Trouble with new Megasquirt (fuel or wideband most likely problem child)

[digger]

Ok, here's my first observations/suggestions. I think you ARE indeed going lean. AFR starts creeping up just as soon as afterstart enrichment goes away, and your VE is in the low 30s, which I certainly don't consider 'high' for an M20:
Next, DISABLE closed-loop idle and EGO feedback. At least for now/while tuning, you want as few variables as possible to simplify things first. Add the fancy features back in for FINE tuning once you get this close on open-loop.
Thirdly, you can't *really* run pure Alpha-N with an idle valve. As the idle valve moves, engine load changes, but since the ECU doesn't 'see' the TPS load change, it keeps fuel the same. You can see this as the abrupt drop in AFR right when you idle valve closes down some, while PW stays constant @ 4.4ms. You can sorta fudge around this by carefully compensating warmup enrichment against your idle valve position, but the 'right' way to do it is to change from Alpha-N to ITB mode.
ITB mode uses a 'blend' of TPS and MAP signals for load, but it involves a bit of work to set up the ITB Load Settings curves and redoing your VE table accordingly. Will give you the best result in the end though, so I'd recommend it.

One way to use AN is just add a fuel compensation table so when the IAC moves it adds or subtracts fuel, this also helps limit the cold start enrichments that get built into that table because the ECU doesn't know the load is changing as IAC opens up to create the fast idle when cold. Its not perfect but helps alot.

[Austrianvespaguy]

Yeah that's fine, and you can certainly run just alpha-n while 'running,' it's just tuning idle together with the idle valve that gets tricky!

[whodwho]

...you can't *really* run pure Alpha-N with an idle valve. As the idle valve moves, engine load changes, but since the ECU doesn't 'see' the TPS load change...

Good call I had forgotten about that, it has been a few years since I ran my ITBs and I was setting it up for boost and only ran Alpha-N very briefly to get them up and running before changing over to ITB mode so I didn't fight that battle.

[Austrianvespaguy]

Ok, here's my first observations/suggestions. I think you ARE indeed going lean. AFR starts creeping up just as soon as afterstart enrichment goes away, and your VE is in the low 30s, which I certainly don't consider 'high' for an M20:
Next, DISABLE closed-loop idle and EGO feedback. At least for now/while tuning, you want as few variables as possible to simplify things first. Add the fancy features back in for FINE tuning once you get this close on open-loop.
Thirdly, you can't *really* run pure Alpha-N with an idle valve. As the idle valve moves, engine load changes, but since the ECU doesn't 'see' the TPS load change, it keeps fuel the same. You can see this as the abrupt drop in AFR right when you idle valve closes down some, while PW stays constant @ 4.4ms. You can sorta fudge around this by carefully compensating warmup enrichment against your idle valve position, but the 'right' way to do it is to change from Alpha-N to ITB mode.
ITB mode uses a 'blend' of TPS and MAP signals for load, but it involves a bit of work to set up the ITB Load Settings curves and redoing your VE table accordingly. Will give you the best result in the end though, so I'd recommend it.

[whodwho]

Hey all, here is a bit of recap and observations of my involvement

Just in case someone is reading this when installing ITBs it was the IAC and the TPS connections that were switched, the IAC reaches the TPS way in the back much better unfortunately 8^(

He was originally running a M50 TPS without switching pin 1&2 but would surprisingly calibrate with a decent 0-100 sweep. I don't remember it calibrating that well when I was originally running a M50 TPS without the pins switched. After finding the switched up connections I fixed the idle control circuits and gave him a used TPS and an adapter I had on it now and calibrates about the same but a bit smoother signal.

The first major issue I was seeing was false accel enrichment at idle with no TPS change, I increased the TPS lag and leaned out the accel settings as it would spike from ~5-6ms to 16-18+ with no throttle change and would die out. Leaning out the accel made that a bit less issue but aggravated the tip in as it was a bit too lean on the fuel mapping in the lower load r3v range.

Seeing the O2 going to the extreme lean and smelling rich we figured it was excessively rich and leaned it out until it stalled and then gauge never registered again until a restart.

A quick test to rule out that it was still not controller related after the short we tried my old MS3 prototype that I was running on and changed the req fuel and getting similar results

I ran his ECU in my car and would run ok with a similar stumble which turned out to be lean cells in the r3v range and roughly richened those up to test it on his car. I didn't have the false accel enrichment with and without the TPS adapter and verified that it had the pins switched correctly on the adapter.

I gave him my O2 sensor and controller, the controller was the same but the sensor unfortunately has a different connecter.

He swapped out the injectors back to the stock ones so we have similar test beds and we are back to his ECU on his car and did some tune by ear/nose and have it running and r3vving better but far from tuned. With the accel leaned out and the fuel map richened up we aren't getting the big enrichment spikes but are seeing the TPS signal changing by 1-5% not like noise or in a steady fluctuation but in jumps up and down and is screwing with getting the tune dialed in any further. We noticed that pushing on the throttle linkage it would idle down and TPS would obviously change and about the same as we were seeing the fluctuation.

He got his laptop all connected and software installed and took a couple log sessions so I still need to get the logs from him and give them a look over

So he is getting a new TPS sensor, O2 sensor and looking into the linkage issue and we will give it another go, I want to hear this rip! 8^)

[firemanory]

Current Msq car is running. https://drive.google.com/file/d/1n53...ew?usp=sharing

[firemanory]

Here is last 2 logs from warmish startup. Clink Google Drive link below. Fuel pressure is 38ish at prime, 40-42 psi when running and add some throttle. One thing I noticed was that the TPS gauge on Tuner Studio bounces around when car is turned to on but not running. Goes from -1.0% to 4% before I turn the car over. As you will see AFR goes from 13-14 ish and bounces up and then down and then car dies. Will be swapping out O2 sensor next and then run car and log and upload here

Best thing you can do to get some real answers rather than more guesses is to take a datalog while it's running rough and upload/post it here, along with a copy of your current .msq tune to check out. Then we'll have some true data to base further recommendations on.

https://drive.google.com/file/d/18a1dXy5BbpnzbyBXBEwUC-ICmquSt_rW/view?usp=sharing

https://drive.google.com/file/d/1S-ZtOfXohHy-Ks_yhfGxPwDJ1WLGN6pC/view?usp=sharing

[firemanory]

Running alpha-N, so just IAT & variable TPS. Will get a datalog posted up later this week. Thx for help guys!

If you have a MAP sensor then it's probably shot anything in the 5V circuit might be damaged.

[Austrianvespaguy]

Best thing you can do to get some real answers rather than more guesses is to take a datalog while it's running rough and upload/post it here, along with a copy of your current .msq tune to check out. Then we'll have some true data to base further recommendations on.

[ADEN]

If you have a MAP sensor then it's probably shot anything in the 5V circuit might be damaged.

[firemanory]

We did remove the O2 sensor as an input from tuner studio and it still didn't run well. Have the AEM wideband that was brand new... wondering if when we backfed the 12 volts from the wrong harness connector (I thought was the TPS connection but was actually emissions one) and it fried the TPS it also fried the wideband. Won't be back home to try it till Thursday so we'll see the. Will get a log to post up then as well.

I would suspect the wideband sensor or controller (or both, if it's junk like Innovate). Checking fuel pressure is a good idea as well.

Run a data log and post it here. Posting the MSQ is always helpful, too.

[firemanory]

It doesn't do anything very well, but it does run and idle, rev it up and it mostly wants to die or it hiccups a bit and then gets some motivation to throttle up a bit but definitely can't hit the higher rev's without dying.

Forget about the O2, Is the car running,idling and free revving smoothly?

[firemanory]

Yeah, we changed required fuel to suit the 19lbr's. We were guessing that the rich condition was actually showing up as lean as well.

Have you updated the required fuel to suit your injectors?

Remmer that a rich misfire can show up as lean. As the air isnt burnt so the wideband will report lean.

[Digitalwave]

I would suspect the wideband sensor or controller (or both, if it's junk like Innovate). Checking fuel pressure is a good idea as well.

Run a data log and post it here. Posting the MSQ is always helpful, too.

[ADEN]

Forget about the O2, Is the car running,idling and free revving smoothly?