AFR values not adapting (rich), injection time too long

[slaughter2k]

Hello Porsche enthusiasts ?

My first post so I give a short introduction. Have a 2001 986 2.7 and am from Germany.  986er is completely stripped for track use. I rebuild nearly the complete engine and did also some modifications. 

The point:

- Durametric log shows rich mixture (o2 sensors at 0.85V confirmed by wide band sensor)

- FRA is sticked to 0.99 even after battery disconnection

- RKAT is sticked to 0 even after battery disconnection

- injection time is higher than spec with 2.7ms (spec 2.2ms)

- MAF is ok with 17.8 @ idle I guess

- comment: warm engine stalls sometimes at traffic lites, MAF is original, engine running in general OK

 

modifications:

- lightweight flywheel 

- 996 throttle body

- race cat with new exhaust (because of this there is currently no post cat o2 sensor connected)

 

Hope someone of you has a hint for me. I hope my ECU is not toasted ? (due to frozen FRA RKAT value)

[Ahsai]

Welcome to Renntech.

 

Have you tried resetting the DME (by clearing the codes) using Durametric?

Did you really mean the O2 sensor reading is stuck at 0.86 and not moving at all? Does it even swing back and forth between 0.2v to 0.8v?

You may also want to re-calibrate the e-gas by turning the key to the last position before crank and leave it there for 60s or so (with foot off the gas pedal), then turn the ignition off and wait for 10s.

[Ahsai]

Also, how about the FRA and RKAT vaules of bank2? Ideal value of FRA is 1.00 and RKAT is 0.0%

Edited February 26, 2019 by Ahsai

[slaughter2k]

codes cleared with durametric.

No, both more or less: FRA RKAT values are completely frozen. O2 sensor value is moving very little (0.85-0.86V). No swinging as it should (also not when switching to the visual mode/ graph).

Calibration is done...know what you mean. Btw: throttle pos. at idle is 1.6%.

FRA RKAT is the same for bank 2

 

thanks for first help ?

Edited February 26, 2019 by slaughter2k

[Ahsai]

Perhaps just drive it for a few days and see what error codes come up?

[slaughter2k]

I just ordered two post cat o2 sensors to complete the system again.

Aditional comment: I disconnected the complete secondary air system including electric valve. Might be an issue?

picture shows idle and 3000rpm. MAF values makes sense, O2 voltage goes down when engine costing down. Injection graph is normal?

Edited February 27, 2019 by slaughter2k

[Ahsai]

Removing the SAI should not affect the normal close loop control.

Have you verified the fuel pressure? Are the coolant temp and intake air temp reading where you expect them to be?

Can you switch back to the stock throttle body?

 

It's so strange the injection time goes down as the rpm goes up. Almost looks like the DME is still in open loop control.

[JFP in PA]

1 hour ago, Ahsai said:

Removing the SAI should not affect the normal close loop control.

Have you verified the fuel pressure? Are the coolant temp and intake air temp reading where you expect them to be?

Can you switch back to the stock throttle body?

 

It's so strange the injection time goes down as the rpm goes up. Almost looks like the DME is still in open loop control.

 

It almost looks like the DME has still not adapted to the throttle body change.

[slaughter2k]

Update:

today I installed the post cat O2 sensors because I read that missing O2 sensors are triggering a rich save mode. --> no change. No fueling adaptions, engine still stalling when warm.

Tomorrow I will adapt the smaller original throttle body to the bigger 987 manifold.

In addition I read that a unplausible fuel pump status might also cause injection time elongation.

Bosch me 7.2 system description (based on landrover)

 

Edit:

I forgot to check coolant and air temp today, will do that tomorrow prior throttle body change.

Edited March 4, 2019 by slaughter2k

[slaughter2k]

temp sensor values checked at room temperature. Btw: I couldn't find the coolant temp after radiator. Engine temperature is "engine coolant temperature",  correct?

Plausible...

just re-adapting the original throttle

Edited March 5, 2019 by slaughter2k

[Ahsai]

Yes, "Engine temp" is measured from the coolant temp sensor plugged into the "front coolant console" that's between the engine and the engine support bracket.

[slaughter2k]

Update:

original throttle installed. Engine not stalling any more on traffic lights when warm. Fuel adaptions will still not performed ?. Engine running rich. Problem 1 solved, problem 2 pending

Remark: Reinstalled 996 throttle again with success. Issue was that I mounted it initially 180deg. rotated which caused stalling when warm :)

-->what issue will not stored as a failure and might block fuel adaptions? Any ideas?

Edited March 5, 2019 by slaughter2k

[slaughter2k]

made a 3rd gear wot yesterday with a wide band O2 sensor on board.

Idle is running at afr 12.9, wot at 10.5. Ignition angle is retarded at 4100 rpm ?

I guess there is an additional major issue.

[slaughter2k]

Anyone has any advise?

Don't know what to do else: Ecu is not adapting fuel trims and retards timing over 4k rpm

[Ahsai]

Re-reading your MAF numbers again I have a question. Did you increase your engine displacement or it's still at 2.7L after rebuild?

 

The reason I asked is at idle 17.8kg/hr @700 rpm sounds very high for a 2.7L.

More importantly, a 2.7L max MAF should be ~652 kg/hr (based on my computation with some approximation with IAT and atmos. pressure) but you got 700+. Have you tried a different MAF?

[Ahsai]

Any other intake mods other than a 996 throttle body?

[slaughter2k]

Should have summarized them accurately before...modifications:

- Aasco light weight flywheel 

- 2.7 987 inlet manifold adapted

- 996 plenum (t-pipe) with 996 throttle

- straight intake piping through the fire wall into rear trunk

- original MAF housing (cutted from org. air filter housing)

- Filter cone (slightly oiled for prevent contamination of MAF)

- deleted SAI (completely. closed also the head bores at the manifold flange)

- complete new exhaust (equ.length headers, 200cell converters, new O2 sensors

- low temp thermostat 

- underdrive pulley

- new ckp sensor with controlled gap (1mm --> had "signal below threshold" failure before)

- vaccum port at plenum / throttle closed

- crank case evacuation via exhaust and moroso valves

- tank venting and charcoal canister system removed completely

- left tank vent valve, SAI pump and SAI valve attached to prevent ecu failures for now (no failure codes currently except SAI pump)

- new spark plugs and coils

- 986 2.5 liter transmission

- stripped wiring harness. Removed all wires which are not needed including all relays. (nothing engine related should have been cutted)

- hm, that's all I guess beside mechanical maintenance and slight improvements

Not changed/touched:

- stroke

- ecu mapping (for now)

- MAF housing diameter or MAF itself

 

--> I will post some detail pictures of modifications in about four hours

Edited March 8, 2019 by slaughter2k

[slaughter2k]

summarized idle values and a MAF plausibility calc. that I did some days ago based on any 996 3.4 porsche dyno log I found in the web (for sure don't know the accuracy)

Edited March 8, 2019 by slaughter2k

[Ahsai]

8 hours ago, slaughter2k said:

Should have summarized them accurately before...modifications:

- Aasco light weight flywheel 

- 2.7 987 inlet manifold adapted

- 996 plenum (t-pipe) with 996 throttle

- straight intake piping through the fire wall into rear trunk

- original MAF housing (cutted from org. air filter housing)

- Filter cone (slightly oiled for prevent contamination of MAF)

- deleted SAI (completely. closed also the head bores at the manifold flange)

- complete new exhaust (equ.length headers, 200cell converters, new O2 sensors

- low temp thermostat 

- underdrive pulley

- new ckp sensor with controlled gap (1mm --> had "signal below threshold" failure before)

- vaccum port at plenum / throttle closed

- crank case evacuation via exhaust and moroso valves

- tank venting and charcoal canister system removed completely

- left tank vent valve, SAI pump and SAI valve attached to prevent ecu failures for now (no failure codes currently except SAI pump)

- new spark plugs and coils

- 986 2.5 liter transmission

- stripped wiring harness. Removed all wires which are not needed including all relays. (nothing engine related should have been cutted)

- hm, that's all I guess beside mechanical maintenance and slight improvements

Not changed/touched:

- stroke

- ecu mapping (for now)

- MAF housing diameter or MAF itself

 

--> I will post some detail pictures of modifications in about four hours

 

Wow, glad I asked. I have highlighted what I think are suspicious items that should be checked/understood.

 

Your intake has extensive modifications. I understand you retained the MAF and stock MAF air tube so it should still be measuring correctly. However, that tube (where the MAF plugs onto) relies on smooth flow so any mods before it could cause turbulence that throw off the MAF reading.

 

The CKP I'm not sure if modifying the gap makes a difference but I remember the DME does need to learn the flywheel and adapt to its pattern (Durametric has two values for this if memory serves). Not sure how these play into fuel control.

 

You cut off the crankcase ventilation and EVAP from the intake and  I think this would be a major one. The reason is the stock DME does expect these flows AFTER the MAF/throttle body and the stock fuel mapping takes into account these "leak" which are not detected by the MAF. The crankcase contains mainly air and very little fuel but the EVAP ventilation could contain significant amount of fuel, depending on the state of the charcoal canister.

 

Stripped wiring is always suspicious. Are there any error codes reported by the DME?

So is your engine still a 2.7L or it's not? You said the stroke was not modified but you never mentioned the bore...You just need to tell me if it's still stock displacement or not if you don't want to disclose the actual displacement.

 

 

[Ahsai]

8 hours ago, slaughter2k said:

summarized idle values and a MAF plausibility calc. that I did some days ago based on any 996 3.4 porsche dyno log I found in the web (for sure don't know the accuracy)

 

Without knowing your actual displacement, I assume it's 2.7 (stock). The MAF reading is definitely too high for a 2.7 engine (idle speed was proper at about 700 so I assume there are no other loads on the engine such as a/c).

 

There is a recent thread here about MAF reading of 996 3.4 and 3.6 engine that were taken by the Torque Pro app. The max MAF for 3.4 was ~213g/s (767 kg/hr) and for 3.6 it's ~241g/s (867.6 kg/hr). One difference is these numbers were measured from the OBDII port and yours seem to be from a Porsche specific scanner.

 

It looks like your engine is in OPEN loop fuel control. Can you read the status with your scanner? I know that OBDII reports this status "Open or close loop" but not sure if Durametric or alike also reports the same.

 

One thread I found about a "stuck" DME is here and the problem was water intrusion into the immobilizer. Some similarities but not exactly the same. May be useful to read up too.

[slaughter2k]

Hi,

detailed feedback...thanks for that.

 

first: yes, displacement is still 2.7l

 

Because of smooth flow I already thought of installing a grate from an other MAF to smoothen the air flow

 

Problem with the CKP was that the shutter wheel from the aasco flywheel was not accurate. In addition the distance "aasco shutter wheel to the ckp sensor" was 0.8mm bigger than "stock flywheel shutter wheel to cpk sensor". After measuring and correcting nearly all teeth of the aasco shutter wheel the CKP failure code was gone. I red from 4 adaption values in bosch me 7.2 technical documentations. But I did not find anything how I can delete this adaptions for the porsche ecu. As I red from many boxsters running fine with aasco flywheels I hope I can exclude this suspicion that the open loop issue is localized here.

 

Evap system is an influence but it should be visible at the fuel trims somehow. My ecu does not make any attempt to regulate nor highlighting a failure code.

 

My big suspicion is also that he is stuck in open loop because of anything which he is not seeing as a reason to highlight a failure code. I already found your linked thread and was excited first. During reading it again I found out that the immo problem and the open loop problem was not related to each other. The open loop problem solution has not been diskussed unfortunately. 

 

I will check now:

- loop status (open, close)

- a different maf housing and a grate

- suspicious ecu pins with open wires from harnes stripping

- ...

 

Edited March 9, 2019 by slaughter2k

[Ahsai]

Don't have much to go on but can you check this?

- Reset the DME using Durametric

- Disconnect the battery for 10minutes and reconnect it

- Reconnect Durametric but do NOT start the engine

- Read all the FRA and RKAT. Should be 1.0 and 0.0, respectively

- All 4 O2 sensor voltage. Should read ~0.44v on all 4

 

I just want to see if the DME has been reset and if the O2 circuits are healthy.

Btw, are the precat O2 sensors stock with 4 wires each?

 

Also, it feels like something preventing the DME from entering closed loop. Will need to verify DME gets all it's looking for: MAF, intake temp, coolant temp, CPS, cam sensors, O2 sensor heater, O2 sensor, knock sensors, e.g. are all these sensors connected and providing the correct signals.

 

Not reporting failures could be because some failures require closed loop as a pre-condition before running tests.

 

[slaughter2k]

My durametric (old one) has no button for ecu reset, just for failure code delete.

I will check O2 and fuele trim values after disconnecting/reconnecting the batterie.

 

What I did today:

Soldered a potentiometer into maf signal line to manipulate the voltage to get 14.7 afr in open loop.

What to say...O2 sensors (wide band external and regulating sensors) reacted immediately. Adjusted to afr 14.7 (was just a very minor amount of adjustment) and got

700rpm

2.7ms injection (accurate measurement)

Afr ~14.7 (accurate measurement)

18kg/h maf (hand adjusted via potentiometer)

2.4% engine load

2% throttle (accurate measurement)

This looks like I have an increased amount of mechanical load at the engine or maybe a compression or timing issue (??!!)

 

What I'll do tomorrow is trying to get rid of the sai failure code and reinstalling sai pump relais (maybe this is preventing closed loop mode)

MAF: guess it is giving reasonable values, even if they might be a bit off

intake temp: confirmed with reasonable values via durametric

coolant temp: confirmed with reasonable values via durametric,

CPS: new sensor installed, no codes any more (still suspicious due to aasco flywheel)

cam sensors: just one in row engine but this gives me 3-4deg. offsett: Should be ok

O2 sensor heater: sensor are perfectly responding on maf modication and coast mode. Bad heater should also throw failure code

O2 sensor: similar too heating

knock sensors: that's something good I wanted to check but don't know really how to.

Will upload a engine running video for you

Edited March 10, 2019 by slaughter2k

[Ahsai]

That's great progress. It's very interesting. Compared to your first post, the rpm and MAF numbers are very similar, with or without the resistor and yet with the resistor, the DME entered closed loop fuel control.

[slaughter2k]

Hi,

no, It's still not in closed loop. I just adjusted the maf voltage by hand until the external wide band o2 sensor showed adr 14.7 in idle. That's it.

The narrow band o2 sensors from the ecu also reacted during hand adjustments. (jumped from rich to lean when i was near 14.7 as they should). but everything in open loop, no automatic fuel adjustments. 

--> currently it looks like he really needs that ammount of air and gas at correct afr to maintain 700rpm idle ?

Edited March 11, 2019 by slaughter2k