Anyone have experience with scaling the injector constant in the Motronic. I'm putting in larger injectors for my S2 Supercharger project and only needed to scale these 90% even though they are supposed to flow >25% over the S2 injectors. Seems odd to me.

See the SC thread here:

https://rennlist.com/forums/944-and-...ld-thread.html

I have adjusted the injector lags, but perhaps the additional 6 ohm ballast resistors the S2 use meaning the turbo injectors have about 10.5 ohm R total and the S2 8 ohm makes a difference. I'm using the brown tops (Ford Mustang SVO) which are supposed to be 36 lb/hr vs the S2 29 lb/hr.

I'd welcome anyone's experience with swapping in larger injectors.

Cheers

 

Bump

 

...............

 

Eric, Your best bet is talking to josh about this or a specific forum for the software.

Sorry i cannot help

 

What are you using to tune with?


And, Josh would probably be the best person to talk to about this....

If it helps though, I have 90lb injectors in my NA car (I needed new injectors, so instead of buying a set for the NA motor, then another set for my upcoming turbo motor, I just bought the big ones and am using them on my NA...), with a 3bar fpr, and tune with TunerPro.

Setting the scaling is kind of trial and error. Bumping the dead time one direction and the scale the other, until everything lines up the way its supposed to.

 

Using Tunerpro RT and Ostrich. I've mucked around with the dead time as well.

Surprised you are getting reasonable idle with 90# injectors?

 

Runs great on Gas and e85.

I'm not sure how the S2 differs from the 2.5 DME, but it really was just trial and error.

The dead time is based on engine voltage, and mostly effects the idle/lower portion of the PT map. The FQS scale is mostly the upper portion of the PT map, and the WOT map.... so its a balancing act to get the AFRs on tuner pro to match up with the "real" (WBO2 gauge) AFRs.

The dead time was the most difficult for me to set. I'm at 60 and 32 for my injectors. Yours will obviously be different, but don't be afraid if the numbers get pretty spread out. I know some injectors call for ~100 and 25... Set the large number with as little load on the electrical system as possible, then the small number with high load (headlights/brights, radio, AC, wipers...everything on), and tune to the idle map.

Then just move the FQS scaling around until you're happy with the actual WBO2 AFRs.... It took me a few long days of driving to get it to where I was happy. Bump the scale a few numbers, then the dead time would go off, so you bump that, and the scale moves. Its a balancing act that you'll eventually zero in on.

It also doesn't help if you have a funky AFM. My AFRs bounce around at idle, and its because of my AFM. Running a test SD tune from Josh, and my idle AFRs are rock solid. So, sometimes you also have to just accept a small discrepancy from the 25 year old equipment.

 

Did some testing tonight and here are some observations:

Previously my fqs factor had to equal 0.9 to get WOT AFRs right (match stock injectors).
I disconnected the battery to clear the trim tables in the ram and then ran the engine in O2 off mode (no cat coding switch jumpered).

Immediately it ran rich. Setting the FQS back to about 0.82 seemed to clear this and get the AFRs at PT back near 14.7. BUT, the FQS switch has zero effect on the idle AFR. I then unjumpered the coding plug, but disconnected the NBO2 in. AFRs remained similar. To get the idle fuel right, I needed to factor the idle fuel maps by about 0.8. Seems like the FQS does not scale idle fuel maps.

I then reconnected the O2. At 0.82 the WOT maps were a little rich still, so I needed to scale the FQS back to 0.80.

So it appears that the Motronic not only adapts the closed loop values, but uses this information to also factor the WOT maps. As I could get the same AFRs using 0.8 or 0.9 scaling factor (O2 connected) - the DME just compensated the WOT fuel trim based on what it needed to do to compensate the PT trim.

One more test to clear the trim again with the 0.8 value set, disconnect the O2 and see if the AFRs remain as tuned with and without the O2 connected.

Quite a time consuming process, and this is before I get the supercharger installed. I figure it's easier to get this right now than have to compensate for boost and injector upgrade later.

 

Read my post above... The dead time values are better for the idle afr adjustment.

You don't need to unhook the battery to clear the DME. It resets every time you turn the key off.

Also, how are you running the O2 in "off mode"? Just unhooking it, or turning it "off" in TunerPro?

Its very hard to tune anything with the NBO2 hooked up, because the DME will always try to correct the AFRs using the O2 sensor. Leave the NB in open loop, tune EVERYTHING how you want it, then when you're happy, turn the NB back on.

 

I did adjust the dead time values, but I still couldn't get the AFR right. Unhooking the battery resets the trim values in the RAM. Simply restarting the engine does not clear these values. My point is that if you don not disconnect power, the DME remembers the previous trim adjustments.

NBO2 off either by disconnecting or running in the "no cat" maps of the DME using the DME coding plug.

I understand the need to tune with it disconnected, my point is that you need to clear the ram each time you adjust the maps or it remembers the previous closed loop sessions trim values. So yes you need to get everything right in open loop first, but you need to disconnect the battery before doing this.

 

I built a Eaton SC'd 85 944 in 1999. I used a Corky Bell rising rate fuel pressure regulator and it was spot on after about 45 minutes of trimming the rise gain. I know its old school as heck- but it worked 110%.

Ive always had issues trying to work within an OE ECU that was not designed to measure boost pressure- you always pay a bad price somewhere- idle, mid transitions, WOT, where ever it may be. The RRFPR is great since it only raises FP when boost is present- (or at atmos or sooner for accel pump effect). Once you get the gain close (amount of FP per PSI of boost) the OE ECU can use its normal injector on times. The car will idle perfectly, transition well and have load of fuel on boost, WOT etc.

Just my .02

 

Well, you can "turn off" the O2 sensor with tunerpro, so you don't need to mess with the coding plug. And, you don't want to just unhook the sensor, because that actually sends a "lean" signal to the DME, so it will correct for that.

I guess I don't understand what problems you are having....? Are you trying to get the map values to correspond with the actual AFRs on your WB gauge? Or is there an actual problem with your tuning process? (Or, maybe I can't read? )

I didn't do anything, or go through anything you are describing when tuning my car.

Set the "Closed-loop RPM limit" to 400rpm - This effectively keeps the DME from ever reading the O2. There is no need to move the coding plug in the DME, or unplug the sensor.

I also set the "Altitude Fuel Correction" to 0. I found that it would make my AFRs jump around, and set at 0, they are consistent no matter what altitude I'm at.

Don't tune anything until your car is at normal operating temp, otherwise the warm-up enrichment will change your AFRs.

My map AFR values aren't perfect either. I think my idle maps are about .5 point above my actual AFRs. Meaning, in TunerPro, I set my idle AFRs to 16:1, and the engine is actually running 15.5:1 via the WBO2.... That's what I mean by "compromise". As long as you tune the engine to the actual WBO2 values, the numbers in TunerPro are kind of irrelevant (Within reason, of course).

It took me a few long days to get everything to where I was happy with it, so don't get discouraged when your tuning for one night isn't coming out perfect. Weather changes, engine temp changes, what road you're driving on, will all effect your tune.

 

Mark, Eric has an S2... so the code/DME/software is a bit different.

 

The S2 has a leter generation Motronic M2.1 which are a step beyond the ML series used in the 951. The Lambda control and adaptive measures seem a step above the ML3.1.

I thought I had the closed loop rpm and load limits sorted, but it seems there is more at play in the algorithm for this than just a single value.

In any case, I think I now have the FQS factor at a point that makes sense.

As you say it takes time. I'll need to go away and tinker some more.

Thanks for the advice.

 

I thought it might be, but you weren't answering the thread

I had a feeling that's why I was lost with what Eric was saying....