rebuilt engine fluctuating AFR

[ollykiwi3]

hi
I have rebuilt the engine over the winter. now 306Ci ,10:1, AFR 165 heads, X303 cam, 40lb bosch Inj Hiz.
this is the first start up ive done.car started fine and had to do some quick fuel tuning for idle.
tried autotune ,but it didn't really make a huge difference.
its running crazy rich. pumping out black soot. I tried pulling fuel out and got to point of it being lean and almost dieiing but it still seemed rich.
could 40lb inj be too much and getting to non linear? I researched a lot on inj, and most people seemed to think they would be ok? running 45psi. maybe run it at 42 to see if I can get more fuel out?

im using a innovate MTX wideband and the reading is stable ish and kind of what I expect to see, however on TS dash the afr is swinging rapidly and varying by at least 3 points( 12-15AFR) while the MTX is fluctuating but no where near as much or as wild(11-12.5) and also a few points lower. looked at the real time display and checked AFR volts and they were varying. it seems the MTX is correct, as it running rich, but TS dash seems really lean. checked earths,all seems ok.
so I guess question is , is it the volts coming from the sensor(have tried second sensor with same results) doing this or the output from the MTX to MS ECU, which is displaying the volt change?

engine is running 15"vac(50kpa) and pretty unstable right now around 950-1000rpm. it seems like ign not stable. sounds like it maybe missing. ign timing set using wizard with no problems. have tried using timing light on leads to see if they are consistant, and sometimes it seems to miss a beat on the flash but not always.

tables are rough at this stage but just trying to get idle sorted for now before driving. was hoping to use my original table from before the rebuild and tweek it, but it just didn't work, so now have to start again.

I guess the autotune wont work too well seeing as the AFR swings so much.

any thoughts on basic setup at this stage would be great.
cheers



engine run

[elutionsdesign]

Your injector time at idle is about 1.4 mS and the dead time for the injectors is 0.98 so you are in the non-linear injector region, that isn't going to help but I'd try a few things first. And double check those injector specs to make sure your dead time is correct.

First, you have to make sure the AFR reading on the gauge and in TS match, the gauge is generally correct and TS must be made to read the output correctly. Make sure the sensor ground(not power ground) of the gauge goes back to megasquirt like the CLT, MAT and TPS grounds do already. Pick the correct wide band in TS, if you can get it within a couple tenths of an AFR you can work with it, you may need to select "Custom Linear WB" and put in some factors to get it reading correctly.

Second, timing, I assume you've used a timing light to sync timing between the controller and the physical timing on the car? That is done by setting the timing to fixed timing and I'd use something around 10-15 degrees. Then leave it as fixed timing for now until you get a smooth idle, it takes one more variable away as having your timing oscillating can easily contribute to the MAP cycling and subsequent AFR cycling.

[ollykiwi3]

ok, so I had selected custom wb and entered manuf data.
all grounds are good and going back to MS grounds.
the datalog attached is sitting with nothing running. the MTX gauge reads 13.8 AFR, and I have a multimeter on output from sensor, and its reading 2.54v and pretty stable, which should equate to 14.7 AFR. the TS dash shows rapid volts change from 2.1-2.8v (14-16AFR) at the same time as multimeter reading. something inside ECU is triggering voltage fluct on TS dash?? input into ECU at DB39 plug is stable voltage.
so the MTX gauge is off for starters I think., however, if you look at the exhaust when running, it would suggest the MTX is right?
I checked it in free air and was reading 5.56v and MTX said 22.4 which is correct. I fired the engine once to get the above readings.
I have checked plugs,pins,wiring,solder joints,and all connectors for anything wrong, but all seems ok.
where could such a voltage change come from??
it would seem that the TS dash reading is more in line with sensor voltage and AFR chart, but I need to make it stable.

If I was going to look inside the MS ecu to find an issue with this voltage swing, where would I start??? I built it myself, but don't consider myself an electrical guy by any means.

timing was set at fixed timing, but I then switched to table. will try again with fixed timing. inj dead time was just a generic start point. I did change it up and down to see when it got better or worse, and it seems reasonable where it is by sound only.

[ollykiwi3]

just run another log idling with AFR sensor disconnected seeing as its not working anyway, and run it at fixed timing of 10deg.
seems reasonably stable and have dropped fuel pump press back to40psi to see if that helps rich idle.does seem a little better.
have dropped a little more fuel out of it at idle. if I drop more it becomes irratic.
checked timing with light and is pretty close but it does fluctuate a couple degrees . maybe timing chain slack? it is a new double row set.

im not sure what to expect with this cam combo, but everyone seems to have a different idea about how smooth it will be and expected vacuum.
it seems a little rough up to 2000rpm, but haven't gone above that yet, and yes its stationary.no drive yet as im unsure about timing and afr right now.

although the tables are rough, does anyone have an opinion on the timing?does it look kind a close?

fuel will be sorted by veal hopefully if I can sort MS output voltage issue with AFR.
cheers

[elutionsdesign]

You are running an older firmware and it might be a good idea to download and install the latest version, it has some additional noise filtering for the various inputs. Do you have a stim so you can better test the O2 circuit?

[ollykiwi3]

Does the ms controller use the same data it displays on the dash or does it use the raw data from the sensor?
In other words is what I'm seeing on the laptop dash the same as what the controller is using or is it displaying another output with regards to afr voltage?

Will update firmware tonight and can borrow a stim.
Thanks

[ollykiwi3]

Ok. So it seems that new firmware has sorted the oscillating afr voltage issue. Lot more stable now.
The new layout has taken a bit to get used to in the ts dash.
The mtx gauge is still out but based on the voltage output from the sensor it seems it always will be.
The actual voltage and what ms sees is a lot more in line with the linear chart theoretical numbers so I'm happy that ts is working ok.
Cheers for your thoughts.

[elutionsdesign]

The answer to the question is yes, what you see in TS is what the MS controller is seeing, not sure why your MTX is reading differently. There are settings for the MTX for different fuel types, you might check and make sure it is set for the application.