After coming off the gas for any moment, whether driving or sitting in neutral, the idle speed bounces from 500 up to 1k and repeats over and over for a few seconds, before usually spiking to about 1200-1500 and then evetually comes back down and idles normally at about 800. Im at a loss for what could be the issue and looking for more ideas.

Cleaned MAF and K&N Air filter and ESS
No vacuum leaks, checked using smoke
Coils and plugs new within 5-10k, recently changed to BHR kit ignition kit
Cat seems fine, took out for a spirited drive and checked underneath, not glowing or anything
As far as i can tell no major power decreases anywhere along the rev range.
Very occasionally will stall maybe 10% of the time this happens.
No issues cold start or warm start.
No CEL or codes

recently did have some issues where accelerating up to 4k rpm would cause kind of a sputtering effect, but cleaning maf and air filter has gotten rid of it, and im just left with this issue.

Any ideas? Thanks in advance

Edit: I should note that this also happens when cold, but the idle 'bounces' higher up on the rev range then between 500-1000

 

Have you reset the fuel trims and the ESS after making all these changes?
What is your LTFT at idle?

 

Thats the 20 brake stomp procedure right? I did not think to before, since it was happening before i cleaned my ESS, but i did now, ltft was -22.1 before and after reset, but i assume any changes would take a few drives to set in?

Also noticed before the rpm spike, Stft was at 0 and then jumped to +25.0 and thats when the idle smoothed out.

 

-22.1 LTFT? That is at idle? You are running extremely rich. What are your MAF readings at idle? My only guess here is one of 3 things. 1. The MAF is malfunctioning, 2. You have a leaking injector or 3. Your precat O2 sensor is not working properly. Then again I could he totally wrong but that is the direction I would take. I wouldn't discount ignition issues either, i.e. not burning the fuel properly. Go through the ignition also.

 

Is the K&N filter attached to a K&N Typhoon intake? If so, throw it in the garbage and go back to stock. See if that fixes it.

Otherwise if you really have MINuS 22.5LTFT, you could have a leaking injector or an ignition coil has failed (as pointed out above). Or, the injectors could have been wired incorrectly if there was engine work done recently. There was a case like this recently and IIRC that was the cause.

Fuel trims are not reset by 20-brake-stomp, pull the battery for a minute for that.

 

I said K&N but I meant AEM intake I'll try disconnecting the battery then to see if that helps. What's weird is i feel like if an ignition coil has filed there'd be more symptoms or a code or something? Last time a coil failed I found out by a misfire, but nothing like that. No work done on the fuel injectors and hopefully it's nothing to do with them, that doesn't sound like an easy fix

MAF at idle is about 8g/s I believe. I'm not real familiar with torque so I'll take a screenshot of the dials and post it to make sure Ive got the right one

Edit: screenshot, taken after a short drive not quite warm, but not cold either

 

Funnily enough, that's only at idle, if I'm driving its around 0. its like the second I drop below 1500 rpm, it shoots down to negative 22 and then after a few seconds the stft shoots up to 25 to compensate. Perhaps that's the reason I'm not getting a code for running rich?

 

Don't assume you'll get a code.
Here's the other case I was referencing. Turns out it was his ESS. Have you messed with it lately?

https://www.rx8club.com/series-i-tro...0/#post4845224

6g/sec at warm idle is roughly the correct amount of airflow.

 

Yeah that was the first thing I did, last time I had idle and stalling issues I cleaned and reset the ESS and all was good.

I reset the fuel trim by disconnecting the battery and after a long drive it seems to have normalized to about +6-8 when idle and the problem seems to have gone away. Maybe the ecu compensated when the MAF sensor was malfunctioning and after I cleaned it, it just didn't know what to do?

I'll watch it over the next week to see if it resurfaces
​​​

 

Ok. LTFT should be zero, so you probably still have a vacuum leak.

 

Let the car idle for 10 minutes. Don't press the accelerator at all. Monitor your LTFT it should be nearing 0. + Or - 5% LTFT is no big deal. If you want to be more proper, reset the fuel trims again and let the car idle for at least 10 minutes and see where you're at.

If you are over 5% start looking for a vacuum leak. Get a can of carb cleaner and go over all the vacuum lines when the car is idling. If you hear a difference in the idle when you spray an area then investigate there for a leak.

The main lesson learnt here IMHO is that whenever we change/clean MAF, throttle body, change air filter, O2 sensors, ignition, always reset fuel trims. The ESS needs to be reset especially after wires Sparks and coils. I had 2 guys here that were getting misfires after changing coils wires plugs, this would go on for months and they wouldn't believe me when I told them to do the 20 break stomp, they thought I was nuts. Well it solved the misfires in both occassions.

 

Okay, my little two pennies. Long story short I was having the same issue and did/change everything you did.

Problem: When coming to a stop at low RPM the needle will fluctuate and (sometimes) it will stall.

Fix: Clean ECC (Eccentric Shaft Sensor) + reset NVRAM (20 brake stomp) + Disconnect the battery for 10 mins. [Vital]

The true fix for me was to disconnect the battery to reset the fuel trims. Because I cleaned my ECC and I did reset the NVRAM but the issue still occurs. ONLY after I disconnected the battery for 10 mins the car when back to normal idle when coming to a stop.

I really hope this helps someone because I was almost ready to give up on my 8. (Sorry to say but it's the truth).

 

Long term fuel trims are constantly being updated. Simply clearing them in and of itself isn't a long term fix because the PCM will just re-learn them based on the state of the vehicle.

If that did fix your problem long-term then it's very likely something else changed around the same time.