hunting for idle (engine revs cycles rich and lean)

I have a 3sge beams vvti with the G4+ link ECU. Its been months of trying new setup/adjustments to get the engine to run smoothly.

For some reason i believe the fuel trim is cutting off and causing a lean condition and giving be a dead spot in power while driving around. It is very strange because the engine feels great under rich or stoic but then drops to lean condition where the power dies off.

While warm up the engine revs seems to be stable however when i start to reach the operating temp (~165 F and above), the AFR meter and engine speed starts to cycle from stoic(14 to 22.4) & 1500 psi to 2500 psi.

I have an Innovate MTX-L Wideband Air/Fuel Ratio Gauge & wideband o2 sensor. Could I be having issues with the CLL setting (Stoic - wide band) to settle it down?

FYI:

A throttle delete kit was also installed so i am only using AP(main) Analog input as the throttle.

I have attached the a log recording and the run and my ECU setup file.

I used MGP (PSI) for all the ignition, fuel setups.

Need help, anything to see if i have a good general setting.

From your configuration and the log, it's clear the ECU is going in and out of over run fuel cut. The threshold is set to 1500 rpm with a delay of 1 second and if you add Overrun Fuel Cut Status to your logging you can see when it becomes active. If you simply disable the overrun fuel cut function I'd say the engine will respond how you expect. Is there any particular reason it's idling so high though? Unless it's a pretty wild 3SGE I'd expect you should be able to achieve a stable idle at 800-1000 rpm.

Thank you Andre!!!

I turned the over-run fuel cut off and the engine does not cycle in revs anymore.

As for the high idle, i manually set the throttle screw high on the throttle and now have it at 950 rpm.

I took the car out for a drive and powerband is back, but i am still experiencing a lean condition every 10 seconds.

The car idles at 14-15 AFR then drops to 17,18,19 before jumping back to stoic.

My CLL signal gain is now set to 2 and Fuel trim +/- set to 15%.

Could it be that i have the closed loop gain signal too low and perhaps need to increase the fuel trim?

If you need more than a 15% trim, I'd be revisiting your main fuel table - I like to see the trims no more than +/- 5% typically. It's easy to see what's going on though if you log the car at idle and look at aspects such as your closed loop trim, injector pulse width etc.

Andre,

I have attached two logs and my map setup.

1. First is startup at cold, the engine bounces between rich and lean then stays lean. i had to tap the throttle to keep the revs somewhat up.

2. engine idle smooths out at 950-1000rpm as it warms up to ~171 deg F. I have CLL on at 165 deg. I can see the CLL activates at the set 10%.

As the temp comes near 162 163 degF I notice the Lambda 1 signal just spikes to AFR 29 lean.

Currently i have the fuel table set at approx values at 27 to 30 for idle and slow cruise Is this considered low?

Not sure how to go from here.

Hyperollah,

Quickly looking at the map, i see you have the CLL gain set to 6, this will cause a large swing in the closed loop correction strategy. Also, I believe your base fuel map needs revisiting. With 2% warm up correction and no CLL your actual AFR is between 18~20. This is in the region of 20-40% away from your target AFR.

The AFR signal is also doing something that i would not expect. You will notice that the ANV 9 error value is 2 Lambda, from the log you can see the signal go to this value at around 2:25 seconds into the log. It continues to cycle in and out of the error state. I imagine it is hitting the an error high value and defaulting to the error state.

Could the AFR signal lean spike be a cause of the low fuel map values?

I will lower the gain on CLL and increase fuel enrichment in the warmup state cold to .

Should I setup CLL to activate during warm up?

I suggest you tune the main fuel table. I can't comment on what numbers you should expect as i do not have enough information (injector size etc.) From the logs i see the inj pw is around 2ms which isnt a large a mount if these are standard injectors. Increase the main fuel table numbers until the AFR reaches your target. It is important that you tune the main fuel table before you enable CLL.

Update,

I increased the Fuel Table and was able to see better Lambda numbers. (Idle Fuel values up from 27 to 38%). There are still little spikes of lean conditions. Notice the log shows MAP(PSI) cycling with the lean events. Hmmmm? Could this bad MAP signal be giving me the erratic idle?

I also see that I forgot to turn off the CLL.

Lastly when I tap the throttle as shown later in the logs @ 1:55m you can see the lambda signal just cuts off at high Lean values and engine trys to die. Maybe need to look at the Accel enrighments for this problem.

I will try another fuel table with higher values and turn off the CLL. I will also try to increase the sens of Accel enrichment to 55% 0-2000rpm.

I am running a stock Altezza engine and stock fuel injectors (~365CC)

I will be trying higher Fuel Table Values in my next runs.

Attached is my program and logs

Happy Easter!,

I found my idle cycling issue, my fuel pump voltage was not stable and so I traced it to a damaged alternator S (voltage regulator) wire.

After repairing the splice, the engine idles is smoother and does not cycle between rich and lean.

I took the car out for a drive, the engine seems to idle ok but as soon as i input throttle, the engine leans out.

I attached the most recent log and ECU program. Here are my setups:

-Ign: Idle @15 deg, 2000 rpm@20 deg, above @ 30 deg

-Fuel Table values 30% to 70 % (still working on this)

-Master fuel 15 ms

-CLL turned off

-Accel enrichment 40% (0-2500rpm) then tapers off

The log shows when AP(Main) is active the AFR leans out, also the engine still has a boggy gurgle sound when reving up. no power 2k rpm and above. Need advice. how to keep the lean out during acceleration.

One Question:

Is Accel Load correction supose to start at value 1 and <1 as the MAP (PSI) increases?

The accel enrichment is essentially a correction applied to the main fuel table value. What this means is that if your fuel table isn't accurately tuned then it wont matter what you do with your accel enrichment as the fueling wont be right. The log you've appears to be just idle with some stabs on the throttle which isn't overly useful. While the lambda does go lean on a stab of the throttle, you can see that with a more gentle application the lambda is better. The difference is in the area of the fuel map that you're accessing. Tune the fuel table fully and then deal with the accel enrichment last.

I'm not sure if you're aware but you have the open loop AFR function enabled too. There is nothing wrong with using that function but it doesn't look like the AFR table is filled out with likely targets.

Check out this webinar on accel tuning on the G4+ platform - https://hpacademy.com/previous-webinars/038-configuring-and-tuning-acceleration-enrichment-on-a-link-g4/

I have attached a log of a drive around my block if it helps any.

I will concentrate on the fuel tables and look into the open loop AFR

thank you

From the looks of it you only have the accelerator peddle position sensors working, there is no values from the throttle position sensors, make sure your throttle plate is working properly as it doesn't look like it.

Just took a double take at your map file, is the electronic throttle system switched off?

This engine is fitted with a throttle delete kit from japan. Only the foot pedal is active. yes Tp man is turned off

That could be why your not getting acceleration enrichment then, the ECU uses the throttle position rather than pedal position.

Try changing your APS to TP and see how it behaves

Chris,

I have the Accel enrichment configured to follow MAP, will this not work?

I never even looked at that, it won't be anywhere near as responsive as throttle position. Andre explains it well in the webinar he linked up earlier

The only time you'd use MAP as an accel parameter in the Link ECU is if you had not TPS sensor - In which case my advice would be to fit a TPS sensor and do the job properly :)

You just won't see the same sort of results on this platform as using the TPS. You have a TPS already so there's really no reason not to use it. Do as Chris suggested and reconfigure your input from APS to TPS.

Good spotting by the way Chris ;)

the reason I setup the Accel enrichment with MAP is because the throttle body is fitted with a throttle delete kit:

http://global.rakuten.com/en/store/premiumjapan/item/kit-002/

I was under the impression that the TPS would not even give the proper signal.

I will try by turning on the TPS(Main) analog input and set the accel to TPS.

Thank you

Cheers Andre :)

your saying it's got a throttle delete, from the look of the kit it is a conversion so you still have a throttle plate, it's just cable driven and not electronic any more?

From the look of the data logs you're getting a full range of movement though your pedal positions sensors, you should also get a full range from the throttle position sensors. If not, and this situation is not ideal, you can use the pedal position as your throttle position input, but the best way is to have a sensor directly on the throttle.

I plugged turned on the TPS signal on (AN Volt 5 & 6). I logged a drive around the block after warmup and noticed no activity for TP(Main) as well as accel correction.

Can I setup AN volt 5 & 6 with Foot pedal APS and keep the Accell enrich mode & load axis on TPS?

thank you Joseph

Use the AN 2 & 3 that you currently have as APS, designate them as TPS. We can see in the logs that these ones work so should activate Accel enrich so long as that has been set to TPS