markusparkus

iTrader: (1)

I've been looking at a lot of pics of mod'd turbo builds and I finally noticed that a lot of them have the air filter connected directly to the turbo inlet and that the MAF is missing. Is this because you guys are using an aftermarket ecu and a wideband O2 sensor and then tuning the fuel maps based upon RPMs?

RotaryRocket88

iTrader: (7)

A standalone EMS allows the AFM to be removed, since the tuning is based on RPM and MAP signals.

gear_grinder

iTrader: (4)

my maf is on my 97 ford explorer, i dont run a maf on my near stock TII....

I do however run a AFM, i have a 45 degree elbow off of the turbo then some duct tubing from the inlet of the AFM to the filter.

arghx

iTrader: (3)

the Rx-7's use volume airflow sensors, not hotwire (Rx-8, 240sx, etc) or hot film MAF (Corvette) sensors

Evil Aviator

Yes, a well-built car will have an aftermarket ECU (Standalone EMS) rather than the old 1980's technology stock ECU. Most boosted engines with a standalone EMS are tuned in speed-density mode, which uses engine rpm vs. manifold pressure for the base maps. O2 sensors are primarily for tuning economy (cruise and idle), although they can also be used to speed up the tuning process, or as a general indirect guide to hopefully avoid detonation.

Also, the mass airflow sensor array on the 2Gen RX-7 is called an Air Flow Meter (AFM). Most people on this forum don't know that the AFM is a type of mass airflow sensor array, so they will be confused if you use the MAF sensor term that usually applies to a hot wire/film or cold wire type of sensor found on more modern engines. It would be similar to asking about the stock supercharger, which would confuse the masses who don't realize that a turbo is a type of supercharger.

markusparkus

iTrader: (1)

Thanks for the clarification on terms.

I am still surprised by this though. It seems like the tuning is just done using inputs, and no output measurement. It seems like variance in some of the inputs could have a significant impact: injector size, fuel pressure, clean vs dirty injectors, timing etc. What is the feedback for adjusting some of these parameters?

I'll happily take an url to a faq or readme about this stuff as an answer

jjwalker

Nothing does in speed density, except the 02 sensor. You have to tune it correctly with a wide band then be good about maintenance.

arghx

iTrader: (3)

Are you talking about on systems with standalones and speed density conversion? Because you are somewhat right there. Usually with an aftermarket computer the "learning" capability is pretty limited. If you look at the factory speed density computers on 90s and later Hondas for example they have long term fuel trims and other ways to adapt to changing ambient and engine conditions. Aftermarket computers don't do this so well. The stock Rx-7 computers are not learning computers either because they are so old.

For the stock ECU ("Australia" means normal USDM/JDM type cars):





search for the factory training manual, that's where those diagrams came from.

markusparkus

iTrader: (1)

Hmmm....

I guess what's still confusing is this: I've read where people are trying to tune so that they are leaning out the A/F mix so that only the fuel that get's burned during combustion is being injected. I guess you might call it perfectly lean, or some such. However, since leaning out too much, especially at high rpms is a major cause of engine death through detonation, this process seems dangerous. That's why I was assuming that there must be some sort of output measurement.

I'm guessing that what people are doing is putting the car on a dyno, measuring the HP/RPM curve and then tuning the A/F maps (leaning out, since you probably start with a rich/safe map) and then doing another dyno run. I suppose when you you start to see a loss of power from your tuning, implying that there's not enough fuel for the air charge, then you've found your razor's edge and then take a tiny step to the rich side.

Am I totally wrong?

sharingan 19

iTrader: (16)

Actually, that's pretty much the gist of it...

The "output measurement" you are referring to in most cases comes from a Wideband 02 sensor. Some tune exclusively using EGT senors, but I haven't seen man posting about that these days.

It's pretty much been collectively agreed upon that 11.0 - 11.5 afr is a pretty good compromise of safety and power (excludeing ridiculous boost and/or timing) . So typically most tune to achieve those numbers (or slightly richer) and then take it to the dyno and see if there are any gains to be had by pulling some fuel or advancing timing or whatever little trick the tuner chooses to use. The dyno is importtant because while you can tune 1st 2nd and maybe 3rd gear on the street, full 4th or 5th gear pulls simply aren't gonna happen so those areas of the map tend to stay pretty rich until they can be tuned in a controlled environment.

Detonation is the death of rotaries and while it seems that most often this occurs from running too lean (pre-ignition of the air/fuel mixture) it can also happen from running too rich, although this seems to be the minority of cases

j9fd3s

iTrader: (3)

the rx7 is a pretty well known combo, we know if you take a stock t2, put a big exhaust on it, we have to add more fuel.

with something more adventurous its safer to start rich than it is to start lean

new2mazda

iTrader: (1)

where those measurements in lambda?

j9fd3s

iTrader: (3)

the 11:1? that would be air to fuel, 11pound of air to 1 pound of fuel.

lambda is actually probably slightly better to tune with, as gasoline varies, and some of us use alcohol too, stoich on every fuel is 1 lambda, but the stoich afrs are different for different fuels.

WingsofWar

in response to the thread topic, anyone using blow by setup?

new2mazda

iTrader: (1)

i just worry because i know im running lean at top end but my aem has be avg runnin around 12-14 at reg driving... that horrible?

john ward

iTrader: (7)

12 being lean?^ ummm

new2mazda

iTrader: (1)

on my aem wideband it says lean is 16

WingsofWar

for rotaries 12:1 is fairly lean even if its richer than stoich.

new2mazda

iTrader: (1)

the minimum is 11 on my scale... so unless ur talking in lambda that means my wideband isnt supposed to be able to measure what my car runs?

RotaryRocket88

iTrader: (7)

You're all making this too complicated. 14.7 AFR is the stoichiometric ratio for gasoline. Anything lower than that is "rich", and anything higher is "lean". But these just relative terms and different conditions mean different target AFRs. Mxiture at WOT is very different from what you'll see during cruising, idle, etc...

j9fd3s

iTrader: (3)

different conditions require different mixtures.

for example the stock ecu has different zones, these zones have different calibrations, as some are accelerating, some are decelerating, steady state, etc

idle: without the airpump is like mid 12's, mazda runs it rich so it does not misifre, and then adds air pump air for the cats. with the airpump air afr will be in the 16-18:1 range

cruising; if you're just running a constant speed, load is low, fuel economy is important, so it runs 14.7:1

acceleration; load is high, we don't care about MPG, we want power, and it has to not blow up on crap fuel, so the mixture is rich. the leanest stocker, the s4 na, runs mid 12's, everything else runs richer