The IPW in open loop is based mainly on load which looks up the AFR map and uses injector size to reach a final value.

Using the IPW and RPM you can then calculate the IDC. The injectors won't introduce more fuel once you get beyond about 95% IDC because they have no time to close.

The "real" size of the injector doesn't matter here, it is only the electrical duty cycle of the injector drivers.

 

Forgive the late entry into this, but I did not look very closely at the original post when the thread started. I figured it was just a missunderstanding about AFR tables. IE, I've made mods to exhaust, induction, fuel injection, etc., and I need to adjust fuel, how come the numbers in the table are so innaccurate? Answer, because the table is just a guess, the only point the ECU can really measure accurately is 14.7:1 (the typical closed loop target)...

But I just looked at the post again. I'm sorry, but 12.4 seems pretty darn lean to me. That is about .84 lambda. With the right timing, that may well be about as much power as the engine can make. Running at so-called 'Best Power' does not have to be a bad thing. In fact, I've run a turbo charged, 310 HP aircraft engine at .85-.86 lambda for many, many hours. But that was 75% power cruise. If I tried it WOT down near the ground, like in an air race, I'd never make 8 laps.

I suppose with a great oil system and especially even fuel distribution, you could run an auto engine there with boost, but it seems to me that there would typically be more heat than the engine can handle. Which leads me to wonder, is the vehicle really running that lean?

Even if we assume that the sensor is new and in good shape, there could be some things that are giving you leaner readings. Detonation is one of them, but in the absense of knock counts we can probably put that low on the list. But there are some other possibilities. For example, spark plugs are very sensitive to temperature. So sometimes mods can lead to occassional miss fires. On most current based lambda meters, even a single miss fire is reflected as a fairly long stretch of artificially lean readings.

And, of course, there are things like air leaks up stream of the sensor, hairline cracks in the sensor itself, and so on. I'm not saying that there is absolutely something wrong. But if you find that you try adjusting fueling and continue to get lean readings I would definately recommend looking deeper.

-jjf

 

Ok I was mistaken about a few things, but have been able to get the car to richen up, I just did not understand why the values in the fuel map were so off from the actual AFRs. I have changed the boost curves a bit (look at signature and have richened up a tad more) and will test later today if the weather is OK.

Also, I am seeing AFRs start around 14 @ 2200 WOT dropping steadily to 12.6 @ 3000 and then at 11.3 @ 3600 and then staying basically 11.5 up to redline (there are 2 areas that I need to tweak).

I am wondering if a real 24 / 25 peak psi and 19 psi @7000 rpm (Zeitronix boost sensor) is too much for 93 and if I'd be better off with 23/24 peak psi and 17/18 psi @7000RPM. Thoughts??

 

I thought you were on 22 PSI at red line? Dropping to 17-18 is a bit dramatic, but now your sig says 19 PSI at 7000 RPM, which for my setup is optimal.

 

JCS -- I have been testing / playing with MWGDC and boost, but think 22 @7K was way too much for 93 and the stock turbo. Do you think 19 psi @ 7K is ideal? Also, what about peak boost on 93 do you think would be ideal? Thanks...

 

If you are getting consistant quality 93 and you aren't knocking, I wouldn't be that worried about it. However, I would definately revisit the tune when you stop seeing winter blends at the pump.

-jjf

 

I see best results on 99RON (approx equivalent to about 94 PON) by running up to about 25 peak, 24 held, 21-22 at 6000, 19-20 at 7000. Depends on atmospheric pressure and other conditions what it will manage at the top. I don't have the feedback very strong to keep it smooth. I have forced it higher and maxxed the injectors in the low-mid 11s:1 (about 0.76-0.78 lambda) but had to retard the ignition excessively.

I will probably force it a bit higher again with 10% methanol in the tank although the injectors will be maxxed out, so I'll stop when I get to about 0.80 lambda and then try to get the rest of the gains from timing advance. I can't be bothered with the hassle of a methanol injection kit, needs too many safeguards/monitoring for my hassle free daily driver, but throwing in 5 litres of meth with a fill up is easy for me. I used to get about 10% increase in power and torque on the Subaru doing this, I suspect the Evo will be similar.

 

I had some fun with 10% methanol/90% V-power today (99 RON, about 94 PON?)

25 PSI boost at 6000 RPM in 4th gear, 8 degrees ignition timing, 0 knock sum, 0.78 lambda at 100% injector duty cycle.

Spec: IX FQ340, Walbro, 3" exhaust, airbox lid mod, AVC-R. I have not adjusted the actuator.

Further gains in power may come from a bit more ignition advance, usually I run 9-10 degrees at 6000 RPM at 1.5 bar on 100% V-power.

Sadly I wasn't logging the real airflow meter reading at the time. Doing so would help me to see the top end turbo efficiency.

It is about 40F with highish atmospheric pressure today. After 6000 RPM it can richen slightly to about 0.76 lambda with 100% IDC which I'm happy with on such a mild methanol mix (10%). AFRs did not lean at the top of 5th gear.

I know that running 100% IDC isn't clever, but the stock FQ360 does it with a 3 year warranty

I think the turbo would give more, the AVC-R was at about 75% duty to get this boost at 6000. I have a feeling that increasing further on such a mild methanol mix would not be wise, but it feels on the ball as it is

 

JCS -- Awesome results....I have wondered why folks here in the US don't just add Meth. to the gas, as it would seem to be safer than running a separate fuel system and having to worry about failsafes, etc....

Others -- I seem to be hitting 280 - 300 ecuload calc. from about 3000 - 5500 rpm and at ~230 @ redline...and seem to have to be 8.7 or lower in fuel mapping to achieve mid 11 AFRs, does that sound right for such basic mods (see signature)? Thanks

 

I just looked at the stock 9 Rom that I have and 8.6-8.9 seems to be a stock AFR target in the cells that you are hitting. So I am surprized that you are running stock like targets to hit 11:1. Unless ofcourse the map that I DL is not a stock map.

 

Well, one reason might be to avoid purging gas tanks. Meth is a moisture magnet and in may parts of the US will readily turn into a gelatenous goo. Another thought would be the extra intercooling effect you can get introducing the meth via the induction system.

That said, I'm surprised that it is not more common here as well.

-jjf

 

I think it was hit right on the head.. Meth is quite a bit more corrosive, and it absorbs water, two things that actually happen pretty quickly if you use methanol alot. The other reasons are consistency, an external system plumbed into the intake pipe has a pretty significant intercooling affect which is of greater benefit than using it as a fuel, and you have control over the amount of injection (as much control as these systems reasonably offer)

The truth is Methanol used to be used very frequently in cars, It was called "Dry Gas" it was used in small amounts to remove moisture from gas tanks and fuel lines.. EDIT: Nevermind, its Ethanol or some blend, one of the few car facts I got from my dad turned out to be wrong LOL

 

I ran 10% methanol mix in the gas tank on my Subaru for some time, temperature range -5 to +25 Celcius. No problems with corrosion, cold start, separation, funny AFRs but I can't guarantee others won't in other climates with different fuels.

I found I gained more power from having it in the tank than I did through an Aquamist pump. I feel that most of the benefits are in-cylinder rather than intercooling if your charge temps are under control.

It is quite popular in the UK Subaru community.

 

I guess it really depends on the quality of the gas your dealing with to start.

I honestly have never heard of anyone having problems running it, just that its difficult to keep a consistent blend in there (The same peeve I have with mixing race gas into your regular pump fuel) as the blend will always vary a bit.

10% is actually not very much Meth, but Most fuels in the USA already have 10% ethanol blended in there, so there may be no difference, or very little.. Plus that would add about 20% alcohol total (ethanol already in the gas can vary 10-30%, and methanol)

 

Most of our fuels are not blended at all, the main choice for the performance driver is 99 RON Shell V-power and it is refined that way, not based on alcohol blending from lower octane fuels or additive packages AFAIK. We have no other options available unless you want your race fuel delivered at about $40/gallon. The high budget drag racers use VP import and the like, but it is prohibitive for the average daily user.

You guys in the US don't know you're born when it comes to the ease you have with the price of cars, gas, availability of cheap parts etc. Things end up costing double in the UK once you've imported from the US - ie about 1 GBP on the dollar!