[skunk]

I just started taking a stab at tuning my wife's 89 4cyl mustang converted to LA3 Turbo 4cyl. It's been probably 15 years since I did the swap and never bothered to tune it since it's completely stock.

The reason I decided to play with it is I've never been happy how it ran and the kids are wanting to get their hands on it. It always seemed run lean under boost but never verified. The tell was detonation with light boost. I wasn't too worried since she doesn't drive hard and I maintained the knock sensor and boost limited it for safe guard. Ignition timing is stock but the head has been resurfaced and it's possible the belt driven OHC is out a little.

The issue I've discovered is everything indicates it's transitioning to OL except it appears the OL fuel multiplier table which apparently modifies pulsewidth and not Lambse is not being refrenced. Meaning it is always delivering enough fuel to achieve a 14.64 AFR and no more. I dialed in Air Transfer function to compensate based on my wideband using a FTOL tune except this only Band-Aids it to whatever AFR I tuned for. In my case 11.5 AFR. So as soon as I enter the MAF/Vaf voltage range of where OL fuel should be taking over, I get the enrichment. Unfortunately that AFR is not always appropriate and could be leaner.

I have gone over the mechanical, sensors and even several ECUs with no luck. I have not gone over the wiring for the swap since it was very minor changes and all components to trigger open loop appear correct in logs. The last thing I think I need to try is to set my wideband up for NB emulation to see if it's in some sort of limp mode. I only have the stock bung to work with so I can only tune for either CL or OL separately to ensure any kind of accuracy down the low end of the Air Transfer.

So my question for the code gurus that have looked at this strategy......can you see the calls and jumps required to initiate the OL fuel multiplier table so I can try to figure out what's missing?

Based on the current definition file I can only see Load (cylinder aircharge) vs RPM. There is a TPS threshold for WOT and an associated Aircharge Clip for WOT but that doesn't necessarily mean you get OL. I will have to take a closer look but I do believe the WOT fuel multiplier is working.

Thanks guys....I look forward to any clarification or suggestions.

John

[skunk]

Did a little investigating into my cam timing today. Due to the way Ford set up the timing pointers for this engine its not easy to set up properly once you move away from stock thickness gaskets and a resurfaced head (belt driven SOHC non-interference engine). In hind sight it may have been easier to remove the water pump and align the keyways. I did verify that my cam is about a half tooth off retarded which might explain why I had to remove fuel down low and add so much up top. Not sure....speculating and trying to wrap my head around the actual affects of overly delayed valve events on a batch fired turbo application. Anyway as a test I decided to move the cam the other direction, the same half tooth advanced. It started and ran but while I was setting the ignition timing the 30 yr old tensioner gave out. I may have put a little too much pre-load on it trying to lessen the advance. So unfortunately I was not able to see if there was any change in fueling.

At this point I have a new tensioner, timing belt and adjustable cam gear on the way. I will update as things progress for anyone that is interested.

John

[oharris]

I am very interested in what you find. I'm having some weird problems with my 2.3T.

[skunk]

I should have the parts in hand sometime this week. In the meantime what sort of issues are you having?

John

[oharris]

Mine seems to run one of two ways, too rich or too lean. I have my wideband set to lambda and when it's running too rich, it seems to be at about .80 until the boost comes on and then the mixture richens to the limit of the gauge (.62 or so) and of course it stumbles out due to too much fuel. When running too lean, it sits at 1.14 which causes it to stumble when the throttle is opened until the boost comes on then it richens to about .8, where it has good power. No idea what's causing it, I replaced the O2 sensor and even upgraded it to 4 wire, but it seems to always ignore it. It seems to pick a mode upon startup and stays there until you restart it. Usually it's too rich. Last year I took it out to Minnesota (I live in Colorado) and every time I started it but once it was in the lean mode. Back in Colorado it it picks the rich mode most of the time. That makes me think it may have something to do with the BAP sensor, but no real idea.

[skunk]

I think your on the right track with the BAP since it is a big player for calculating airmass. Have you altered the Airmass Transfer?

If you want to post your tune I can have a look.....maybe fresh set of eyes will spot something.

John

[oharris]

I'm running the stock tune. I have a quarterhorse installed and I'm trying to use TunerPro with it. I was hoping to look at what the ECU is seeing, but so far I haven't found an adx file that will let me do that. I'm not having much luck getting the quarterhorse to do much of anything. I tried changing the idle speed, but nothing happened. I tried changing the injector slope, but that didn't do anything either. I'm using the T23B.xdf definition, which is supposed to work with the LA3, but so far I'm not having much luck.

Eventually I want to run SVOJohn's Speed Density tune, but I really need to figure out what's wrong with it first.

[skunk]

Update...... setting the cam properly seems to have made an improvement. A quick run at WOT seems to be rich 5-8% compared to the retarded tune.

Unfortunately it's still about 11% inflated over stock. I have to rig up a fuel pressure guage to see while I'm driving. I need to be sure it's climbing with boost since that same 11% is roughly how much larger the injectors are supposed to behave with 10# of boost on the regulator. I didn't like seeing a 43# base pressure on a stock regulator so I'm suspicious.

Oharris....this could also be something you should look at.

John

[oharris]

Thanks John!

My cam timing is stock, unless it's slipped a cog. I'll doublecheck but that doesn't explain why it runs lean sometimes. Yours was off? Do you have an adjustable cam sprocket?

[skunk]

My cam timing is stock, unless it's slipped a cog. I'll doublecheck but that doesn't explain why it runs lean sometimes. Yours was off? Do you have an adjustable cam sprocket?

Yes mine was half a tooth off. I now have the adjustment sprocket installed to compensate. If your going to verify cam timing you really need to pull the crank pulley off and align the dot on the gear with the indicator. Don't rely on the plastic cover to be accurate.

I was actually suggesting you check your fuel pressure during running conditions.

John

[oharris]

I think your on the right track with the BAP since it is a big player for calculating airmass. Have you altered the Airmass Transfer?

If you want to post your tune I can have a look.....maybe fresh set of eyes will spot something.

John

I replaced the BAP, and it runs much better! Pretty much stays at stoic now during part throttle. Still rich at idle and still gets really rich under boost, but that may be normal.

[skunk]

That's good to hear. Yes these ECUs have a heavy rich bias in closed loop. I ended up removing in the neighborhood of 10% around idle with a 43# base fuel pressure. WOT you can expect around 10.5 with the multiplier, otherwise high load part throttle should be around 11.1 above 3000 RPM.

I think I found a solution to my issue. There is a WOT Airmass Clip that was was holding my fuel back. I always wondered why the engine felt great if I eased into it vs WOT. When I maxed it out everything seemed to fall in line. I still had to add around 8% airflow up top of the transfer function over stock. I also had make a few other tweeks to timing and the BCS routine to minimize detonation issues. I'm starting to wonder how much was shaved off this head or if it's possibly a NA head causing a higher compression ratio. Im dumbfounded trying to understand how others claim to shove 20# of uncontrolled boost into this engine with the Knock Sensor disconnected AND without tuning.

I suppose in closing this thread ..... I've got the car running very well at the moment ...best it's ever ran and actually alot of fun to drive now. I may actually have to de-tune a little before I turn it over to the kids.

John

[oharris]

Im dumbfounded trying to understand how others claim to shove 20# of uncontrolled boost into this engine with the Knock Sensor disconnected AND without tuning.

I'm told that the VAM maxes out before you get to the redline, so they enriched the mixture to protect it. Adding Boost leans it out enough to run correctly. Haven't actually tested this theory, but it does make sense.

[skunk]

That theory seems opposite to what I am seeing with my engine. Mine was rich below 2.5v and lean above. At 6000 rpm I am around 3.8-4v @ ~15 psi which leaves plenty of overhead. The biggest issue I've had is detonation at peak torque 3000-4000 RPM. There is an acceleration spark table that adds up to 6 degrees on increasing RPM that's always in the mix. I've pretty much had to nullify it in that RPM range and slow the BCS to ease me past that zone. I suppose I should actually check the voltage sweep of the VAM to see if I am getting a full sweep.

Doing rough calculating.....if I was targeting 14.64 WOT the curve would have been accurate which makes sense since OL and WOT fuel is a multiplier of stoich. WOT adds another 2-5% on top of that after a certain RPM. The Airmass Clip I mentioned actually is supposed to keep things in check up to 3000 RPM where it then allows the full measured airflow up to the limit of the ECU.

John

[teal95]

My understanding is that the VAM is maxed by around 3k rpm (or nearly so) so it goes OL at that point and just runs rich. The only validation I have of this is the time I was out west with the higher speed limits my fuel economy dropped off a cliff. At 70 the car would get 30 mpg. Running 75 it was 23. Higher speed does drop the mileage but not that much. To back this up in another Thunderbird I had with a V8 T5 and 3.73 gears (which enabled much lower rpm at cruise) it was still getting 28 at 75, because it was under 3k rpm.

This is an area that could be improved by going MAF or SD since it would be able to measure the airflow and therefore keep the AFR under control even at higher mass flows.

steve

[tvrfan]

Just reread this and I have a comment to all ..... Are you sure the VAM maxes out that early ? 3k seems too early for a 4 cyl OHC airflow.

Several European engines with VAF/VAM setups ran OL ALL THE TIME as they were still leaded gas, and had no EGO at all.
They never ignore the VAM, but add fuel (via lookup table) when the throttle pot reaches defined WOT point, which is 90% full range (as I remember...). And their gas mileage was good even at 'fast' motorway/highway speeds (like 80mph, which would be over 3k even in fifth)

Only saying this as a 'sanity check' in case it might stop you going down the wrong path.
Of course with an EGO Ford might have used a smaller VAM to get low end emissions improved, but wouldn't the EGO do that too ?

note - some EU V6 engines (2.8 and 2.9 Cologne) used twin VAFs, which I guess would achieve the same thing for low end ?

Just my 2C - this could be a red herring. I'm not sure where peak airflow would be on these various engines, But I'm guessing it would be somewhere around 5K on a 6K redline ?? Um.....not for turbo ..... ??

(and I am assuming the VAM for turbo is bigger than the NA one....)

[skunk]

Its not so much that the Vam maxs out at 3k, the code throws it into OL at 3k regardless. You can play with the values to hold CL into higher RPM if you wish.

John

[sailorbob]

One of the options the LA3 ecu uses to determine when to go into open loop fuelling is a simple rpm vs air charge function together with a global air charge multiplier. It probably looks like it always goes into open loop at 3000 rpm because the air charge value at 3000 rpm is less than a sixth of the air charge value at 1500 rpm.