Ok guys so I have a questions about the EMB unit. Everywhere I go I read that the EMB can only tune for 305cc injectors. But I also have come across posts where epople got 450cc injectors to work with it. My question is, does the EMB get limited to 305cc injectors when using the yellow ***** in front? And once you use support tool you can go up to 450cc, or is that the downfall of an EMB and 305 is all it can tune for?

THE EMB CANT SUPPORT 440'S ON A 1600 regardless of how you tune it...youre limited to about 360cc whichis enough for most peoples targets,,,

yeah, the instructions say an injector 150% times the stock size, so 350cc range seems about the highest you'd want to go.

otherwise you'll have trouble manipulating the PWs enough to get it to run correctly.

if you need more power utilize an FMU to increase the pressure in the rail some, and you're now controlling 400-500cc injectors.

ok Perfect.

Well actually I have 1.8 '05.

And im not going to be using my 450's hehe those are for the RB. I got some 370's for the Miata.

I was just double checking cause I knew there was no way i can only use 305cc with the EMB ..

But thats the answer I needed. Now it's time to wire this bitch in and figure out the stupid resistor crap ...

This has all got me curious about something...

Conventional wisdom and experience tells us that the EMB can only handle injectors up to about 50% oversize. This is due to the fact that EMB compensates for injector size by trimming the AFM/MAF signal going into the ECU, rather than the INJ pulses coming out. This AFM/MAF trimming also affects the load calculation and thus the ignition timing as well, which is undesirable.

However, Bell sells the XEDE with either 440cc or 550cc injectors, and yet the XEDE trims fuel in exactly the same way as the EMB- by altering the AFM/MAF signal, rather then trimming output pulses ala EMU.

Now, in all seriousness, I cannot imagine that Bell would sell a solution to its customers that caused terrible idle problems and poor cruise performance. So from this we can infer that it is at least *possible* to accommodate 440/550 injectors on the stock injector drivers by altering the airflow signal.

Discuss.

__________________

well i data logged the car and played with the afm correction and the pulse widths stopped getting lower once the afm voltage reached around 3.9 volts at idle..any more trickery ov voltage over 3.9 volts made NO difference to the injector pulse width....that was the limit of the stock ecu's correction.....and funnily enough, the 3.9 volts came in at around the 340cc mark.
so those people who claim that 440's will work obviously have cars that defy physics

I used to preach the same as everyone else until I saw 440's running with my own eyes. I noticed that the base fuel pressure was around 27-29 psi on the stock pump on this particular car and chalked it up to that being the reason it ran so well. Most cars idle pressure is in the 35-45 range.

YMMV

Regards

I think I'm going to do an experiment. If I zero out the I/J Correction factor on my EMB, and then take all the cells in my AFM Output map and multiply them by 0.46, this should be roughly equivalent to what the Xede and others are attempting.

I'll try to find some time this weekend to give that a shot.

Well another thing that has me baffled like you is this.

My friend recently purchased a full Celica turbo kit for his car.

Craziest part is the kit comes with 410 cc injectors and a emb with a base map for them.

I know that the Celica has same size injectors as the 1.8 Miata. And his car idles like a dream.

Another possible solution is unless your running WI you have 4 spare Injector channels.

There have been a few EMB users that have fitted 4 extra injectors to the intake manifold, that way there's no real restriction on Injector sizing as you'll run both sets sequentially.


Cheers
Mark

Hey Joe,

Will you be attempting using a stock AFM or an RX-7 one?

-t.t.

Neither. I am running without an AFM presently, although the substitution map I am going to modify is based upon the stock AFM. Actually, I guess I could just pop the AFM back in temporarily- it's not hard to get in and out.

It really should not matter, actually. The procedure in question seems to be whether the voltage from the AFM can be trimmed high enough to get the ECU's injector pulses down low enough to idle a set of 440's without external injector correction, so it's fairly agnostic as to which particular AFM is involved.

This would tend to address the feasibility of running such large injectors on the EMB / SAFC, as well as answer a personal question I have as to how in the heck BEGI is getting them to work on the XEDE. (I still can't see how they've got the XEDE working on the 1.6 at all, since there's no MAP sensor and I know for a fact that the 1.6 AFM runs out of range after about 6 PSI. But that's another issue...)

__________________

Wait the SAFC can tune 440cc no problem. Especially on the 1.8 with the 280cc ones. You can tune with the SAFC until around 550cc with out any major hic ups.

I know cause I ran 550cc on my 240 which comes stock with 270cc injectors, and that was back in the day when I was getting in the tuning game and using my SAFC, BTM combo ...

Right now it's Rom tuning FTW

ive said it before.....the 1600 miatas ecu is PRIMITIVE compared to most cars of its age...back in 1989 it was probably the LAST production car ever to use that old bosch L- jetronic system which isnt very flexible.....the latter cars with the hot wire afm's tend to be way more flexible with regards to engine tuning...with the NB cars you could run 365cc EASY....
NISSAN AND HONDA ecu's were light years ahead of mazda in terms of technology

I agree, except for some reason Mazda has always had superior Timing control on their cars.

Go figure huh.

Well everyone, hurry up and sell your EMUs now before word of this gets out and the market collapses. We naysayers were wrong.

I put the AFM back into the car and switched from the Airflow Output map to the Airflow Adjust map. Then I fired it up and idled it to get a baseline:



So I see at idle I'm getting 2.4 - 2.6 ms pulses in from the ECU, and outputting 1.1 - 1.2 ms pulses to the injectors. So there's my target

Then I cleared out the injector size adjustment parameters and started fiddling with the Airflow map. It took a couple of tries to get the car to start and idle, but it did happen. Here's idle with a value of "-70" in the Airflow adjust map:



Amazing. I could tell that the ECU was struggling to maintain proper AFR- the engine went very rich every time I tapped the pedal and came back to idle, then after a few seconds managed to settle into a good mixture.

There was a point at which changing the AFM signal further produced no effect other than to stall the engine. The -70 value seemed to be right on the edge.

What's interesting is that I never did get the INJ pulsewidth down to 1.1 ms. Here I'm seeing 1.8 ms pulses, and yet the mixture is acceptable. Perhaps the ECU has a mode whereby it fires the injectors only once per cycle as opposed to twice as usual?

So I went out for a drive, and it stalled immediately. Ok, back to the map. I entered -50 into most of the cells, leaving the -70 value in a block of cells around idle. The attached .ZIP contains the whole log of that drive. Amazingly, it worked. There was some hesitation in places, and the same deal with the ECU obviously struggling to lean out the mixture, but fundamentally it worked.

I suspect that with some additional tuning it could probably be made driveable. I still prefer trimming injector pulses directly, but I'm not going to say that it's impossible any more. Only that it is non-optimal.

Attached Files

IJ out AFM adj in.zip (266.1 KB, 5 views)

__________________

Here's an example of what I was referring to by the engine "struggling" to maintain idle AFR:



This log starts just as I lift off the pedal and shift into neutral. There's the normal fuel cutoff for a moment (206.7 - 208.5) and then it goes into idle mode. At 209, the mixture is down at 11.0 : 1, and leans itself out gradually until it finally crosses stoichiometric at 219, and settles into the usual oscillation (off screen to the right). So it took about 10 seconds to work its way into a proper idle mixture, running very rich all the while.

I wasn't able to tune this phenomenon out. If I trimmed the idle cells any further, then the engine would simply stall.

So, as Adam & Jaime might say, I'm calling this plausible, but not recommended.

__________________

WOW thats some awesome info dude.

Thanks a lot for taking the time to do this. It has brought some hope to my EMB set up lol.

Either way, I hooked up my EMB and 370cc injectors, Im driving the car NA right now on that setup, and seems to be working pretty damn well....

Yes, I think that as the injector size goes down in the direction of stock, overall performance will improve. 440's seem to be in excess of the upper limit of "near-stock" performance in terms of idle mixture quality, but within the realm of "acceptable, with reservations."

550 is probably out of the question. I imagine that 370 is probably going to still be slightly outside of the "near-stock" range, but undoubtedly much closer to it than 440.