Control EGR?
 

So I plan on going full standalone with my MS3 and was curious about EGR functionality. Basically I'm curious if I can control the EGR valve like the stock computer, and if not, am I really losing anything? I was planning on going parallel to avoid issues like this, but I feel it may be more of a hassle than its worth.

I just want to make sure I keep my gas mileage the same if not better and want my girlfriend to be able to drive the car and think everything is just like stock.

Any input to this?

and you are worried about controlling emissions stuffs?

If you are thinking of using a parallel install to let the MS control the EGR, I don't think that will keep you from throwing a code. This may not be what you are getting at but i am trying to cover all the bases.

From what I read, the stock ECU will open the EGR (coils) upon deceleration and will look for a corresponding change in MAP.

If you are going complete standalone, I don't see any reason to keep the EGR, personally.

Well the only reason I'm concerned about it is gas mileage as I've always heard they may slightly increase fuel mileage when cruising.

This car will have to do an annual OBDII inspection so I already know once a year I will be required to swap the stock computer back (and injectors once I put my turbo on). All of this is for the sake of gas mileage and keeping the car's operations as close to stock as possible.


I guess I should clarify, all of this is regarding to my 2000 NB.

This is just my opinion... Any milage decrease that you see from losing the EGR will be offset by your engine management.

Do the local emissions control people check for the turbo itself or do they just look for an OBDII and see that it's happy?

According to some tuners I know up here near Chicago, they ignore all aftermarket mods as long as the thing passes OBDII.

If you are willing to do the work up front you may want to do a parallel install like what I am working on. Its a pain in the butt, but once the install is in an debugged, I am hoping to never have to remove anything, testing or no. (though I will unplug my gauges and throw a block off plate in my dash when I go test)

EGR helps fuel economy AND reduces knock. but the MS3 can't control it.

Yet?

How does an ECU decide (for lack of a better term) when to open the EGR? What inputs would you need for an ECU to control an EGR effectively?

I am nowhere near smart enough to program MS myself, but I am interested in how it works and how it all works together.

Years ago I yanked the EGR out of a car I was building and it never ran as well after. Mind you, that was in the days before MS but I am in no hurry to loose EGR if I can keep it with a reasonable amount of work.

It opens it when you're cruising. And that's really about it. EGR at idle will stall the car (or make it run like shit), and EGR at WOT makes no sense.

My MPG's went up so much during cruise with MS, that it would have made any losses from ditching EGR irrelevant. EGR's main purpose is to lower NOx emissions. Not to raise your gas mileage. That's just an occasional side effect.

If you notice in my sig, I've already done a parallel install on my 97. I'm fully aware of the work involved but I've come to the realization after doing my 97 that it truly isn't worth the effort for that once a year inspection. Everything has to be just right for the stock computer to not throw a CEL with Megasquirt in the picture. It truly would be easier to just swap injectors and wire the wastegate open for that one time a year. When I first did my parallel install I thought I was saving time by letting the stock ECU do idle, A/C, EGR, etc. but in reality, I still had a constant CEL due to my missing MAF, primary O2 sensor running too rich (in boost), and the headache of sharing sensors and attempting to get the bias resistance correct, etc. I wish you the best of luck doing your parallel install, and hope everything works out great.

I just want my car to be just like stock in functionality and if others drive my car, I would want them to not even know the stock computer has been swapped.

I'm curious though, does anyone know if the computer is smart enough to realize if the EGR tube has been blocked off, leaving just the functioning valve?

If that's the case could you just have it activate EGO based on a MAP (or even TPS) window as an aux output? Just set WLED if MAP > 35 and < 70 or the like?

i'd start with at least MAP AND RPM. otherwise you might activate it at idle.

probably 2000 < active < 4000

I knew/know about the former, but not the latter.

How does EGR effect knock propensity? I was under the impression EGR was only activated during cruise conditions.

You can knock in cruise. I suppose there's an argument that could be made about using EGR in cruise to reduce the chance of auto ignition by reducing the amount of particles that could auto ignite. Not something I've tried.

There's no special code written for EGR control, but I don't see why you couldn't use one of the spare outputs. I haven't checked if the Miata's EGR system is on/off or PWM.

I'd like for my EGR to function too. I'm running parallel and thought that would keep it together, but it turns out it ain't so. As I understand it, the EGR needs input from the MAF (which I no longer have). This also causes the stock computer to throw a code, even on OBDI.

It is a dual-PWM valve on the 94-97 and a a 4-wire stepper on the 99-05.

I wonder if it runs closed loop (EGR flow is metered), or open loop and the ECU just checks if its working within certain parameters.

Anyone know if the stock computer is smart enough to realize the egr pipe would be sealed shut and not feeding any exhaust to the intake? (Leaving everything else functional)

AntiLag

Use EGR system to push air into the exhaust manifold

The control system changed over the years, though I'm pretty certain that none of them "metered" the flow of gas with any degree of precision.

The '94-'97 valves had a potentiometer-type sensor on the valve body to judge the position of the valve. From '96 onward, they used an "EGR Boost Sensor" (essentially a MAP sensor connected to a chamber within the intake manifold casting where pressure can be observed to vary with EGR flow.) '94-'97 had both.

I can't find any solid descriptions of the control system, however on the NB there are two EGR-specific codes, P0401 and P0402, for "EGR flow insufficient" and "EGR flow excessive", so it's obviously capable of recognizing at least three distinct states: too little, just right, too much. This at least suggests that the system could be operated in closed-loop mode. (It could also be that the "just right" state is deduced rather than observed, by looking for predictable transitions between "too much" and "too little" in the same manner as is done with a narrowband O2 sensor.)

What puzzles me slightly is that there is a two-way solenoid valve between the pressure sensor and the manifold. The only thing I can guess is that it allows the system to expose the sensor to ambient pressure periodically as a calibration reference, possibly to ensure that the sensor returns a reading approximately equal to barometric pressure as a test on the sensor itself.


See above.

This is correct and not at all uncommon in OEM controls. It's also sometimes a source of problems, as some poorly tuned ECUs *cough MAZDA cough* look for a new sample too quickly after changing state of the solenoid, causing a momentarily incorrect reading and resultant bog or hiccup...

I doubt the OBDII computer is controlling EGR flow in closed loop. I don't even think it's looking for the pouridge to be just right. I think it's just looking for evidence of flow when the valve is open and for evidence of no flow when the valve is supposed to be closed. This may be more sophisticated in newer cars, but I don't see how just a single analog map sensor can be used to accurately meter EGR flow other than detect on and off.