My recently acquired Formula seems to be running rich (based on the exhaust). One piece of the troubleshooting I'd like to work through is analyzing the data logs I collected today. I'd like to determine if anything in the data points to a running rich condition, or if anything else seems amiss. I'm on the hunt for some guidance on what to look for, but I figured it cant hurt to reach out and have a trained eye take a look if possible. I'd be grateful for anyone who could do that, and/or point me to some resources.

The car is a 1990 L98 A4, and as far as I know it's all stock.

I took 3 logs:
1-cold start with ~10 mins driving (the fuel pump had primed a few times before making this start while I tried to get my ALDU1 talking to the car and TunerPro)
2-started a new log with the car parked and running. Mostly cruising, a few stop lights, and a few WOT moments
3-idling in my driveway

Ambient conditions:
76 degrees, 71% humidity, 29.75 in pressure

Side note, I've done minimal mechanical testing so far other than listening to the injectors firing, which they all seem to be. I plan to test the fuel pressure (hopefully tomorrow) and see what the plugs look like once I get a little more time to dive in.

 

This exchange between Traxion and RBob is exactly the problem I'm having. My exhaust is not bearable, but mine does have a bit of fuel smell too.

https://www.thirdgen.org/forums/diy-...ml#post1610915

 

For me, if you can export those logs from tunerpro to xls I can take a look.

 

Thanks! I've added them to the first post.

 

I just looked at your "warm" log. Your BLM is 140, INT is 129. When you start to move BLMs go even further LEAN.

You also have quite a bit of knock at low throttle.

Two big red flags. I wouldn't drive the car much until you sort it out.

 

Thanks for looking. So, I assumed my smelly exhaust was rich but I'm lean, interesting. I still need to test my fuel pressure and go from there. Also, this jogged a memory: the PO claims he advanced the timing to get it to run faster. I'll check base timing and reset, then see what it's doing for knock.

Two questions: (Editing after doing a little more thinking/reading)
1) Can you confirm I have this right? A quick google tells me that both the BLM and INT targets are 128. If INT is >128 = computer thinks there is a lean condition and is calling for more fuel, and INT is <128 = computer thinks there is a rich condition and is calling for less fuel

2) Is the rapid variation in the O2 sensor voltage normal? If not is it safe to assume it may be related to the high INT/BLM condition? In the graph below O2 is dark blue (red is TPS, black is INT, maroon is BLM, green is RPM)

 

Yes on BLMs.

o2s looked good to me when I looked at them. The fact that the car is learning/closed loop/crossing o2s is a good sign.

the TPS, MAT and MAP looked normal as well. So I don’t think you have a sensor issue. I did NOT look at your idle map readings. That could help tell you if you have a vac leak.

double check your distributor timing advance and make sure it matches what your tune says you have it set at. I leave mine at 6 degrees, but others don’t.

after that check for vacuum leaks and exhaust leaks.

 

Also if you’re speed density you may have interest in my custom OS.

One of the several improvements I made it will flash the check engine light at you when you have knock above a tunerpro set threshold. i set mine to .7 degrees. And it’s very helpful to have.

 

Thanks, I have plenty to check on. I'll grab some new data once I fix any identified issues.

My car is SD. I'll revisit this once I get a little more immersed in the tuning thing. Thanks!

 

I did not go beyond the first page in that link, but Traxion is completely wrong with that first claim. The speed density code is absolutely based on 14:7.1. The entire fuel model is built around a single assumption that Stoichiometric AFR equals the AFR at which the O2 sensor switches. That's it, that's the whole design. SD code calculates BPW using; stoich AFR constant, VE, MAP, IAT and injector flow rate. 450mV is not an AFR target, it is the bias point used for O2 cross count detection, Integrator direction, Rich/Lean switching logic. The ECM does not aim for 450mV, it is just the mathematical midpoint of the O2 sensor's switching curve. GM could have chosen 400mV or 500mV, it wouldn't change the AFR target. If the O2 tables defined a different AFR target other than stoich, BLM's would drift endlessly, INT would never stabilize, a change in the stoich constant would break closed loop, but more importantly... the ECM would not be able to maintain emissions compliance.

- Rob

 

The way I’ve always seen it:

Two data points.
One tells ecu the o2 mV target to achieve desired afr.
The other tells the ecu what that mV is supposed to represent in AFR.

You can change the mV to fix the sensor if a wideband says it’s actually targeting 15.0.
Or you can change the 14.7 value so that when the ecu wants to target 13.5 it knows how far off it is from that based off the AFR you tell it the mV actually represents.

I could be wrong but it’s made sense to me.

 

Thanks. This is some good info. My reference to that thread was more about the discussion of the exhaust smell that and permeates my clothes and skin. Whenever I drive the car Im usually ready for a shower afterwards, especially if I have the windows open which is most days since the AC does not work right now.

I'm leaning towards an exhaust leak causing the issues (high BLM and INT) due to said exhaust coming in the car when the windows are open. Also under load the exhaust sound odd, so I'm thinking its leaking out of the hacked up welding job done to the gutted cats on my car. It's got a scratchy "husky" sounds for a lack of better words. I know what typical exhaust leaks sound like but this is different. Either way it needs cats so I have 2 cats on the way and a new O2 sensor on hand, just because it's cheap enough and I'll have a wrench right next to it. I still need to run the fuel system tests, after which aim also going to install a new fuel filter.

 

Unplug the O2 sensor and force it to stay in open loop and see if that helps any. ​​​​​Set your timing to the factory base timing and then unhook the battery to reset all the learned adjustments. Unhook the O2, hook the battery back up and then try it out. A leaking AIR tube or malfunction in that system can throw the O2 off as well.