MS3 started leaning out at high rpm, fixed after restart

[lowgman]

I ran into a frustrating issue while street tuning today. I had a good 10 minutes of tuning at higher RPM and high load (going uphill). I left auto tune enabled on Tunerstudio, and went for the drive.

At near the 10 minute mark, my car started hesitating, and eventually starting going full lean at any rpm above 3000rpm. I pulled over, idled fine and then tried to drive again. Same thing, anything after 3000 or so, it would lean out and start bucking.

I pulled over again, looked at the dash on tuner studio, everything looks normal. I shut off the car, started it up again and everything worked! Car felt totally fine to redline.

Unfortunately, I wasn't data logging. I will attach the msq later on in the evening.

Anyone have any ideas what may have caused that, if tuner studio makes lots of changes during the drive in auto tune, is there a limitation?

I was thinking maybe fuel pressure, but then why would it have been fixed by a restart of the car/MS3. It seemed like something was wrong the the ms3 to me.

[billr]

I'm guessing you had VEAL set so that TS was not doing a periodic "burn" to the VE table in ECU memory, so those changes were lost when you powered down.

[lowgman]

Hmmm, I'll look at the settings for that. However, I can't imagine my tune being that far off from the pre-auto tune settings that the car would be undriveable.

[lowgman]

Hmmm, I'll look at the settings for that. However, I can't imagine my tune being that far off from the pre-auto tune settings that the car would be undriveable.

I've attached the current tune and a datalog when the car was acting more normal.

I noticed that the voltage does drop down to 12.5V at idle when I have my headlights on from a healthy 14.x when the RPMs are higher than idle. Is this a sign of a bad alternator? I'm assuming the voltage does something to the PW calculation?

[atomic6]

Almost guarantee it is a fuel pressure, problem.

Check pre and post filters.

When you restarted the megasquirt you restarted the car(right), some times the filters get clogged, and when the pump is turned off, the "junk" falls out of the filter media, then it takes alot of flow to "suck" is back in. So idle and part throttle are fine again, and it seems like it is fixed, but push for some power, and the problem comes back.

I have encountered, this symptom a few times, and the first couple times, I though the MS was doing something wrong, but it always turned out to be Mechanical problems.IE fuel pump, fuel filter,injectors,fuel pressure regulator, etc.

[lowgman]

Hmm, it's a brand new fuel pump/in-tank filter (I hope it's not the in-tank, that is a NIGHTMARE to get to) - I could change the fuel filter in the engine bay easily, so I will start with that.

I am thinking I should splurge for an adjustable fuel pressure regulator with a gauge, just if it ever happens again, I'd like to know.

[billr]

"Splurge" on a sensor for fuel pressure instead, then you can log FP and see what it is doing at all times (without taking eyes off the road to baby-sit a gauge!).

[lowgman]

150905-LeanCondition.msq

Welp, I thought my problem went away. I installed a new fuel filter and changed the fuel pressure regulator. I didn't end up buying a fuel pressure sender unit and wiring up the fuel pressure sender to the MS for logging, because I thought it fixed it.

I took a drive again today up some mountain roads (the same one when I experienced the problem back in March)- and sure enough at about 2500ft of elevation, my car exhibited the same problem!! It really hesitated to go up the mountain, the conditions were at full lean when I gave it more than 10% throttle. I had cyclists beat me up the hill!

I pulled over to a safe place, and then pulled out the laptop to log it. I have a short log attached to this post where it exhibited the problem. You can see when I gave it throttle, the AFRs go to full lean.

I shut off the car, restarted it, and everything was normal again! I logged a few minutes of driving after to see if I can find some differences

Now it got me thinking about a few things:

[1] I noticed that the Baro Correction was at 90.1 on the second start near the top of the mountain vs 100 when I started the car at sea level. Does this make that much of a difference to throw off the fuel calcs??

[2] The PW of the injectors seem to around the correct area (~5ms) when the full lean condition happens. Makes me think that it's still a fueling problem.

The engine is a 20V 4age, with open air ITBs. Could it be possible that the fuel pressure regulator doesn't work well during high RPM usage? It's vacuum driven, but when I have the throttle plates open past 30-40% for an extended period of time (climbing the mountain) would that cause a massive dip in fuel pressure?

[3] Could this be an issue with firmware? I am running MS3 1.2.4

Thanks in advance!

[lowgman]

It just hit me! Last time, it was pretty close to the same place on the mountain where I had this issue before. I looked up the altitude and it's about 2800feet, which translates to about 90kpa, which is my ITB switchover point for my ITB mode on MS3

Could this be it? It's pretty consistent. Unfortunately, it's about a 45 min drive to get to the mountain. May have to wait till next weekend before I can test again.