Stubborn Lean Spike in Mid-Range, inconsistent AFR (uS)

[MikeK]

Hi, first post. I have been running my 1979 Volvo 242 on megasquirt for almost two years. Last year I was running on ms2 v3, stock ignition and it was running great. I decided to make some upgrades and in the process built a wiring harness and switched to a Microsquirt V3 and have been trying to get my tune back ever since. I have been using the help of a very gracious Volvo guy who has a lot of MS experience during all of this. Most of the tuning work is done by him.

Setup:
- 2.3l 8valve 4 turbo (~220 kPa max right now)
- Micrsoquirt v3 | 3.4.0 firmware (reloaded 8/17) | TS 2.6.14
- Wasted spark, running 4 d514 coils with coils paired on IGN 1 and IGN 2
- 60-2 VR flywheel input - stock
- 1000cc injectors
- NGK powerdex WB
- Lots of silly go fast stuff

For edification, the following are the major changes since it ran great on ms2: Microsquirt swap, new wiring harness, ignition upgrade, swapped to high flow fuel pump -6 lines and a bigger fuel rail, big sheet metal intake manifold.

The car was actually running fairly well a few months ago but I noticed high injector duty cycles under boost. Tracked it to a fuel pump which wouldn't push over 55psi and replaced it. Since then the tune has been more of an issue. I had tuned around the fuel pressure issue unknowingly so I was pig rich under boost. Also I was having a major lean spike during normal acceleration. Looking back in older logs I can see some of the same behavior but it was less drastic and less noticeable in the car.

The problem: I have a very consistent lean spike around 2800-3000rpm and 90-100kPa. It will go from normal acceleration sharp to very lean then drop off a cliff. We've also been trying to get the +5,000rpm part of the map nailed down post fuel change. My tuner has been gradually increasing the fuel in that portion of the VE table but the behavior is very specific and strange. Mostly we have narrowed the lean peak, knocked it down a bit, and get rich condition around it. Going up a hill holding at 3k it will sit lean to the point of breaking up unless I "push through" it and get the AFR to drop back down.

You can pretty easily see in the VE table the direction we've been going. Even with this big island of fuel I'm still getting lean swings around 3k, and the behavior seems generally inconsistent.



Rolling slowly into a 3rd gear pull



Consistent throttle, consistent MAP, gradually rising RPM and the AFR is very inconsistent.



This has got me, pretty confused. At first I thought maybe it was some VE "sweet spot" with the new intake manifold, but now it's just a hassle! I have done the following mechanical checks: Fresh plugs, calibrated all the sensors, verified steady fuel pressure/tracks with boost, compression check is OK, no big glaring air leaks, plug wires have swapped.

Any help or thoughts on what might cause this kind of behavior is appreciated. I've tried to do my due diligence with the mechanical stuff, something else I could check? Otherwise I am pretty green with the tune. I really feel like something is OFF, but don't know what it might be. Do I just keep working fuel into the table to try and smooth it out?

Tune file attached. My latest log is too large to attach so taking my first shot at google docs (will fix if broken)... https://drive.google.com/file/d/0B_LBSR ... sp=sharing

Thanks!
Mike

[Matt Cramer]

I would first try cranking up the VE in that range.

[VcrMiata]

Your observation is correct.

Your VE map needs tuning (smoothing).
Perhaps you caused this in an attempt to fix your problem?

It would be interesting to see what your pulsewidth is doing during this. It would be interesting to see if it stays consistent with VE.

Edit: removed some crazy talk due to too many wines.

[kaeman]

I noticed that your ego correction is taking fuel away from the engine at each of the lean spikes, try looking at ego corrections and see if the lean spikes are being caused by the ego correction settings.

[MikeK]

Hey guys,

Bumping this up because this is still something I am fighting on my Volvo. I have covered it in a bit more detail in this thread on turbobricks: http://forums.turbobricks.com/showthread.php?t=324033

As detailed in the OP, this condition presented when I converted the car from ms2>microsquirt v3. Last year, with the help of an experienced tuner via email, I got it to a point it was good enough and I just drove the car. It was never quite right at 3k rpm, but it was alright. I had attributed the difficulty truly smoothing it out to the inherent difficulty of email tuning combined with my really limited skills (I'm a much better mechanic..)

This year I have put in a fully built volvo 2.3L engine in the car along with a much higher flowing head and a bigger cam, really different than the old stock engine. As I was re-baselining the tune the 3k rpm headaches came front and center again. I have exhausted the help of my email tuner, and I have not had any success addressing this with the VE table. The basic issue is the same as the OP, as I come through the 3k rpm range the car swings lean at a variety of loads. Please note my wideband (ngk afx) has a range of 9-16:1 so it will peg the gauge lean and the car will hesitate noticeably.

I have had several people review the tune and dozens of log files and I have been told repeatedly it must be a mechanical issue. I have done the following diligence:
- Checked and re-checked grounds multiple times.
- Plugs, wires, and the ls2 coils have all been swapped/replaced.
- Fuel pressure verified and as expected
- Fuel injectors sent out to Injector Dynamics and confirmed good/cleaned
- Crank sensor replaced, and now running a BMW VR sensor in case it was an issue with the Volvo sensor.
- Boost/air leak tested and leaks addressed.
- Addressed a voltage drop issue
- Replaced a slightly sluggish o2 sensor with a new Bosch unit
- I even temporarily swapped my big plenum intake manifold for the stock Volvo intake. The favorite theory on turbobricks was some kind of flow reversion at 3k rpm due to the plenum.

While this was all time well spent, it's done nothing for the issue. I am at an impasse and getting pretty low on ideas. Here is some data.

Here is a typical behavior log:
https://drive.google.com/file/d/0B_LBSR ... sp=sharing

And the .msq to pair with that:
https://drive.google.com/file/d/0B_LBSR ... sp=sharing

Ref record 3385 for crossing that area under mid load and the obvious swing at 3100 rpm. 4825 on you can see it run lean under cruise.

Here is a current .msq. Obviously I (and autotune, and others) have been trying to increase VE in the relevant columns to flatten the curve. Generally, the car runs pretty well. Idles, cruises, and the top end is pretty decent. It's just the specific rpm that causes issues.
https://drive.google.com/file/d/0B_LBSR ... sp=sharing

Yesterday was the first cold day here, and the colder ambient temps really had the car running poorly.
https://drive.google.com/file/d/0B_LBSR ... sp=sharing

Ref record 1275 for a prolonged drift at low load. 191 for pretty bouncy AFR including a peak at 3300rpm, I found this interesting as it is at a higher rpm than I typically see (temp??)

In the garage in neutral, I tried holding the car up in that rpm range. Ran, not smoothly but will hold the rpm.
https://drive.google.com/file/d/0B_LBSR ... sp=sharing

I then tried doing the same and increasing VE and it did drop the AFR. Again it will hold but it is not running as smoothly as it does at a lower rpm.
https://drive.google.com/file/d/0B_LBSR ... sp=sharing

Summary:
I feel I have done a lot of mechanical verification without any red flags coming up. I believe the fuel side of things is good. There are two options here; there's nothing wrong with my setup and the tune needs a better set of hands, or I still have an underlying condition. At this point to me that would be a bad microsquirt, or an issue with the wiring harness or something that is causing an rpm specific misfire. I would happily bring this car to a professional but I'm unwilling to do so until I can be 100% confident I'm not bringing the chef bad ingredients.

Any help will be hugely appreciated. Like I said I've been pouring time and money into this and would love to see some progress!

Mike

[TT 03 Cobra]

Was this issue ever resolved?

[radbrad_dsm]

Also requesting if this was ever resolved. Have been experiencing the same exact problem ever since going standalone MS2 with 1000cc FIC High Z injectors. 2500-2800rpm I experience a HUGE lean spike, engine completely falls on its face to about 16:1 AFR. Both hard acceleration and steady state causal driving. I have been increasing VE in all cells within the range which has helped a little but seems like the wrong way to go about this. Some cells I've needed to increase +20! Only possible solution I've found elsewhere is to install a fuel damper to smooth out fuel pulsations within the fuel lines/rail. Another is switching to sequential injection instead of batch fire. Would really love to hear if this was solved. Thanks!