Evo anti-lag ECU disassembly
 

Before my vacation over the last two weeks I disassembled the code in my IX ECU that controls the secondary air solenoid. I'm not sure on the units for time because timed events go deep into the timer units on the SH2.

Since the US cars don't have the hardware this is mainly for interest.

The minimum speed is 30km/h (30mph on mine because Mitsubishi UK fit a km/h-mph converter).

The minimum coolant temp is 81C.

Then it appears to need 2 sec over load 150 to start the triggering process.

Then if the load drops below 80 within 2 seconds of the above, there appears to be a 2 second delay before there is a 0.4 second pulse of fuel, and a zero pulse of air through the valve (this is mapped from 2000-6000 RPM in a table). It will run for up to 2 seconds.

With ignition retard in the lift off areas and some pulse times put in the table it does the pop and bang thing, although I've not tested it personally because of concerns of exhaust valve, turbo and manifold longevity. It also doesn't have an EGT disable feature.

The 2 second times as standard do seem long, and I may have the time units incorrect. If I get around to testing it running I'll be able to correlate what I see with the times I think I'll see. All the above tables show in Ecuflash for my ROM.

if it comes factory then its probably mildly activated
so i dont think there would be much in the form of damage
i have thought of trying to find a complete setup and adding it to my car but you can never find a complete factory secondary air injection/antilag anywhere
interesting find though

It seems to be deactivated as standard with the air injection time at zero.

interesting read. I too have been looking for the Anti-lag hardware, but ran into a dead end.

If we could control lift off ignition retard more readily and easily(without altering main ignition maps).....there is a way to have a somewhat working software anti-lag, by retarding ignition timing a lot and adding a lot of fuel on lift-off. it would not be nearly as effective or as cool....but i can happen. But, it's not going to happen very easily with the stock ECU.

Anyway.....keep up your research....i have been really intrested in the JDM and UK anti-lag set-ups and how the ECU is controlling the anti-lag harware and igntion/fuel on lift off.


CJ

I don't know if there is some link between this and those strange "anti-lag" fuel and timing tables that show in the ECUs. They appear to be more to do with closed loop timing and fuelling adjustments.

The other "boost enhancement/anti-lag" stuff is the same as the lean spool enable/disable stuff which seems to just lean out the mixture as per some tables for a certain time.

It is feasible to make the ECU use a different ignition map in the circumstances in which the ALS is activated. Maybe it even does this already.

Maybe we can find a way to use these ignition and fuel maps for a software anti-lag set-up...that would be functional, without have the hardware to inject air into the manifold. Obviously, having a air pump and lines would make anti-lag truly functional.....but, like i stated before, you can have lower grade functionality anti-lag using just fuel and ignition timing. Most standalones have this option as well. Now, if we could only dissassemble and find a way to use what the ECU already has, that would be a good start in the right direction.

CJ

John,

This can be done without retarding ignition.

Also can be controlled during certain rpm too depending on load (lift-off). You get the noise and the turbo does stay spooling - only under certain load calculations.

:D

Do you notice less effect on lag without the ignition retard?

From the map you're running can you tell if my time units are correct? ie how long you have to be on the throttle for before it works and how long the delay is from lifting before it pops/bangs?

what anti-lag systems use an air pump? was that just a general statment or do you know of one... interesting if so, seems it would be a little over coplicated when all the goodies are right there for you.

plus it has to be possible because the group N ecu's encorperate ant-lag and they have to fit in the factory ecu casing so i would assume it is less complex than we are thinking, dont know though????

standalone ECUs use the software to retard ignition timing(-10 to -15*) and add 10-15% fuel on liftoff....of course, this is user editable.

The factory Mitsubishi Evolution uses actual hardware that is connected to the exhaust manifold, which includes a air pump. This is the "hardware" we reffer to when we speak of the factory anti-lag system.

http://www.webwombat.com.au/motoring...es/evo-8-7.JPG


If you look close, you can see the lines that run just under the valve cover along the head.....they bolt down into the exhaust manifold.

CJ

John,

Alot less lag = smoother drive. (bit like driving with no BOV).

You can actually hear the turbo stay spooling too.

I think the ignition retard is a gimmick and old-skool which is not needed unless you want the noise all the time. The air system is all that is needed really.

Your time units are spot on{thumbup} I have spent too much time on this...
Once the boost/water is at temp. setting it works from there at above 3k rpm. Although I had this working at 2k rpm too. Minimum load I set to is 80 not 45. 45 is good, very mild, also 70 is good load but 80 (fq360) load is better. If you gun it you can get it to pop/bang at every gear change (change gear at 4-5k +/- rpm). If you red line it you get the usual pop but no real need for anti-lag there. It bangs between 3krpm to 6krpm and keeps turbo spooling. You can also drive civilized without the need for the pops and bangs.

I returned to an earlier map and you can feel the difference during spooling. I will email you mine.

SK:)

It might be not needed as much on the JDM and UK EVO using the factory hardware. BUT, it is absolutly needed and has to be imployed to have a anti-lag system using a set-up that does not have the hardware and a air pump connected to the exhaust manifold.

Anti-lag is when a turbocharger is subjected to a very late burn causing the exhaust to expand in the turbine housing. This energy is concentrated at the turbocharger causing a very quick turbo spool up or to keep the turbo spooled. The basic idea is to fire the ignition after TDC to cause afterburn in the exhaust to spool the turbocharger.

Ignition Retard: The more the ignition is retarded, the more afterburn and heat will be put into the exhaust.

Additional Fuel: Added fuel will also aid in the anti-lag process by insuring the fuel mixture will be there throughout the engine cycle. A typical addition is 5-10%

CJ

I should have mentioned my AFR is 11.5.

Yes you are right about the air system. I am not sure if you will get it to work the same without the air igniting the fuel (SAS) unless of the exhaust mani. heat.
You can try the below timing numbers on all your high octance timing maps from 2500 rpm to 7000 rpm from load 0 to 20. Let us know if by retarding timing you can get the turbo to stay spooling.

-14 -14 -4
-10 -10 0
-10 -10 0
-10 -10 0
-10 -10 0
-10 -10 0
-10 -10 0
-10 -10 0
-10 -10 0
-10 -10 0

Cheers!

OK.....well, i got a somewhat working anti-lag set-up going. I'm still working on it, but it's slowly getting there. -10 * ignition timing and lots of fuel on liftoff..... If there was a way to lengthen the injector over-run cut-off delay....this would be perfect. It pops and bangs and shoots some nice flames and spools the turbo, but there is no way to keep it going unless you get back on the throttle again, then lift.

I'm still working on it.

CJ

The timing numbers given are very mild. You can either extend the load area to load 40 (progressively) or decrease current timing to -15.

Let us know how you get on?

I guess it's for mainly changing gears and not keeping the turbo spooling constantly in case of slowing down etc.