Forum » Road Tuning » Methanol Injection

Methanol Injection

Road Tuning

Discussion and questions related to the course Road Tuning

Page 1


I just finished installing an Aquamist methanol injection kit on to my EVO X. I am looking for any tuning advice on the subject. Attached are to logs, one is running no methanol injection and the second on is running methanol injection. I did not do any table editing or tuning yet. I wanted to reach out to you and find out what I should be looking at and what should be considered. Or if i need to put in a bigger restrictor to spray less? I have the kit at the lowest flow rate right now.

My car is Stock Injectors (ECUflash shows 513 scaling)

From the Aquamist I am using a restrictor of .7mm that splits into -

Nozzle 1 - .8mm

Nozzle 2 - .6mm

Both nozzles go into the upper inter cooler pipe.

My AFR's drop pretty rich after spraying...

I am use ECUScan with RAX logging to data log my car.

AEM wide-band for AFR's

Also attached is my ROM and the Tephra XML to read it.



Attached Files

When you're dealing with an OEM platform, occasionally there are some aspects that are difficult to deal with and this is probably the case with what you're trying to achieve here.

If you're spraying pure methanol or even a large percentage of methanol in a water/methanol kit, you are introducing another fuel into the engine. The methanol will combust while obviously the water doesn't. The methanol therefore affects the air fuel ratio.

This can be accounted for by reducing the fuel delivered by the injectors, however that in itself represents a problem with an OEM platform as ideally you want this to only occur when the water/methanol kit is injecting. If you run out of water/methanol then you end up lean. That being said though if you want to take advantage of the methanol injection you will typically be adding timing and/or boost, so this really needs to be dealt with carefully.

Water/methanol injection is a great technology but it needs to be treated carefully because if your tune is dialled in to take maximum advantage of the kit, you will be in some serious trouble if the water/methanol stops injecting for some reason.

Looking over your logs you're AFR is quite rich both with and without methanol injection although as you have noted, the AFR is richer with the injection operating - As above, this is to be expected. Looking at your timing and the knock sum, it's probable that you could add a little more ignition advance to really take advantage of the methanol injection.

Thank you Andre, I'm sacred to add any timing advancement. All the stock engine EVO X I have seen don't do well with over 350 TQ they tend to break something. I was hoping to up the boost to 29 - 30 psi with the injection to add HP. I'm even thinking I might have to take a few degrees of timing off at that PSI.

My AFR's are rich I tried to get them to be as close to 11.1 as I could get it. (I was hoping you could help with the Target maps and Fuel Calibration maps, Ill post another thread on that) They ended up being a little richer but not showing any knock! As you can see I am peaking 25.9 PSI with out methanol.

is there a dual map option for evo x computers?

could use the on signal from the water injection to trigger a second leaner map to maximise the water injection, have it set so when the level switch is low and the system shuts down, it deactivates the second map and you return to you're basic tune?

never worked on Evo's but I know older models could have dual maps. the method above is similar to how haltech recommend setting up progressive nitrous control. scary.

Thanks Josh,

And yes there is, its call Tephra V2 its by David Shirley he lives in Melbourne Australia! I have it on my car and I am working with him on getting that map switching fail safe active.

He has it using the Fuel Temp Sensor wire 115 on the ECU. Then I have to cut the wire going to the sensor.

The Aquamist has a brown and white wire for fail safe. I have the white wire connected to the 115 ECU and the brown wired connected to a 200 ohm resistor going to ground.

If you can use a second map for the water injection you will find your AFR is much more stable once the maps are dialled in correctly.

As far as boost vs timing, both are effectively achieving the same aim - Increasing cylinder pressure and hence engine torque. If you have a specific torque level that you don't want to exceed for engine reliability reasons (which is totally sensible btw), you could exceed your limits using either technique. Personally I always prefer a little more boost and a little less timing when I'm running on a quality fuel as this gives a margin to the knock threshold that you may not have if you have the timing dialled in to MBT at a certain boost level. That's just a personal preference but it's how I always approached tuning high boost drag engines.

Since you are on pump gas though and still at a relatively modest specific power level I would suggest trying a little more timing (very carefully) to find out where the knock threshold sits. You may find that you can still add 2-3 degrees with no hint of knock.

What is the proper drop in AFR to expect when spraying? Is .3 to 1.0 a good window?

As Andre said, you need to be careful with water-meth or any kind of water-alcohol injection kits, as trouble may very well arise when you run out of mix or get some kind of controller or pump failure.

However, something I do want to add (perhaps it's more of a question really)'s all good and well to say that, because you're introducing a secondary fuel, your primary fuel delivery should be tuned to be a bit leaner. That part makes sense. But, I've also started thinking about the chemistry behind it all.

I might be completely wrong in my assumption or way of thinking here.....but let's say you're aiming for an AFR of 12.0:1. If you check the wideband sensor 12.0:1 should be 0.857 lambda.

Now, my question here is, how exactly does that lambda value work? What I mean is, how does it calculate that lambda value?

The reason I ask this is, a lambda value of 1.00 is 14.7:1, which is stoich for Petrol. But what about when you introduce a secondary fuel such as alcohol which has a stoich of 9:1? Or perhaps an easier way to address this would be to ask, if you were only using alcohol (i.e in place of petrol), would an AFR of 9:1 = lambda 1.00? Or is the lambda sensor specifically designed to read, calculate and output a value tuned specifically for the stoich of petrol?

If this is indeed the case, my theory would be that it would be far more important to look at the lambda value, rather than the AFR value. Meaning that you would have to take into account exactly how much petrol is being put into the motor, what the specific mixture of water-alcohol is, and the stoich for that secondary fuel.

An example would be to assume that we are injecting an exact equal amount of petrol and alcohol. In that case, would I be correct to say that the final stoich value would be 11.85:1? If I am correct in that, would lambda be 1.00, or 0.806?

hi this is zaid owner of Instagram account and shop "Ta7akum_motor_sports"

what I do normally on stock ecu's for example ls3 with e38 ecm, is I put the IAT after the methanol injector therefore I would see changes with IAT and pull timing and add fuel accordingly this way what happens is u tune the car with same AFR with or without meth and you add and remove fuel and timing from different tables. AFR vs IAT & Timing Adv vs IAT and you can hook a boost safe output lets say your meth pump failed you can hook your boost safe to shorten the IAT signal to ECU and will send it to limp mode or reads maximum IAT and therefore pulls hell amount of timing.

lol I dnt know If I made sense.

Theo53, the O2 (or lambda) sensor, like it name imply only measure the oxygen on the exhaust gas.

It doesn't care of what kink of fuel you use.

You need more fuel with alcohol to consume the same volume of air (hence the different AFR) but once burned you end up with the same % of oxygen in the exhaust gas, and that's what you lambda sensor measure.

Thanks for the reply.

But what I was asking is how the 2 fuels will interact with each other, i.e if you used equal amounts of alcohol and petrol, would I be correct to say that the stoich AFR would be 11.85:1? It seems like you're saying that the lambda sensor would read 1.00 (if my assumption of 11.85:1 being stoich is correct).

But what about the actual AFR gauge? At lambda 1.00, would it read 14.7:1, or 11.85:1?

If you have a non programmable gauge (AEM for example), at lambda 1.00 it will always display 14.7:1 no matter if it's pump gas, E85, methanol or a blend.

If you have a a gauge that you can calibrate (Innovate for example) you could take into account the fuel you use.

But the calibration is up to you and if you do take into account the fuel you are using, you have to remember what scale you are using.

If you start to make a blend of different fuel and calibrate the AFR scale based on that blend, headache guaranteed !

That's why Andre recommend to use the lambda value instead of the AFR, it's less confusing if you tune with different kind of fuel (or blend) and you only have one scale to remember.

The key part to understand is that the wideband doesn't know (or care) what fuel mixture you're running. It also doesn't care what the stoichiometric AFR for the fuel is. All it's doing is looking at the oxygen level in the exhaust.

Basically a wideband will natively read in units of lambda which is a scale 'relative' to stoichiometric. It doesn't matter what's going on in the combustion chamber, a value of 1.00 will always be stoichiometric.

The aspect many don't properly understand is that to read AFR, the wideband is simply multiplying it's native lambda value by whatever the stoichiometric AFR programmed into the unit is.

I personally see no upside to tuning in units of AFR and multiple potential downsides. This is why I strongly recommend using lambda for tuning.

Thanks Andre and Ludo86...that does clear things up for me