RalliartN

iTrader: (1)

Why don't you ask ROCK instead of Hackish?

ModernRacer

iTrader: (4)

if he knew what he (ROCK) was talking about then im sure he could answer

OttRalliart05

With no disrespect to either ROCK or HACKISH and it sounds like they both know their stuff....but i want to know if if anyone has their kits running properly??? and when I say that its easy enough to get the thing running but from my readings if you don't have all the previously said functions in check it wont work properly for long.....I have read the 5th will lean out and people out west have to get re tuned every so often....my source a tuner in Edmonton Alberta. I have tried to find one other company that uses the 5th and I can't find one...so far but I am looking....I am not trying to bash the set up but I am wondering why its being used? I know this Ralliart is not easy to tune but I am just skeptical of how well it works in the long term. The way it was described to me is your driving along under normal "stock" ride and then you punch it and once the turbo spools up and you hit 3000RPM-ish??? whammo its full boost and power.....thats seams hard on the stock internals.....again I could be wrong.....

OttRalliart05

Again I am asking people with either ROCK or HACKISH set ups. Any vendor will defend their set up vigorously as they should and that why I am asking guys and gals who have their kits and have actually tested them over time....so when I eventually go turbo I want to know what kit to get and what is safe.

DangerousDan

iTrader: (8)

RRM has been making the kits for a couple years. there are guys on here with the kit running that long, but I don't believe anyone is up to Michael or Robs level of understanding; that limits what those people can tell you. the guys who COULD tell you have built there own kits, such as thunder-rush (using RRM's NA piggy for two years; I hope he posts up) and modernracer (using megasquirt and now split second iirc).

excellent thread hackish, did you ever look into pmr's o2 adjust and MAF elimination features? between that and the FIC or comparable injector controller (yeah, too many units) you should be able to trick all the systems you want getting further and further from the stock ECU control though I am in the final stages of my move so I will be back on track (hopefully) with my car soon. hopefully that means more steps forward with the mapecu2, I know the guys there have to be frustrated with me because the last time I was in contact I was getting ready to send them some maps for review gotta take care of me first though, they'll understand

hackish

I can't comment for Rock - he will have to talk about his kit. I have worked on a few. They were all running a significant difference in AFR between cylinder 1 and cylinder 4. This is a well known problem with an extra injector and it's been known for years. The trouble is that air/fuel does not have the same distribution across the cylinders.

The way an additional injector kit works it will by definition lean the ralliart out. There is no easy way to get around it. The only way you can avoid it is to rail the longterm fuel trim at -25% but then if you reset the computer it will have a longterm fuel trim of 0% and if you floor it the car will run 25% richer and like a bag of ****. Also, railed at -25% you will throw a fuel system lean check engine.

If you never run boost at part throttle and always floor it when you get into boost then there is more of a chance that your car will not lean itself out. Now that I understand how the O2 routine in the ECU works I can tell you that it's not as simple as making a circuit to simulate a happy O2 sensor - the factory ECU will figure it out and end up with excessive fuel trims in either direction.

I suppose I've already let out enough hints to others who are working on kits that they will understand what's happening. Probably I should have just kept it to myself.

-Michael

hackish

Hi,

Yes I looked at it. As I told you quite some time ago in it's current configuration the mapecu will be unable to do the job. The factory ECU does a lot of trickery to determine if the O2 sensors are doing what it thinks and it's not a simple task to react properly to everything it does.

-Michael

DangerousDan

iTrader: (8)

I don't recall that particular statement, but we'll run with that and I'll suggest a revisit at a later time

hackish

Dan,

The difficulty is not in making something work at the instant it is in making it work for the long-term. If you can keep all the fuel trims and such in check while achieving the target AFR then you are making progress. Part of the reason it's taking so long to develop the turbo kit is that I want to see a minimum of 5000km of mixed driving without any CEL or fuel trim problems. You can stick an extra injector in there or 4 extra injectors and they will achieve your target AFR but come back after beating it for 2h on the track and we'll see how good the AFRs are. That sort of problem does not bode well for the longevity of one's engine.

-Michael

DangerousDan

iTrader: (8)

I agree, but adding or subtracting from a specific signal is not the same as emulating said signal. if you don't try and emulate the o2 sensor to trick it then the ECU should read any and all changes it makes just as it would without the change

unless I am severely misunderstanding how the o2 adjust works on this unit this is the case here: you leave it at zero and the o2 sensor functions normally; if you give it a positive value it will add that much voltage, conversely a negative value will subtract that much voltage. you should be able to shift your baseline either direction without the ECU seeing the change while still letting the ECU make any adjustments it would normally make, ie cold start or testing the sensor.

your other problems, ie MAF signal and timing adjust, are still there as the mapecu2 is still in development (not my main concern, and I don't believe anyone else wants to come out of pocket for a new one just to try and see if it works) for the MAF sensor and long term has not been tested for timing control. since it intercepts the igniter signals going to the coil packs and triggers them itself the only issue would be if there is some kind of test routine for the ignition (specifically the coil packs themselves) system like you say there is for the o2 system.

again, this doesn't really solve your problem, especially if you didn't want to have four or five piggybacks on the car but at the very least it will put some new ideas on the table for you.

if I am wrong please explain what the issues are. we are all struggling for this information and I know you are doing this for your profit, but look at how many people purchase retail kits even though there are members on here with custom kits. it's not like anyone who could realistically use the information would be buying a retail kit anyway...

hackish

Your understanding of the O2 routine is not correct. I will be offering a complete turbo kit as well as turbo kit components and the much needed solution to the ralliart problems. I can try to answer specific questions but you can rest assured others who have problems with their kits are paying close attention.

I'm not looking to "take advantage" of the ralliart community because I'm the only one who can fix the problem, just get a fair return on the number of hours invested in disassembling some of the rountines in the stock ECU and figuring out the parameters in the ECU.

-Michael

DangerousDan

iTrader: (8)

ok, that's fair enough. why can't you add or subtract voltage to the o2 sensor circuit? as far as I know we don't have current based sensors so they can't be all that accurate. if the ECU only adds/subtracts fuel to check the sensor, and merely needs to see a heater circuit upon startup, then there is definitely a big difference in what I think and what you think (can't say know, because you haven't proven it )

I don't work for RRM, and that's clear from some of my posts. I do think they do what they think is best for the end user while still considering their business, but I don't agree with everything they do either. is there any way we can take this to PM? you can obviously see I abide by my verbal/textual agreements, legally binding or not, so don't be worried about me blabbing on about what you find or tell me. this is all for my little brain to learn new things. I'm tempted to trade the RA altogether (ready for something a little more luxurious) except I don't think they'd give me a loan to replace it with (poor credit still) and I have quite a bit of time invested in research and labor on this damn thing(not to mention parts and the mapecu2 still in development that PMR has spent loads of time helping me with).

if you're comfortable with it I'd like to continue this conversation, just PM me if your cool with it. if not, make that money bish just remember we're all hoping you come through with a much better solution for the RA... while still being "reasonably affordable" for the RA (is that possible?)

Kristopher

Michael,

5th injector setups have never been desired in many applications for the reasons you've noted. However most applications will also be within range if they've been relying on the 5th injector's fuel source even when not present. They'll see a 'lean' condition when the 5th isn't activated then immediately begin modifying trims to compensate. My thought is that the 5th injector setups are initiated too soon and not allowing the ECU to add enough of its own fuel via the stock injectors first. Instead, it's taking over and the stock ECU is having to pull fuel which confuses it since the airflow is increasing yet the TPS isn't moving. Thus over time the stock ECU will trim out a lot of fuel and you 'rail' it causing a code.

For what it's worth, the Haltech Miniceptor does indeed work with our MAFs. There is a running RA as I type this. Did you give up on your Haltech?

But I also have a question for you which you can PM me if you'd like.
Are you certain the TPS is read as a percentage in the case of the RA?
Most other mitsus only refer to TPS to determine when a change has been made for tip-in. It's not directly related to fuel input in that sense.
The only explanation I can see is the ECU might have a coded limit such as "IF Airflow=X, then TPS >Y" and if that's not the case then it trips a TPS code. So, there's another simple way around this too; larger injectors.

hackish

When the 5th injector is not squirting then the ECU is trying to maintain 14.7. If you have it floored then it uses a fixed fuel value plus or minus whatever the last longterm fuel trim was. So if you hold 90% throttle and boost the 5th injector will be happily spraying but the ECU will be going 14.7 14.7 14.7 as it subtracts more and more fuel trying to achieve the 14.7 it so desparately seeks. It will eventually rail itself at -25% and trip a CEL. Before that if you punch it (chewy) you hit open loop but with a negative fuel trim. The 5th injector setup has no way to know what the factory ECU is putting in there and subsequently will fail to correctly fuel the engine.

With the addition of a certain component the Interceptor does work with the ralliart - in fact the NA build we did used one. The problem with the interceptor besides having no control of the closed loop fuelling is that it is unable to control larger fuel injectors and is therefore unsuitable for a turbo application.

When you install larger fuel injectors you can crank the airflow down a little but the fuel scheme within the ECU will switch to the MAP signal below a point because the MAF signal is unstable at low airflows. This means you have to modify both signals properly and that gets complicated and unpredictable. As an engine tuner I require that every modification is reliable and predictable otherwise I should be called an engine guesser not an engine tuner.

I worked with the Haltech engineers for quite a while and many dollars were spent on the telephone to Australia. They tried very hard but ultimately were unwilling to make any changes to satisfy the requirements of the ralliart.

-Michael

Kristopher

Wait a sec, I just had to re-read what you wrote and edit my response...

The ECU will trim up and down depending on a variety of factors. As all other Mitsu ECUs go to my knowledge, at WOT (not necessarily 100%) they don't reference the fuel trimmed tables. It used to be my belief that they did, however over time and in speaking with multiple people who've reverse engineered some of the ECUs it was found that when in Open Loop the ECU just uses a set of parameters based on Airflow, Coolant Temp, Air Temp and a couple others. The Trim calcs don't get introduced in this scenario.

Let's not get Closed Loop confused with Stoich. Closed Loop only means the ECU is referencing the O2 sensor and applying changes to the output (trimming).

When the Airflow exceeds X and the ECU references the TPS and sees it's not at 100% I suppose it's possible to set up a circuit that would send 4.99v to the ECU so it's happy with the airflow and not suspecting the throttle plate is wide open. This might serve to solve the trim issue you speak of as well since the ECU should certainly be in Open Loop when the 5th is active.

The MAF signal at low rpms shouldn't be any different and have nothing to do with the injector size. The piggy (Haltech for example) will just modify the signal so it's not a fault of the MAF itself. I think what you're saying is the ECU doesn't like the lower airflow values. I'm not sure I see the problem with this, but I'll assume you've tested and found it to throw a code. My suggestion would be to idle a bit rich.