• Hello Guest, welcome to the initial stages of our new platform!
    You can find some additional information about where we are in the process of migrating the board and setting up our new software here

    Thank you for being a part of our community!

lean spike in boost, just like lh2.4 except this is ms3

740atl

740atl

Gear acquisition syndrome
Joined
Apr 17, 2010
Location
USA
Gang, I have a curious problem that's appeared.

Tuning the 16v+t at about 75% ethanol in the tank.

This is a flex fuel setup where I blend VE1-VE3. I tuned the VE1 on pump gas and multiplied the entire map by about 50% to get it right until the blended afr matched what I was seeing on pump gas. Needless to say, I never saw the lean spike on pump gas.

This thing is spiking hard lean when I slam on the throttle. pumping up the VE in those spots on the MAP does not make any changes. THe lean spike is dependent on what RPM I floor it. It reminds me of my lh2.4 days. I've already tried playing with acceleration enrichments but that had no effect.

I noticed the spike in the logs when I started tuning with e85. Now that I'm at 75% the spikes are more pronounced. I've never been this high before... normally I only ever get to 55% (I live 45 min from the closest e85 station).

I tried a bunch of things.... decreasing gap to 0.022 (a 0.55mm guitar pick) thinking it was spark blowout or a lean misfire, nothing doing. I could perhaps go smaller on the gap. (dsm coils)

Posting a datalog and a snapshot below.

I'm not logging fuel pressure... I don't even know what the fp is... assuming it's all kosher but I am going to pick up a transducer and start logging.

Your thoughts?

bad regulator? dying pump? Spark blowout?


167686727.jpg
 
Last edited:
I'm no expert with e-85, but I do know I had some issues with too much fuel causing detonation and apparent lean spikes on the afr gauge, there was a post by Bobbyxyz that explained that the afr gauge is actually a lambda gauge and that too much fuel can give a false reading.
 
Actually what's the spiky line on the same graph as the as afr? It's that the e-85 concentration? I think I remember seeing a thread about crazy readings from the alcohol sensor causing all kinds of mixture problems.
 
Fuzzy line is knock input.

The deal with e85 is that you leave the wideband alone and just multiply the ve table by 40-50% until the afr you’re expecting shows up on the wideband. There are a number of ways to do it and I won’t go into it. You can’t just dump in a bunch of e85 into pump gas and expect everything to be fine. To go from e10 to e85 like I said you need about 50%or so more fuel. Do that on lh2.4 without any changes to the map and you’ll run lean enough to burn a piston.

That’s not what’s happening here. My pump gas and e85 maps are sorted out and blend together based on the ethanol percentage.

I’m thinking the lean spot I’m seeing is either related to a pump that can’t keep up or a bad regulator or spark blowout. The lean spot only happens when you floor it, not during the rest of the run. It moves around to different VE cells depending on the rpm you floor it at. Normally I’d say this was an acceleration enrichment problem but this happens after the enrichment is over. I’m typing this on the cell else I’d go into more detail on the e85.
 
I need to see TPS and MAP, to see the other influences.

it appears the transition from NA to boost tossed a huge surge or over correction into the mix.
Maybe because it is loosing spark advance it goes richer and then leans over correcting.

If you are tuning with a 31% authority EGO correction enabled that might explain why your VE tables look so choppy.
Turn off EGO correction and AE when tuning so you are not doing battle with the enhancements.
eg: the afr table is lean and you throw fuel at it in the VE table to try and bring the afrs down, then O2 correction takes out the fuel you just put in and the motor goes back to lean

Your AFR tables are lean in the idle area, #1 is way lean in cruise. And both awful flat.
 
VB242 said:
I'm no expert with e-85, but I do know I had some issues with too much fuel causing detonation and apparent lean spikes on the afr gauge, there was a post by Bobbyxyz that explained that the afr gauge is actually a lambda gauge and that too much fuel can give a false reading.
Click to expand...

There is no AFR gauge it is displaying Lambda with some math, an approximation.

And because of this when you get enough fuel in it to cause misfires the gauge will show lean.

A real AFR calculation should take into account both HC and CO along with O2.
 
Thanks! I'm still learning about tuning and I'll file this one away as: "if the AFR doesn't match the PW, check the pump!"

Looking at your graph, the peak PW of 13.8ms and 6100rpm means that the injectors are on for ~70% of the time (please check my math). If you've wired the injectors in banks (firing order of 1-3-4-2 so banked as 1-4 / 2-3, I think?), you might try changing to 2 squirts/cycle alternating. This should even out the injection a bit. By definition, 1 squirt per cycle means that each cylinder sees injection over a different span -- e.g. cyl 1 sees injection from compression through exhaust, cyl 3 sees injection from intake through power, etc.)

What's your cam and intake/exhaust setup? Turbo appropriate or +T?
 
You must log in or register to reply here.
Back
Top