djessence

Hey Crew,

Sorted out my surging and now I seem to be running lean on moderate to heavy acceleration (15-17 on wideband AEM AFR gauge). Car seems to be running fairly well - maybe lacking power but its tough to tell since I am doing pulls on a gravel road as the car isnt licensed for on road use. Idles around 12, light throttle seems to be between 12-14 and "cruise" at ~3-4000 rpm is around 14.5 ish.

Any recommendations on where to start? A bunch of the threads I've read have dealt with the various emissions equipment which we do not have.

Specs are:
- S5 engine, fresh rebuild
- Emissions stripped
- Denso FI fuel pump from surge tank
- first gen chassis.

Any tips are much appreciated.

Jesse

j9fd3s

iTrader: (3)

check for codes, any of the fuel codes will trigger odd limp homes and different mixtures.

might be the O2 sensor too

djessence

O2 sensor only operates on cruise does it not?

djessence

Also what fuel codes would those be?

This is a unique application in that its an S5 in a first gen racecar. The harness I got was fairly hacked apart and since there were some missing bits, its fairly barebones. I did work with the first and second gen forum folks on de-pinning what I could but theres no diagnostics, no emissions equipment.

My understanding was only OMP would trigger limp modes. Are there other scenarios which will impact limp modes? Is there a list of what all forms a limp mode?

djessence

Looking for some confirmations - again engine has no emissions equipment on it:

1. Best spot for the pressure sensor vacuum hose to go to - right now I have it on one of the lower vacuum ports on the lower intake manifold - is this acceptable? Should i put it on the one right below the fuel injector or one of the ones on the firewall side of the upper intake manifold?
2. Best spot for the vacuum connection for the fuel pressure regulator - also have it on one of the lower ports on the lower intake manifold.
3. Secondary injectors are controlled by both RPM >3800 ANDload (ie increased vacuum) correct?
4. Are there reliable aftermarket options for pressure sensor that will work with stock ECU?
5. Secondary injectors should have 12V with key on - is this 12V across the 2 pins? Or 12V from 1 pin, grounding to engine? As well, in the wiring diagrams it says "Grounding time is very short" - does this mean that it will be pulses of 12V even with just power or just a general statement overall? Not really sure how to read into that one.
6. Does the S4 pressure sensor function in an S5? Reading through forums it sounds like no..?

Tested the pressure sensor and I have a feeling its bad. We were only getting ~2.8V with key on, port plugged. Is there a way to confirm either that this sensor is the issue or if it isn't, simulate load and verify the secondaries are working/firing? My thoughts are this should work:

1. rev engine to 4000 rpm
2. Pull vacuum on pressure sensor to simulate load - with a syringe we can pull up to 50cmHg on a vacuum line.
3. Engine sound should in theory change once secondaries kick on.

Should an S5 NA engine have a "variable resistor"? I do not recall seeing one in my many days spent within the wiring diagrams. I currently do not have one so if this is needed, that may be a concern.

j9fd3s

iTrader: (3)

any of the fuel pump speed, or fuel pressure regulator will cause different fueling.

i did have an S5 turbo that ran super lean, and just plugging in the FPR solenoid, got rid of the code, and mixtures came back to normal

the ECU is pretty dumb, a resistor works fine, i usually use the old solenoid because i have them

088

iTrader: (9)

1. doesn't make much difference and where you are sensing vacuum,. The failure mode of the MAP sensor (open circuit) will cause the ECU to assume a fixed pressure of atmospheric (zero vacuum), meaning a "high load" condition (contradicted by AFM and TPS, assuming those are working), so the ECU is adding more fuel than the engine needs, except at WOT, at which point a zero vacuum reading is correct.
2. that spot is fine. Vacuum (low load) decreases the fuel pressure. To take that out of the equation, leave the FPR open to the atmosphere.
3. Secondary injectors contribute nothing until about 4400 rpm. At that RPM, primaries ramp up, secondaries ramp down (by an equal amount), until > 6300 RPM they contribute equally at WOT (secondary to primary ratio is 1:1). See attached for WOT behavior with s5 na ECU (N350/N351).
4. no
5. that simply means they are pulse-width modulated, so a "standard" voltmeter won't see the pulse pattern.
6. no
   
   Disconnect the FPR and MAP from vacuum, do some WOT pulls and see if your AFRs change. They should not, as they are both seeing atmospheric pressure at WOT.
   s5 NAs do not have a variable resistor.
   
   Your secondaries are coming on. If even one secondary injector was inoperational, badly clogged, or its wiring foobar'd, you would feel severe hesitation and power loss at WOT > 6300 RPM because its rotor is only getting half the fuel it needs. It's also possible for them to fail intermittently, resulting in random stumbles as you accelerate. I've had several types of FI failures in my s5 racecar and they mostly fall into these scenarios. AFRs would be far greater than 14.5. At one time, I intentionally tuned a tired engine to 14s at WOT and made slightly more power than with 13s.
   
   I'd start with the obvious -- check the fuel pressure inline. And related -- check/replace fuel filter, ensure you are getting 12V at the fuel pump (are you using an FC fuel pump or aftermarket?), send out your 460cc injectors and get them tested/cleaned.

djessence

Awesome technical response - thanks! I'll try those things out and get the injectors cleaned/tested - that was next on our list

We did do a bunch more troubleshooting last night and found the following:

• MAP tests good as per FSM
• Air Flow tests good as per FSM
• TPS narrow range tests good
• TPS full range "idle" tests high - 1.25V instead of 0.8 V
• TPS full range "WOT" tests good
• Maximum fuel pressure (deadhead) tests good
• Fuel pressure leak tests good
• Inline fuel pressure tests good.

Did a pull accidentally with the TPS unplugged and there was no noticeable change in pulling. But I think that makes sense since my understanding is the TPS primarily used for detecting idle, accelerator pump and full throttle. Don't think we made it to WOT on that pull so not entirely unexpected. I am however curious the impact on being out of spec at the idle setting of the full range is.

Will do more testing and come back.

088

iTrader: (9)

With the TPS disconnected, the ECU assumes that (a) the narrow range is fully open and (b) the wide range is 20% open. It may not matter at high throttle, as the ECU will know from the MAP and AFM that you are in a high load condition.

If the full range TPS resistance fluctuates irregularly as you open it, that *may* cause hesitation on acceleration and when you suddenly change accelerator pedal input. On a digital meter, open the full range TPS slowly to check this; if you have an analog voltmeter that's even easier to see. That said, in my experience, the ECU seems to be quite tolerant of an imperfect TPS.

One more thing to try. Do a half-throttle or quarter-throttle pull all the way to redline. If the AFRs are significantly richer and acceleration is smoother than at WOT, you may well have a fuel delivery problem. (as an example, that's exactly what I experienced with one secondary dying).

Incidentally, the stock mapping causes the mixture to go rich at high RPM. High 13s at mid RPM, dropping down to about 12 at redline (WOT pull). The dyno chart above is for a completely stock s5 NA at 114K miles.

djessence

Exactly what happens. Getting injectors tested this week. Will test the other things as well.

Thanks for your feedback!

djessence

Got the injectors sent out for cleaning and testing - came back as fully functional. Slight discrepency between primaries/secondaries of about 30cc.

We did some more testing to ensure vacuum lines were in the right place - wanted to confirm

• air injector bleeds are the vacuum ports immediately below and inline with the secondary injectors?
  • these T together and go to one of the 3 vacuum ports on the UIM windshield side
• Primary air injector bleed is slightly off center between the primary injectors on the engine - this vacuum port is vertical?

Further - we can be doing low speeds (20km/h up to 60km/h) in third gear with WOT, AFRs sky rocket up to 17/over limit.

So it seems like the signal to increase duty on injectors is not getting back to them for some reason?

Any thoughts?

FührerTüner

iTrader: (5)

Are your 5/6 ports functional? Open all the time?

Also what ECU are you running?

Have you pulled the plugs and checked them out? When you have that lean condition does it stumble and buck? If the ignition is breaking up its possible youre reading lean from the unburnt fuel being pumped into the exhaust, thats why I asked about the plugs.

djessence

• 5/6th ports removed - was following a bunch of info on the site that said should be no issue removing ports entirely or wiring them open - is this not the case? Probably should've mentioned this before. Definitely forgot about that detail.
• N350 ECU (came with the engine)
• It does hesitate when is leans out but not like a cut or drop in power just less power if continuing to accelerate. Haven't pulled plugs but thats a great idea.

Jesse

FührerTüner

iTrader: (5)

You will lose a lot of low end from the removal of the ports. The scenario that you described, 3rd gear 60km WOT pull, is exactly where youre going to feel the difference. It also makes sense that youre seeing a lean spike because the oem tune doesnt expect that much air considering normally the 5/6 ports are closed until 3500ish RPM. Maybe the AFM should compensate? I dont know, Just a theory.

Also, another theory, your N350 is a California ECU which in my experience usually means they detune and lean the **** out of it. Probably not helping your situation.




Edit: I dont know why some font is larger than the other, but it bothers me.

djessence

For some reason I thought N351 was cali spec? I'll double check that I have it in but this is good to know. We have one of both ECUs both N350 and 351.

I realize you lose low end - it's an endurance racing car so we are rarely below 4000 rpm so felt the sacrifice was worth it. That being said, we still have all the parts and could put them back in if absolutely necessary.
Fair theory on the tune not compensating adequately for the ports being open. I wonder why this hasnt come up with others who have done it. Or if they have done it, they haven't had an AFR gauge and were not aware of it.

I'll do some more digging. Thanks for your speculation - gives us something else to think about and we can try swapping ECUs to see if it makes a difference.

Jesse

FührerTüner

iTrader: (5)

Probably because most NA guys with a stock ECU dont have a wideband installed.

Put the sleeves back in and see if it richens up...

088

iTrader: (9)

> Also, another theory, your N350 is a California ECU which in my experience usually means they detune and lean the **** out of it.

Not true. The California (and federal) ECU goes rich after 6000 rpm -- see dyno chart above, which is for a completely stock setup. If you plan to swap ECUs for troubleshooting, both the federal and California ECUs will be fine.
In a naturally-aspirated race car, you should never drop below 5500 on track if you can avoid it. Eliminate the aux port valves / wire them open and wire open the VDI.

Note that the 6PI and VDI solenoids are activated by the air pump (positive pressure, +1 to +3 psi). If you eliminated the air pump and/or ACV (which you should, on a dedicated offroad race car), the VDI valve never rotates to the short intake runner position, and you are missing out on 20 to 25 hp after 5500 rpm (this is the point where the ECU energizes the VDI solenoid, directing air pump pressure to the actuator). Same goes for 6PI, but you already have the actuators wired open.

Which air bleeds are you indicating?
The oil injector air bleeds connect into metal piping that collects to one of the tubes on the air snorkel -- i.e.; *before* the throttle body; do not connect them to vacuum sources on the manifold (the other tube on the air snorkel goes to the BAC valve). Obviously, this won't affect your lean condition (the air inlet on the oil injector and internal valve diameter are tiny), but it will ensure correct oil injection, assuming you are running the OMP (eliminating it is another simplification you can make on a race car, if desired, but on s5 you'll need aftermarket ECU or leave the OMP connected and mounted elsewhere -- the ECU has to "see" it.)
If you mean the air bleed on the primary fuel rail, that controls whether the pulsation damper sees atmospheric pressure or vacuum. You can leave it open to the atmosphere on a race car.

> Fair theory on the tune not compensating adequately for the ports being open

The aux port actuators are fully open after 3800 rpm. You won't see any impact on AFR in your RPM range of interest (>4000) if they're wired open; you would see impact if they're stuck closed (a bit richer).

> Further - we can be doing low speeds (20km/h up to 60km/h) in third gear with WOT, AFRs sky rocket up to 17/over limit.
> So it seems like the signal to increase duty on injectors is not getting back to them for some reason?

In third gear, you would be at 1500 to 3000 rpm between 20 and 60 km/h and fuel is being delivered entirely by the primary injectors in this range. So there is your problem. You've installed known good injectors (tested) and swapped ECUs (assumption). So you have a problem in the wiring. You can backprobe the ECU outputs (pins 3W and 3Y) and look at pulse width or voltage; backprobing the injectors themselves is quite a bit more difficult since they are buried underneath the UIM. Bad wiring means the ground signal isn't being seen at the injector or it causes the ECU not to ground the injector appropriately to keep up with load demands (pulse width too short). A bad or partially bad wiring harness may not be evident, even after you remove it and test continuity at both ends. Once, I've fixed this problem by swapping out the harness. Another time, the replacement harness did not fix the problem because it exhibited the same issue (doh). A third time, the harness was bypassed by running a wire. Which wire? This is where you have to get creative (if you want to dig in this far). There are two wires going into each injector; the thicker black/yellow wire gets constant power with the ignition on. Connect this to battery power on the front injector, then do it for the rear. If the car runs the same in both cases, the problem is in the grounding circuits. Run a wire from the injector (this replaces the thinner of the two wires) to the ECU for one injector, then the other. [Based on your description of the symptoms, I doubt the secondary injectors are playing a role -- but if they were, the harness troubleshooting is the same. You can also swap the wires at the ECU (front primary <-> front secondary; rear primary <-> rear secondary). If the car runs on one rotor, you've identified the secondary injector circuit causing the problem.]

Start with replacing the ECU (if you haven't done so), followed by the wiring harness and pray that fixes your problem.

FührerTüner

iTrader: (5)

Im talking about CA vs Federal maps. The CA maps are slightly leaner than the Federal ECUs to meet Californias emissions standards. That would make it a contributing factor.

Hes not having a problem up top. This whole thread is about low rpm lean acceleration. Low speed low rpm high load. The car still runs and drives, its just lean. Wouldnt you agree that the ports being wide open can also be a contributing factor?

djessence

Reviving this a bit.

Should the VDI valve vacuum port be blocked for a race application or hooked up to vacuum?

j9fd3s

iTrader: (3)

pressure, it needs pressure to open. if you are over 5250rpm its "open" (on, active), in a race car you probably don't spend a lot of time under 5200rpm, so you might as well wire it open, one less thing to fail. in a street car, you want it working

djessence

Thanks.

Wired the VDI open. Still seeing crazy lean under >20% throttle.
What's changed?

Realized intake air temp sensor in the UIM is broken off - i don't think this would make that much difference on AFRs. Can someone confirm?
Fuel Pump relay has been wired up (but doesn't actually do anything because our pump is full on all the time).
timing confirmed

Where I am struggling with this is the air is metered - that means I would think the only item that would lean out the AFRs is unmetered air getting in the system (Fuel injectors confirmed, sensors confirmed). Any thoughts?

A scenario that I can see specifically where things change is the following:

• Lugging the car - 3rd gear, 100% throttle, 1000 RPM (full load, low RPM)
• AFRs are in the 11-12 which I would expect in this type of scenario
• Once 1500 rpm hits, AFRs go to 17+ overlimit

Another scenario:

• 2nd or 3rd gear with <20% throttle, can go all the way up to 7000 rpm with AFRs being <14
• If you blip throttle or try and increase throttle beyond the 20% (ish) mark, AFRs spike and you can hear the detonation.

Thoughts? We are getting close to carbureting this as I feel like we are running out of things to troubleshoot.

djessence

Figured it out. It is in fact in limp mode - found the sequence of limp mode events an our AFR gauge is on the rear rotor pipe so it all makes sense.

Ran the tests - both of our ECUs are fried and not showing output to OMP (per FSM tests). I'm guessing that also means the OMP is fried, though the resistances on it test fine.