Technical Woj's turbo project

[cc1]

The 1.2 engines are in a lower emission bracket in the UK so it makes me wonder about the maps for the 1100. I fitted a Punto MF59 1.2 ECU to the 1.1 and it felt lacking in power.


How much difference in mapping is there between the Euro 3 and 5 engines?
Does the Mani Cat require an adjustment in the mapping to warm it up quicker?

 

[woj]

The 1.2 engines are in a lower emission bracket in the UK so it makes me wonder about the maps for the 1100. I fitted a Punto MF59 1.2 ECU to the 1.1 and it felt lacking in power.?

Interesting. My guess is that both engines are well within these brackets anyhow, so little to do with the mapping. At least fuel mapping, as this is corrected by lambda anyhow. Lacking power - my guess it was simpy overfueled, or if the two engines use different injectors (I would not believe so), then there might be effecive underfueling.

How much difference in mapping is there between the Euro 3 and 5 engines?
Does the Mani Cat require an adjustment in the mapping to warm it up quicker?

I am very dumb when it comes to the Euro numbers (who cares ), but I can tell you that the newer IAW programs have a considerably different architecture and map design than the old ones. And you can call me crazy, but I can actually feel it when driving. I got my maps in the 4AF close to where I want them to be and according to the AFR readings it is OK, yet I can feel under the accelerator pedal the engine having different "habbits". It's this thing when you change cars from one to another, you have to learn to press the throttle in the right moment and to the right depth all over again.

As for the second question, I am not sure if that is required, but the maps are defo made to go to the stoich mixture of 14.7 AFR as soon as the engine can handle it and keep stable idle. The manicat heats itself up quickly from the signle fact of being where it is.

While I am at it, I have been playing with 4AF a bit more, I am getting to make it play along fully. I managed to figure out how important the after start enrichment is - make it too high and the engine chokes a bit right after start but keeps alive, make it too low and the engine starts up like crazy (revs going up and all) to instantly die afterwards. At least this is my experience so far. I have to find the golden middle.

I also found out that this ECU is very aggressive in charcoal canister valve operation. From what I can tell so far, this is what gives me occassional fuel flactuations. My charcoal canister system is defo different from the one in Sei, so the ECU most likely is simply not operating it as it should. My next step will be to try to disable it altogether. (Actually, even with the old ECU I had nothing but trouble getting the canister operation right).

 

[woj]

OK folks, I think I am kind of saturated with the work on the ECU running my engine. I know what tickles it, I am almost sure I can set up all the things I want. It now starts up flawlessly, idles very stable, lambda correction and fuel adaptation is working, there are no DTCs or MIL. The only thing now left is to fine tune all the things I know should be tuned. This is a lot of work, lots of logging, lots of analysing, and a dyno visit too. Without all this it is not going to be as smooth as my 18F, but I think I just can't be bothered at the moment.

What I did notice during my last tests is that this ECU is indeed EURO-X crap thing. The charcoal canister valve operation is one of the basic components of fueling, and playing with it have drastics consequences on the mixture richness/leaninees. It is a very fine game between operating the injectors and the canister. And there is the whole second lambda operation routines that in my case are all switched off. I would be guessing that is one of the things that makes the factory setup with this ECU run so smooth and stable.

I will think through the weekend about what else I can still test and figure out while the ECU is still in the car (any ideas?), but otherwise I will probably just drop my old reliable ECU back in.

 

[AndreaRC]

Hi woj! have you disabled canister in your 18f project?

 

[AndreaRC]

I guess the valves are not the same, but i'm not sure

 

[woj]

Hi woj! have you disabled canister in your 18f project?

I can, but I didn't. 18F has a very lightweight, but very effective auto adaptation mechanism for the rest of the fueling routines not to have to worry about canister operation. The only problem was that I had this mechanism switched off while keeping the valve running, and that was giving me problems that are now all resolved.

 

[woj]

I guess the valves are not the same, but i'm not sure

You mean between the 1.1 MPI Sei and 1.2 P75 engines? Oh yes, moreover, the factory cam of 1.1 and the one that I have are two different worlds. All this has serious consequences on getting the VE cooeficients right.

The major headache is that I started working with the factory 1.1 Sei maps as a base and only rough guesses on how they should be adapted based on what my 18F program has. But that has not brought to be to the sweet spot I was hoping for. What I could use is an ECU with factory maps for P75 engine that is based on the same algorithms as IAW 4AF. I think the 49F ECU might be what I am looking for, but it is, as mentioned above, implemented on an undocumented CPU (Motorola ZC439507MFT20).

 

[AndreaRC]

I spoke about 18F canister valve and 49F canister valve. I'm sorry, I didn't specified that...

 

[woj]

Don't be sorry, I should have figured that out, just being a bit tired

Not sure about the electro valve itself, might actually be the same, but the canisters are different, so surely the same duty cycles in the factory Sei installation and in mine give different mixture.

 

[woj]

I wouldn't be myself putting it on the shelf so soon As a last attempt I made a proper AFR log with the Zeitronix module.

The good thing is - my EBC code works rock stable, gives even better results than the one that I have in 18F, see the attachment. Kicks in without overshot and keeps a flat line.

The bad thing - I still get unacceptable fuel dips, see the second attachment (during what you see the engine runs, but has this very feelable dip and hesitation). Today I was fighting a lot with maps I thought should cure it, no luck. But luckily, I think I have finally found the right map just now, it is as simple as fuel enrichments on MAP sensor reading deltas. Tomorrow I will give it another try.

 

[woj]

OK, this time I really give up, the old ECU is in the car, but not because I do not want to deal with the ECU, I just decided I should give it a rest for a bit and perhaps come back to it later when I will also have some motivation to take it to a dyno.

I still get the fuel dips, the same magnitiude. I started thinking about it, and came to a conclusion that this might be purely mechanical. In the MPI Sei the MAP sensor is mounted directly in the throttle body, in my engine remotely and with a muffler. My guess is that this gives a split second delay in the MAP readings and hence the ECU has no time to react to the load difference and chokes for a split second. In the 18F ECU this is cured by having a fuel correction on TPS reading deltas, here there is only MAP delta readings.

Another possibility is that MAP readings deltas are not properly registered simply because I bypassed the procedure that checks if MAP readings are sensible, perhaps it does something more...

 

[woj]

Big win, and this time for real . I brought back the original MAP sensor serivicing procedure, only had to allow for it to register pressures above atmospheric. That was still not it, though it was a bit better. Then I started thinking about ignition advance, and long story short, because of the specifics of this ECU, how maps are read, etc. etc., effectively my ignition was about ~5-6 degrees too low. After that it runs like a dream. In the AFR log there are still dips on very sudden throttle presses (the kind that you should never do anyhow), but (a) this is not really noticable when driving, (b) I am still convinced this is because of the remote location of the MAP sensor (I tried removing the muffler from the vacuum line, this did not change anything). And even the knock control procedures seem to do something useful.

There is a but to every story - my rear right brake is fubared, it rubs and scratches horibly, this is very audible inside the car. Realising that the last time that I changed shoes was 4 years ago, this is probably it, they are finished

 

[blu73]

thats a big win and only a tiny issue, have a metaphorical pat on the back....

 

[woj]

Just came back from a longer drive, including motorway. How sure I am about this ECU and my work is this: I forgot to take my phone with me and I knew it before I took off, yet I decided without hestitation to hit the road and go with proper speeds No glitches deteceted that cannot (and should) be treated with fine tuning of the maps on a dyno. A bit more experimentation is needed to setup the base ICV parameters to be ideal, the ECU still has problems from time to time to decide what to do to get the target revs when approaching idle, but it is nothing drastic, just very short moments of trying to find itself in what is really going on

I kind of declare this project to be successful and finished for now. If I still have the car in a couple of months I will think about the dyno. ECU program wise, I have to spend some time implementing a proper PID based AFR fuel correction, I have the signal in the ECU, it can be logged and everything, but my short-cut version of the fuel correction algorithm does not work. This will take some evenings of work.

 

[Honeymonster]

Top work me old chap.

 

[woj]

Two things.

First is I am looking at this AFR correction procedures I wrote a while back and I am not sure what the hell I was thinking. It is now clear to me that it had no right to work and have a good idea on how it should go. A PID controller might actually not be necessary.

Second, about my rubbing and screaching rear brake. First I checked the wheel rotation for possible problems with the bearing, found nothing worrying, but then I am not a specialist in diagnosing those. I took the wheel and the drum off and I was expecting to see horror - eaten up shoes, friction layer detached from the shoe or broken into pieces, snapped springs, what not. To my total surprise it was all in place and by visual inspection nothing is amiss. Apart from one thing. There is a clear friction ring on the inside of the drum, outerish on the side of it, see picture. It would indicate to me that the shoes are scratching the drum sideways, probably the metal part of them, not the friction material. And it started happening because the shoes are worn off considerably and the metal part of the shoe is now closer to the problematic area. But I have not seen any scratch marks on the shoes (but then, they are very dirty and I was in hurry to put all this back together before dark).

Does this sound right to you? Obviously changing shoes should help, but somehow I can't be bothered

 

[AndreaRC]

Do you earn noise only when you're breaking? Or always when the car's going on?
In the photo the drum seem to be perfect...

 

[woj]

Only when braking, and during the last stage of it when I press the pedal more firmly to actually stop or almost stop. And the noise is not this characteristic brake noise, it is metal against metal scratch noise

 

[AndreaRC]

Are you sure the noise cames from there?

 

[cc1]

A new clip/spring set might help pull the shoes back. Could be a badly manufactured set of shoes?