

[Shan2nu]

Consider a car whose ECU has not been reset for a long time. Obviously you cannot possibly have been ripping this car for such a lengthy period of time and the ECU would have probably factored in many long-term considerations into its code. It is possible that the ECU would be moving on the wrong track and get you further away from the kind of performance you want. Of course if you take such a car and keep ripping it, eventually the ECU should learn what you want it to learn and correct itself. But it would be simplest to reset the ECU and then rip the car, for then the learning process will be that much quicker (it doesn't need to "unlearn" all that it learnt over a lengthy period of time). Sometimes the reset could even be essential to get the ECU on the right track.

Then why is there such a noticeable diff in my cars performance when i drive it hard for a couple of kms?

Since everyone here is going by the "feel" factor, we can't really say the reset will actually make my car quicker than it already is when it's fired up.

Anyway, what i'l do is, go for a hard run so that the ECU gets fired up, do a roll on test where there's no shifting involved (100-140 in 3rd or sumthin like that), reset the ECU go for another hard run and perform the same roll on again. This should tell me if the car has actually improved it's 100-140 time in 3rd.

Shan2nu

[memo45_m]

Originally Posted by Shan2nu

Anyway, what i'l do is, go for a hard run so that the ECU gets fired up, do a roll on test where there's no shifting involved (100-140 in 3rd or sumthin like that), reset the ECU go for another hard run and perform the same roll on again. This should tell me if the car has actually improved it's 100-140 time in 3rd.

Shan2nu

Please update us on the difference when you do carry out the project, by the way there is just a slight increase cause the map curves are steeper than the base map the ecu reacts in a better manner giving you a slightly increased spark advance and an increased injector on time (or off time in case of DSM vehicles). therefore its a very slight increase in performance, in time figures it could be in the hundreths of a second when run on a 1/4th mile on the dyno.
Will also run the same test on the dyno and let you know the difference this Friday for my Eclipse with regular driving and the a hard drive with the ecu re-set.

[memo45_m]

Originally Posted by rks

I don't think it is the case that you will *always* get a noticeable improvement in performance when you reset the ECU in a stock car.

Buddy they call the feature Adaptive learn for a reason and its because it learns then adapts the change.

Originally Posted by rks

Consider a car whose ECU has not been reset for a long time. Obviously you cannot possibly have been ripping this car for such a lengthy period of time and the ECU would have probably factored in many long-term considerations into its code. It is possible that the ECU would be moving on the wrong track and get you further away from the kind of performance you want. Of course if you take such a car and keep ripping it, eventually the ECU should learn what you want it to learn and correct itself. But it would be simplest to reset the ECU and then rip the car, for then the learning process will be that much quicker (it doesn't need to "unlearn" all that it learnt over a lengthy period of time). Sometimes the reset could even be essential to get the ECU on the right track.

Bud it an electronic equipment so it either works or it doesn't, the ecu is not as smart as the person who programmed it, it just acting on the various inputs it gets and compares it with the base map for certian parameters and reacts as required and yes the easiest way to un-learn is to re-set.

Originally Posted by rks

The stock ECU is supposed to be fairly sophisticated and does factor in long-term considerations. That is why I believe in sticking to the stock ECU, for the manufacturer knows the car best. I have no doubt in my mind that my Santro's ECU has adjusted the ignition timing to extract noticeably better performance out of Speed 97. I certainly wouldn't want to reset it in the near future, because that would almost certainly bring down the performance of the car.

By the way putting in a higher octane fuel than required fuel grade could bring performance down, for a stock car it is very much recomended to stick to what the manufacturer specfies. I am guessing the recomended grade for the santro is 87 octane and if I were you I would at max put 91 under extreme heat conditions.

[jquest25]

Post Deleted by the Team-BHP Support : SMS / Slang is STRICTLY prohibited on this community. We would much appreciate your using full & proper english in posts, for the benefit of other forum users.

Please view our board rules carefully before proceeding any further.

[rks]

Originally Posted by memo45_m

Buddy they call the feature Adaptive learn for a reason and its because it learns then adapts the change.

And the learning is not always short-term learning. Sometimes the learning period can be as much as 5000 kms (see my one of my earlier posts in this thread, where I have quoted from a Toyota Landcruiser website).

Bud it an electronic equipment so it either works or it doesn't, the ecu is not as smart as the person who programmed it, it just acting on the various inputs it gets and compares it with the base map for certian parameters and reacts as required and yes the easiest way to un-learn is to re-set.

Sure, but the stock ECU is a fairly sophisticated piece of code. It is actually more that just a piece of code, for I have read that the ECU can re-program itself.

By the way putting in a higher octane fuel than required fuel grade could bring performance down, for a stock car it is very much recomended to stick to what the manufacturer specfies. I am guessing the recomended grade for the santro is 87 octane and if I were you I would at max put 91 under extreme heat conditions.

That is what I would do for a relatively new car. But now my Santro has done 77000 kms. For older cars, higher octane fuel can be beneficial. And there is no doubt that it actually is beneficial. The performance of the car has improved by leaps and bounds. I estimate that that car's power has increased by 3-4 BHP and the mileage has increased by 0.5 to 1 kmpl. By the way, they don't sell anything less than 94 RON (Research Octane Number) petrol in the UK.

And if I do shift to normal 91 RON petrol at some stage, I would surely stand to benefit by resetting the ECU. That way I can avoid a lengthy re-learing period. But I have no intention of doing that for now.

[Shan2nu]

Please update us on the difference when you do carry out the project, by the way there is just a slight increase cause the map curves are steeper than the base map the ecu reacts in a better manner giving you a slightly increased spark advance and an increased injector on time (or off time in case of DSM vehicles). therefore its a very slight increase in performance, in time figures it could be in the hundreths of a second when run on a 1/4th mile on the dyno.

Hundreths of a sec over a 1/4 mile, is too small a diff to be gauged with a stop watch. We will need a high end GPS device like the VBOX II to catch such a minor change.

Like i've said before, maybe thats the reason why there aren't any performance charts available, the diff in performance must have been found to be negligible.

Shan2nu

[memo45_m]

Originally Posted by Shan2nu

Hundreths of a sec over a 1/4 mile, is too small a diff to be gauged with a stop watch. We will need a high end GPS device like the VBOX II to catch such a minor change.

Not exactly that hundreths could be the winning or loosing number when you draging down the 1/4th mile bro. Been there and lost the last time I did Brackets after getting through 4 elimination rounds.

Originally Posted by Shan2nu

Like i've said before, maybe thats the reason why there aren't any performance charts available, the diff in performance must have been found to be negligible.

Shan2nu

When I dyno my Eclipse tomorrow I will give you the figures, i will zero all values on the E-Manage Ultimate then run the car all day before i put it up on the dyno pull the figures then re-set the ecu then rip the car on the dyno and then re-run it for the figures.

Originally Posted by rks

And the learning is not always short-term learning. Sometimes the learning period can be as much as 5000 kms (see my one of my earlier posts in this thread, where I have quoted from a Toyota Landcruiser website).

I agree that the cars did take as long as 5000kms for learning but you see more cars see advanced technology before suv's and trucks, most cars sold post y2k, have ecu that are capable of doing very good fine tuning and adaptive learning specially Europian and Japanese cars.

Originally Posted by rks

Sure, but the stock ECU is a fairly sophisticated piece of code. It is actually more that just a piece of code, for I have read that the ECU can re-program itself.

Bro the ecu is not a robot to re-program itself and I have been trained to mod/tune ecu codes, I have been in a trade school for the same reason.

Originally Posted by rks

That is what I would do for a relatively new car. But now my Santro has done 77000 kms. For older cars, higher octane fuel can be beneficial. And there is no doubt that it actually is beneficial. The performance of the car has improved by leaps and bounds. I estimate that that car's power has increased by 3-4 BHP and the mileage has increased by 0.5 to 1 kmpl. By the way, they don't sell anything less than 94 RON (Research Octane Number) petrol in the UK.

Dude again the conditions in UK and India are not the same, I even have dyno charts for my 185bhp 2001 Mitsubishi Eclipse with a 3.0l V6, having 77890miles on the odo, for 87octane, 91octane, 94 octane and 97 octane, and at an altitude of 1mile in Denver my car performs best with at 91 octane, but the same car gave the best performance results with 94 octane in California, the octane ratings have be specfied by the manufacturer in the owner's hand book. This was a project I had to do for my Advanced Electronics course.

[Shan2nu]

Not exactly that hundreths could be the winning or loosing number when you draging down the 1/4th mile bro. Been there and lost the last time I did Brackets after getting through 4 elimination rounds.

Ofcourse on a drag strip every fraction matters but when you drive normally, i don't think this hundreths of a sec will be noticeable to the drive. Remember it's hundreths of a sec "0.01-0.09" not tenths of a sec "0.1-0.9".

Anyway, lets see what the dyno figures have to say.

Shan2nu

[v1p3r]

Originally Posted by dadu

Once reset the ECU will put itself into 'agressive learn mode'. When it is in agressive learning mode, the ECU will keep pushing timing and fueling, until it starts to sense knocks, then it backs off. It does at all RPM's, under all loads. And this is done for its continous and non-continous monitors.

This is all assuming your car is equipped with a knock sensor. Which most Indian cars aren't.

Originally Posted by Shan2nu

Hundreths of a sec over a 1/4 mile, is too small a diff to be gauged with a stop watch. We will need a high end GPS device like the VBOX II to catch such a minor change.

I have an issue with this. It's well known that all civilian GPS apps have an inbuilt fudge factor. Added to which, India-located devices are accurate to 15m at best. Also, the average accuracy of a civilian GPS based device, is at best, 15m anyway. I still think a synchronised timer and laser based system is the most accurate. 15m is a hell of a lot in a 400m or 0-100 kmph scenario.

[Shan2nu]

I have an issue with this. It's well known that all civilian GPS apps have an inbuilt fudge factor. Added to which, India-located devices are accurate to 15m at best. Also, the average accuracy of a civilian GPS based device, is at best, 15m anyway. I still think a synchronised timer and laser based system is the most accurate. 15m is a hell of a lot in a 400m or 0-100 kmph scenario.

Use any method you like, the more accurate it is, the better.

Shan2nu