EFI Tuning Philosophy

[Foghorn]

I've completed the build of my manual fuel table. I'm looking for a sanity check on my method moreso than the actual numbers, but if I can figure out how to attach my Global File(s) to a forum post, I'll share them for further input.

Considerations:
1) Exhaust system is side-pipes with a slip-fit collector and about 38" after the sensor. The "mufflers" are straight-through, but not a single large pipe, but rather three smaller tubes arranged in a triangle and wrapped in 'glass. The point of this is there *should* be a small restriction relative to a straight pipe with no muffler. I'm using two WBO2 sensors with the ECU averaging the readings. I expect a false lean at idle due to the slip-fit (no reasonable way to seal them, see my second post in this thread), but once off-idle I expect that the exhaust would be pushing out of the slip fit more than sucking air in, which is to say as the RPM increases so should the accuracy of the AFR reading. BUT...if the collectors continue to leak air in, then the ECU will compensate by adding fuel which is much better than pulling it out (WFO & lean = BAD).

2) The engine will idle with the ECU in Closed Loop, but it's way rich with the Target AFR set at 14.1. And when I say rich, it fouls the plugs after about 15 minutes. I considered just leaning the Target AFR in Closed Loop mode until the plugs looked like they should, but I'm concerned about how the ECU might behave with too much "Kentucky windage" to hit the target.

Background: I have about 7000 miles on the engine with a carb on it, to include a 1500 mile road trip over the course of a week. Based on this experience, the engine would burn about 5 gallons per hour cruising at ~ 3000 RPM and ~ 9 inHg vacuum (I was getting about 15 MPG at 75 MPH). The timing curve was 18° base/20° mechanical all-in by 2500 RPM and another 9° vacuum advance (~ 1° per inch of vacuum). The plugs were shiny black, but not sooty, so it was a little fat, but fat is safe.

Method: I used 6.3 lbs per gallon as my basis for converting gallons per hour to pounds per hour when I populated my Base Fuel Table. Starting with the cell that corresponded to 3000 RPM & 70 kPa (~20.6 inHg or ~ 9 inHg vacuum). I set the fuel rate at 32.5 lbs per hour. I then used one of Holley's pre-configured tunes in the V5 build 141 software to develop the rate of change for the fuel "curve" in both the X & Y axes. Using that Rate of Change and using my known value as a starting point, populated the rest of the table.

FWIW, the fuel table in my start-up "tune" is programming 12.6 lbs per hour of fuel at idle for a 14.1 target AFR. That's 2 gallons per hour of idling (!!!) It is a 482FE, but I think that is a bit excessive. Any thoughts?

Pending any constructive feedback I receive, I'm going to load the tune, warm it up and let it idle for 10 minutes (new plugs, of course) and see what the plugs look like. Adjust from there. When I get the plugs happy, I'll note the actual AFR. Set that as my Target AFR, limit the CL Comp and try Closed Loop just out of morbid curiosity. I'll post how it turns out.

Any constructive feedback is welcomed. (I already have somebody telling me I'm a dumbass for removing the carb, thank you very much).

[andrewb70]

@Foghorn: If you give me more information about your engine combination, that would be helpful. Once you zip your Global File, you can attach it. Andrew

[AndyF]

Yes, 12.6 lbs/hr at idle is too much. Start with 9 or 10 lbs/hr at idle and go from there. If you have a laptop hooked up to the ECU you can do this very quickly in the car. You set the idle area at 9 lbs/hr and then fire the car up. Have the Base Fuel Table open in your laptop and your mouse at the ready. I assume you have the engine in Open Loop. Once the car fires up you highlight a 3x3 or 4x4 area on the Base Fuel Table around the bouncing ball and then use the up & down arrow keys to add or subtract fuel. You should be able to quickly find a spot where the engine is happy. It works best in warmer temps. When it's super cold you'll be fighting with the enrichment curve at the same time. It also works better if the timing is close to correct. For a performance engine you should start off with at least 15° at idle.

[Foghorn]

@Andrewb70 – Thanks for the info on zipping files.
@AndyF – Thanks for the advice.

The engine is a Ford FE stroker. 482” (4.25 x 4.25). Flat top pistons with Blair Patrick aluminum heads (~ 10.5:1 compression). Solid roller cam: 264° intake, 270° exhaust on a 110° center. .535” lift. Blue Thunder high-rise dual plane intake modified for port fuel injection with Holley’s 1000 cfm throttle body. Exhaust as previously mentioned is a pair of side pipes with a slip-fit collector. I don’t know the primary tube diameter without going out to measure, but they’re probably 1 7/8” or 2” and at least 30” long. They are individual tubes that have no joints before the collector. 3” collector.

Update:
I started it on my manual fuel table (out of coincidence I had 9 lbs/hr programmed at idle). Ambient temp was 60°F. It started, but wasn’t happy so I kept it running with the throttle while I increased the AE vs. CTS in the After Start Enrichment table. It was happy with a ~ 18% increase from 107% to 125% at 60°F, which considering I reduced the Base Fuel Table by 40% made sense.

As it warmed up and the enrichment decayed, it stayed happy so I considered that a win. I let it idle for about 10 minutes with no throttle input until the coolant temp got up to 160°F. During this period, I noted the real-time AFR readings which were fluctuating between 17.5 – 19.5:1.

After a couple of hours of cool down, I pulled the plugs and they were all dark gray. Still rich, but at least I can see the porcelain now!

I then started studying Holley’s base calibrations in the V5 build 141 software - I settled on MPI48BB245, which is a 454 Chevy with a 245° duration cam. Pretty close to my setup. I should have started with this one as my base cal when I first started the engine. Live & learn.
With that, I copied the Target AFR table into my Global File & modified it by changing the idle area from 14:1 to 18:1 to match my data (verified by the plugs) and extrapolating from there. I have started it in Closed Loop mode, idled it and revved it up several times. I need to tweak the AE tables. I haven’t read the plugs from this run yet nor taken it for a drive. More to follow.

I do have a question on how the AE tables interact with each other. Which table takes precedence over the others? Does the ECU work through the tables in a particular order? Or does the ECU look at all simultaneously and program the extra fuel based on which table asks for most % increase?

Again, if you can offer any insights or see something that I'm overlooking, please advise.

[andrewb70]

What's the MAP kPa at idle?

[AndyF]

What you're doing with the Target AFR is not a good idea. You need to have it idle in Open Loop and then go into Closed Loop at 1500 or 2000 RPM. Manually tune the Base Fuel Table below your Closed Loop start point. If it idles okay at 9 lbs/hr then you're close. Just follow the process I gave you before. Get the engine warm and manually adjust the Base Fuel Table while the engine is running. Pay attention to the MAP and the IAC while you do this.

[Foghorn]

What's the MAP kPa at idle?

78-80 kPa, idling at 900 RPM. IAC starts high, but gets down < 20% as the Startup Enrichment decays.

AndyF: I know that I'm flirting with disaster with my manipulating the Target AFR. When I was in Open Loop, manually programming 9 lbs/hr at idle, I monitored the real-time AFR (my previous post details this), so there's a method to my madness, and I know it's risky. I can't just keep fouling plugs and hope it gets better. I have done a smoke test on the side pipe collector. If there's a leak, it isn't pulling in the smoke with any obvious effect. I'll attach a couple of pics of the collectors.

Other than that, I find it hard to believe I have two malfunctioning WBO2 sensors that would provide bad feedback to the ECU. That leaves my only real avenue to apply some sort of scientific method and tune to my specific application as best I can. I can provide a safety factor by limiting how much fuel the ECU can pull out in Closed Loop mode in the Learned Compensation Limits table to protect the engine.
Here are two pics of the spark plugs: This first one is from the startup tune, running completely in Closed Loop. (I'll attach the Global Files as well if anybody is curious.)

It fouled while out on a short drive after an extended idle/AE tuning. Total run time was about 20 minutes. This is the second set of plugs I have fouled.
This second pic is from the manual fuel modified AFR tune:

As you can see, it's still very rich, but at least it's not sooty & fouled. This plug has about 10 minutes of idling with a few stabs of the throttle in Closed Loop mode. I'll attach the Global File for it as well.
Since I don't have a warm fuzzy about attaching files, I'll attach them in a follow-on post, so I don't lose all I have typed thus far.

Here are the two Global Files that correspond to the pics of the spark plugs in my previous post. V5B171 482bbf 66lb NA Startup.zip, V5B171 482bbf 66lb NA manual fuel modified AFR table.zip
Thanks to all who have taken the time to assist me - AndyF, Andrewb70, Danny Cabral & Rich (EFIExpert).

Based on this article, I don't think I have to worry about a false-rich situation: https://www.hotrod.com/articles/exha...nsor-readings/. Any feedback is appreciated.

[AndyF]

You should probably just ask for help from a pro. Contact Rich at FAST Man EFI for a remote tuning session. I don't know what your issue is, it usually only takes a few minutes to tune an engine in Open Loop. You just fire it up and use the arrow keys to go richer or leaner. Same with the timing, advance or retard until you have it happy. I've tuned a bunch of different EFI race engines and have never fouled a set of plugs like that.

[andrewb70]

With that kind of low vacuum at idle, your in race car territory. Does your camshaft have a ton of overlap?

[Foghorn]

AndyF: I have talked to Rich. Very knowledgeable and most of all, patient. I'm trying to solve this without wearing out my welcome. ��

Andrewb70 - I don’t have the cam card in front of me, but considering the LSA & that it doesn’t really get up on the cam until ~2800 RPM, and the low vacuum, I’d say it does. If I had it to do over, I would change several aspects of the cam, but that ship has sailed. If I absolutely can’t get it dialed in, as a last resort, I’ll pull the engine and install a milder cam - but I’m not ready to throw in the towel yet.

I took the car for a drive yesterday. Weather was clear and cool, with the ambient temp around 60°F, MAP reading engine off ~ 97 kPa (which now that I’m thinking about it, seems a bit low. - I didn’t think to compare that to the local barometer). Anyways, it started and idled rough while I continued tweaking the After-Start Enrichment vs. CTS.

After it warmed up, I drove it for about 15 miles in varying regimes (2D gear country road, ~ 25 MPH & 2500 RPM, low throttle opening to interstate, ~ 70 MPH & 2800, cruise throttle). I did a couple of part-throttle accelerations up to 5000, but no full throttle pulls. A couple of times in 4th gear, I let the car slow and then part throttle accelerate from 1500 RPM. Engine seemed happy, with a bit of a stumble between 2400 a 2800 RPM, and then it would pull harder even though I wasn’t changing the rate at which I was opening the throttle. Occasionally it'd pop through the exhaust almost like it was loading up.

The plugs didn’t look significantly different that the second pics in my previous post. I'm running NGK BKR6E plugs. Could the heat range be too cold?

I have the system ICF to read both WBO2s and average them. Could I have one bad sensor that is skewing the results?

This will be my last update for a couple of weeks (I had to go back to work). But I would appreciate any ideas that would get my Target AFRs back where they are supposed to be. I’d hate to drop $$$$ on new WBO2s and a couple of HBX-1s only to have no change.