rlynchster

The 86 N/A comes with 460cc injectors leading and trailing. . .Low impendence in my case. .. .What would happen if the trailing injectors only were changed to 550CC?. . .Or heck, maybe just 500cc . . .Injector guys will make you anything you want.

My assumption is that:

1) it would go rich but give more power when the trailing kicked in.
2) I believe the smog process in CA doesn't rev the engine high enough during the test so it would still pass smog..

I know there are downsides and I was hoping somebody would talk me out of it.

I have stock everything except exhaust.

Thanks .. .

WondrousBread

1. When the secondary injectors come online at 3800 rpm, it will run a lot richer than intended due to the larger injectors. This will waste more fuel and probably decrease power, since the power you make is a function of air AND fuel.

Say you make the most power at an air/fuel ratio of 13.5:1, just for example. This means for every 13.5 units of air being aspirated, one unit of gasoline is being injected. Well, what you want is to flow more air + more fuel at the same time and make more power, leaving the ratio unchanged. Just adding more fuel just decreases the ratio, eg. 10:1.

Effectively what you're doing here is adding more fuel, but not to compensate for any increased air. So this will likely result in less power than before.

The larger injectors are only helpful if you've done other modifications that flow a lot more air and the stock injectors can't keep up. For example, if you are running massive ports and you find that you are running very high injector duty and still going lean under high load. In this case, you might see the AFRs start creeping up to 15:1 for example. This means the stock injectors simply can't flow enough to keep the ratio where you want it (13.5:1 in our example), and it's time to upgrade.

2. Assuming it doesn't cross 3800rpm, it would still pass emissions.

rlynchster

Yep. . .That's what I figured but thought I'd ask anyways.. . . It's a bad idea without air for stoichiometry.

Thanks Wonderous.

RL

rlynchster

More questions now that I know a little more.

The ECU only knows air volume/mass from two places.

The Mass Air Meter (vane type inferred from vane opening) and the O2 sensor on the exhaust manifold (single wire in my case)

O2 In and O2 out.

Are there any other ways for the engine to know how much air?

Are these the only ways for the ECU to control duty cycle on the injectors on a stock 86 N/A S4?

Thanks in advance.

WondrousBread

MAF yes. It tells the computer the volume of air based on the vane position. It does not tell it the mass of air, since that changes due to temperature induced changes in density.

O2 no. It's a single-wire narrowband type. It isn't capable of giving the ECU a specific reading like a wideband O2 sensor. All it does is under steady-state conditions tell the ECU "rich" or "lean", centered around stoich. It can't tell the ECU how rich or how lean. Then the ECU trims fuel up or down to try and maintain stoich. This only happens in 5th gear at low throttle, and the ECU will go back into open-loop mode if you step on the throttle or change gears. Under almost all conditions other than highway driving, the O2 sensor does nothing.

Not directly. It seems like the way it works (based on observation and on the training manual) is that the ECU has a fuel table based around MAF opening. This is fairly intuitive; for X amount of vane opening, inject Y amount of fuel.

Then it has correction factors. For example, coolant temp based richening. On a cold day, it applies extra fuel based on the coolant temp reading until reaching 176 degrees Fahrenheit, at which point the ECU decides the car is "warmed up" and the coolant based correction no longer applies.

Another example is that it corrects based on the air temp sensor in the MAF, which probably combines with the volume reading to give it a reading of actual air mass.

Then you also have corrections based on the MAP sensor and ambient air pressure sensor in the cabin.

The TPS seems to take care of acceleration enrichment, and also tell the ECU whether you are in a high load or low load situation. When at idle (determined based on RPM<1000, TPS at zero load, and clutch switch not depressed) the small variable resistor near the airbox applies a correction factor to the idle mixture.

Yes, in normal operation. There are also a number of switches and solenoids which affect fuel metering in other conditions. For example, if the ECU sees no continuity across the two-pin temp sensor on the radiator, it switches the orange solenoid on for 90 seconds after startup. This is to switch the fuel-pressure regulator reference from the manifold off, which causes higher than normal fuel pressure for those 90 seconds to prevent fuel boiling in the rails. Allegedly this helps with hot starting and idle stability after hot starting, but I haven't noticed a difference. This isn't strictly related to your question, but it's tangentially related so it is worth mentioning.

Speaking of tangentially related:

The Rx7 ECU is from a time where they were still working their way away from carburetors and refining what "fuel injection" meant. You can see the evolution from carbs, to mechanical injection, to throttle body injection (basically emulating a carb), to multi-point fuel injection. The Rx7 is at the tail end of this evolution but you can still see them working out the kinks in the system. This is why they included a main ECU coolant sensor, then a separate sensor for the gauge, then a switch in the radiator itself that turns off at 70 degrees F, and on some models a separate fan switch. In a modern car, there would be one sensor for the ECU and then it would pass to the gauge. Maybe a separate one for the gauge or the fan switch. On my Celica for example (2004 model), it has one sensor for the ECU and a separate fan switch. Not four sensors / switches of varying purpose.

I'm guessing having the ECU do all of that switching based on one sensor reading would have been a challenge for the processor, and so to keep costs down they used all these switches and whatnot instead. You can also see this with the intake air temp sensor in the MAF being used for the fuel map, but then a separate sensor in the manifold that operates like a switch to increase BAC duty above some temperature. Logically the ECU could do this from one sensor, but they used two.

Same goes for the thermowax being used for high-idle instead of including that in the ECU (normally the BAC would handle this based on coolant temp), the dashpot being mechanical (not unusual for the time, but easily emulated by the idle valve on newer cars), etc. All sorts of mechanical or simple electrical interventions to perform these functions.

Not directly related to your questions, but interesting nonetheless.

j9fd3s

iTrader: (3)

it is! the ECU is also hardware limited, as the computers get faster we see them loose the extra sensors.

if anyone wants to know more, https://www.rx7club.com/members/henrik-494/
was the guy that pioneered the RTek chip, which required a lot of disassembly of the stock ECU, and he posted a bunch of stuff about it.
as i recall the ecu takes the vane position and temp and comes up with a load, and then with RPM it does the math on the fly to come up with a base injection amount. then it applies any corrections, and fires the injectors.

j9fd3s

iTrader: (3)

i've done a bunch of testing on various stock FC's over the years, and i was a little surprised that the S4 NA runs full throttle AFR's that are pretty ideal. ~4000rpm its at 12.2, and then it actually goes slightly leaner (12.6 or so) at peak power and then after that it goes rich. basically its close enough to ideal that you don't really need to do anything to fuel mixtures (it did like a couple degrees of timing though)

unless

i did my testing on a stock car, IF you have more airflow (full exhaust, intake, https://www.rx7club.com/time-slips-d...update-418805/) then it would probably be a little too lean. in this case the 550's would work quite well, and even better if you had something like the S-AFC or the RTek to fine tune it.

Sideways7

iTrader: (7)

Wow, that thread is a blast from the past! Haven't thought about it in years. That said, its not super applicable since he was running a TII intake and probably some other things. I had a large streetport on my old NA, with a cone filter and full exhaust w/ header, and it ran great with a stock ECU and injectors. It might have been able to make a little more up top, but it requires something to adjust the fuel. If you throw in 550s and don't tune it, it will make it worse everywhere else.