Mustang5L5’s progress thread - Boom

[Mustang5L5]

So, going way rich indicates that you have TOO MUCH 'accelerator pump' tip in. That spurt only lasts for a very brief time period, and then AE & close-loop correction takes over.

My best guess, and caveat first that you're right about getting the VE tables right, is that your initial tip-in is too much causing closed-loop corrections to pull it lean, before it finally re-corrects back to commanded AFR. Two ways to test my theory: pull tip-in out, or command open loop and see if it still happens.

You may very well be onto something here. My instinct was to add fuel to the FE table, but going through some old datalogs where I used autotune, i find it was pulling fuel from some of the cells immediately off-idle. So i do think i need to play with the VE table a bit before even thinking of messing with AE.

Oh, and on your injector issue, my first thought goes to the connectors/adaptors.

Thought has crossed my mind. While i had waited for the car to cool down before restarting it, i did run the injector test a few times on my 30# injectors in one cyl #1 just plugging them in loosly and had one injector not fire until I really seated it snuggly. Entirely possible this is a connection issue.

Quick thought on the injector issue; if you want to repeat it with the 3rd gear WOT pulls, first move the suspect injector to cylinder 2. Then if it happens again, you have the source isolated. I know it's a huge PIA to move the injector, but it would provide a definitive answer.

Did your pulse widths ever go static when calculated against rpm (e.g. 20 ms @ 6k rpm)? That could be the source of a driver over-heat, assuming that the driver stage has over-current / temp protection built in. If I have time, I'm going to try and open your logs at work tomorrow. I should be able to plot it and check if it's just an ASCII based data log.

Are your 30lb/hr injectors EV1 or EV6? If I remember correctly, the EV6s are slightly higher resistance (like +1 or 2 ohms) so they should pull less current / easier on the driver.

Good idea on the swapping. I'll have to wait and see if the issue happens again. I did toss a spare 24# injector i have in the glovebox so that if it does happen again I can test it to see if the ECU is firing any injector on that spot. Thankfully Cyl 1 is easy to get at....however that means it's the injector plug i've most frequently unplugged and plugged back in for various reasons.

PW getting high is a concern, which is why i think I'm underinjected. Hitting 18ms at about 3K RPM, so 45% duty cycle at 3K. Target AFR in that region is about 12.6

The 30's are EV1's. Been considering purchasing new injectors but haven't decided either way yet.

 

[Mustang5L5]

I'm interpretting 'dead time' as injector opening time.

Delay from when the injector gets voltage to open an when fuel starts to actually spray

What the mega tune is showing in microseconds differs pretty significantly from what binary editor shows in the A9L.bin file (in milliseconds):

But what I'm not seeing in what you've posted is the high & low slopes, which is how the strategies in the EEC controls them:

Then, the EEC uses something called a breakover point, which I believe is the time after opening from which the injector acts as a high-slope sized injector to the time it acts as a low-slope injector. I believe this primarily matters at low pulse width where if your computer is demanding very low duty-cycle, the injector will act as though it's much bigger than its nominal flow rating. The break point in the A9L.bin for the stock 19 lbs injectors for all voltages is 0.000030398 lbs/rev. I'm not sure those units make sense, but that's the column header in the table I'm looking at. According to a post on EecTuning.org, "The injector break point is the pulse width value where the EEC changes the way that it addresses the injector. Above the injector break point, it goes to higher slopes, below the injector break point, it treats the injector at the lower slope." I thought the units were supposed to be in time. lbs/rev doesn't add up to that, to me.

I popped open my Big Stuff 3 software (I'm still using the original stuff that didn't require a subscription) & it doesn't have any of the hi or low slope data for the injectors, either... It just seems to have the nominal value for flow and the I/O times. I guess the rest of it is interpolated from some base assumption within the program, or maybe it's just not as important as the EEC makes it out to be.

Anyways, if I were in your shoes, I would use the A9L.bin data for the I/O times. That's a very well-vetted tune. Just need to convert the milli to the micro.

Why did you set it to 200cc? I can see that making the most sense if you're trying to pick a point between the low & high slopes. Just curious about your reasoning.

Tuner studio doesn't have an apples to apples way of inputing the A9L data. I can set the dead-time, a pulsewidth curve and a timing table, but that's about it really.

 

[Mustang5L5]

Something else i need to work on is why my VSS is not registering. No input at all when the wheels spin.

I suspect the issue is polarity. VR sensors go have a polarity. When i converted my car to mass air, i ran two wires from the VSS to the ECU and just connected them to the pin 3 and 6 without paying attention to polarity. They probably need to be switched...

With the MS3, i suspect it may matter. I'm going to test this on a bench with my old MS3X box and spare VSS.

I'd like the VSS work so i can log speed, gear, and it will trigger my e-fan off at a certain MPH.

 

[FastDriver]

Sounds like you might be onto something with "pulsewidth curve" If I'm right about my understanding of Hi, Low, & breakover, then you should be able to use those to input that data. Would help to see that table, unless I missed it in what you already sent. I'm probably gonna have to think through the 2-squirt per cycle thing I saw in your first screenshots.

VSS from the stock T5 is sinusoidal/analog, whereas the MS may need digital... I'm probably not using the right terms, but I know that's your field and you'll get it. Anyways, if that hypothesis is correct, then you can use a dakota digital converter like I have in black jack, I think. I go from digital to analog, but I think it'll run the other way, too. Summit part # DAK-SGI-100BT. Don't have time to dig into it right this moment.

 

[Mustang5L5]

Sounds like you might be onto something with "pulsewidth curve" If I'm right about my understanding of Hi, Low, & breakover, then you should be able to use those to input that data. Would help to see that table, unless I missed it in what you already sent. I'm probably gonna have to think through the 2-squirt per cycle thing I saw in your first screenshots.

That's another thing I need to look into. 4 squirts vs 2 squirts. I see a lot of recommendations to go with 4 vs 2.

All stuff i should prob hold off on anyway. I made some adjustments to my VE table and took some fuel out and adjusted my AFR table. I was approaching this as needing to add fuel, but I think the opposite is true. I just want to see what this does and then i'll call it quits until i swap those injectors

VSS from the stock T5 is sinusoidal/analog, whereas the MS may need digital... I'm probably not using the right terms, but I know that's your field and you'll get it. Anyways, if that hypothesis is correct, then you can use a dakota digital converter like I have in black jack, I think. I go from digital to analog, but I think it'll run the other way, too. Summit part # DAK-SGI-100BT. Don't have time to dig into it right this moment.

I think this is just a wiring issue. I wired in the VSS myself so i didn't pay attention to polarity. In from what i read, with the MS it does matter. SO this might be a simple wire swap here.

Fuel pressure sensor in the mail. Will swap the wires and wire in the FP sensor.

I've been stalling on the injectors i know...won't get to swapping until mid-winter anyway so not in a rush
Debating if I send out my 30s or just go with something like this

Ford Performance Mustang Fuel Injectors - EV6 - Uscar w/ Jetronic Adapters - 47lb (86-04)

Upgrade your 1986-2004 Mustang's fuel system with this Ford Performance 47lb fuel injector kit!

lmr.com

 

[FastDriver]

Have fun. Experiment. Learn. There's no timeline, bud.

Did a bit of study with ChatGPT. Something to consider if my other idea is wrong:

MegaSquirt applies the extra fuel, then decays it away over time.

The decay is tunable:

Time-based decay: enrichment bleeds off over X engine cycles or milliseconds.

Taper curve: you can control whether it falls off quickly, slowly, or blends into a “fuel film” model.



On MS3, for example, you have settings like:

“Accel Time” (ms) → how long until the enrichment fully decays.

Decay mode (time-based or percentage of the added fuel).



---

Typical duration

A light stab: enrichment may taper out in 100–300 ms.

A hard snap: it can last 500–1500 ms, depending on your decay settings.

You want it just long enough to prevent a lean spike, but not so long that the AFR goes rich after the throttle is steady again.

 

[Mustang5L5]

Chased the VSS issue down. Ended up cutting the two wires to add some quick disconnects. Plugged in a spare tps and spun by hand and the speedometer function worked.

Plugged the wires back like they were and disconnected the 60pin. Put a voltmeter in on pin 3 and 5 and set meter on AC voltage. Jacked up rear and spun tires and saw voltage fluctuation.

So something is coming from my VSS, but for some reason the MS doesn’t pick that one up, but picks up my spare unit. I have verified that it reads no matter what polarity.

So…guess next logical thing to do is jack up car and plug the spare unit in under the car and test it. If it works….i swap the VSS.

 

[Mustang5L5]

Also…picked up a fuel pressure sensor. So will need to wire this in.

 

[FastDriver]

That VSS situation is perplexing. But yeah, so long as the spare works, don't overthink, I guess.

 

[Mustang5L5]

That VSS situation is perplexing. But yeah, so long as the spare works, don't overthink, I guess.

Got under the car. Unplugged the VSS from the one in there and plugged in the spare. Fired up TS and did not see any signal display.

Gotta be a wiring issue in the car then. My guess where I tapped the wires in the driver's kick panel. No time to dig into that today

 

[FastDriver]

So, you're seeing it on the ac volt meter but not at the MS? I dunno. I'm back to thinking it might be the square vs sine wave. Maybe I'm dumb, but how can you test this? Does the MS manual tell you explicitly that it accepts and analog or sine wave?

 

[Mustang5L5]

So, you're seeing it on the ac volt meter but not at the MS? I dunno. I'm back to thinking it might be the square vs sine wave. Maybe I'm dumb, but how can you test this? Does the MS manual tell you explicitly that it accepts and analog or sine wave?

FIxed it. Traced out the wiring. SOmewhere i had a bad connection. Remember, i added these VSS wired because i did a SD -> MA conversion before and VSS needed to go to the ECU.


ANyway, traced and ohmed out all my wiring and found the culprit. Now i have a speed reading.

I might need to make some adjustments.....

 

[FastDriver]

Awesome, but damn you been holding out on us with them sleeper mods.

Oh, and the EEC always needed to see VSS if you wanted the extra RPM while slowing to a stop. There might be other features, but that'sall the comes to mind. It bumps the RPM until ~5mph. I'm rusty but I think the cruise signal splits off from the VSS line, too.

 

[Mustang5L5]

Awesome, but damn you been holding out on us with them sleeper mods.

Oh, and the EEC always needed to see VSS if you wanted the extra RPM while slowing to a stop. There might be other features, but that'sall the comes to mind. It bumps the RPM until ~5mph. I'm rusty but I think the cruise signal splits off from the VSS line, too.

Cruise only on the speed density cars, hence why my car lacked the wiring over to the ECU when i went to mass air.

Thing is, i never had a VSS code with the A9L, and I code dumped a few times.

ANyway, trying to sort through the MS settings and it looks like i have this right. Ford VSS work at 8000 pulses per mile. I must have something wrong somewhere in the settings.

FOund this good calculator to help try and adjust it, but even it says 8000 pulses/mile gets me in the ballpark at least

Ford Mustang VSS Pulse/Mile Calculator

ANyway, that's checked off the list. Next up wire in the FP sensor and order me some 47# injectors.

 

[FastDriver]

You might've already read this, but yeah, it confirms 8k pulses/mi. You should adjust for rear gears & tire circumference/diameter.

https://img1.wsimg.com/blobby/go/1975f84f-4935-4131-8404-5a914da1afb7/downloads/1brke6aet_808115.pdf

 

[Mustang5L5]

You might've already read this, but yeah, it confirms 8k pulses/mi. You should adjust for rear gears & tire circumference/diameter.

https://img1.wsimg.com/blobby/go/1975f84f-4935-4131-8404-5a914da1afb7/downloads/1brke6aet_808115.pdf

I just did math based on taking the raw Hz reading from the VSS and doing some math with the reported (incorrect) speed to figure it was reading about 2.2 hz per MPH. So a little more math and I have an adjusted pulses per mile of 39,680. We will test at a later date.

I don't quite understand why this is however when everyone else uses 8000 pulses

 

[FastDriver]

I don't either, unless it somehow came from a newer vehicle. The 03 Cobra trans I have, for example is supposed to be something like 32k pulses/mi.

 

[Mustang5L5]

Not physically the same. Can't mount a gear.

I bought this VSS new. Has a ford part number.

Regarless, i'll just play with the PPM until i get it close. Still..it's weird.

 

[Mustang5L5]

Well I think i figured out the VSS issue. Looks like it's a goof on my part in terms of how I set it up in TS and burned it to the car. I had 8000 pulses set...but I never sent that to the ECU. I had 2 pulses per mile on the ECU because I was trying to verify the VSS worked by spinning it with my fingers.

Did some more driving/tuning. I think I figured out accel enrich a bit better. I didn't have enough fuel. Still not perfect, but WAY better than it was last week. Needs a little more tweaking as tip-in from a dead stop is still a bit more sensitive, but it's much better now.

Also figured out a startup stumble that was occurring. Megasquirt has a cranking pulse and taper, as well as a afterstart enrichment. I had my cranking speed set to 350RPM, but I guess my 92+ starter cranks around 325RPM. The car catches, speeds up to 350 RPM and the MS come off cranking pulse. Problem is the cell it lands on in the VE table didn't have enough fuel, so it would go lean and stumble and then slow down under 350 and heistate and then fire.

SO i added more fuel to that cell and bumped my crank to 500 RPM. Didn't get to test it because it was late and this thing is loud, but if that doesn't work I can extend my crank taper a bit too. I'm looking to get that nice smooth flare on startup that settles into a nice steady idle. I'm almost there.

But i can see why they say tuning has a steep learning curve. There's a lot of little tuning quicks and tricks to learn here. Some guys just want to install and go and I can see why.

Injectors ordered. Got the Ford EV6 47#ers. Should be more than enough. Bought a EV6 set of plugs to convert my injector harness over to USCAR plugs. Probably will tinker a bit with cold start just to see if my hunch on what changes to make is true and then I will stop. No point dialing things in when i'm going to make a drastic change with the injectors and the 70mm TB.

Oh, i did manage to run out of fuel while driving. Fortunately it was on a side-street around the corner from my house. Had a 5 gallon fuel container available so one of my kids ran it over.

 

[AeroCoupe]

Dude you are really diving in! I’m just going to get a remote tune after seeing all of this. I’m way more mechanical by nature and this would try my patience too much. I am going to go dual wide bands right out of the gate and I want to stay mass air so I don’t think I want any part of doing this myself.

I’ll still be following along as I like knowing what’s going on in general.