It did have a knock readout of some kind but it never triggered even when I could hear it pinging. I tested the knock sensor when the intake was off and it seems to be good, so this is another reason I suspect something deeper then a basic sensor error. That or the knock detection sucks on these ECU’s.
In my 4runner the knock detection is really good, you can get away with an amazing amount of abuse because of it lol.
I might toss a mechanical vacuum gauge on it and see what it reads, although so far everything has checked out as matching the OBD readout so loosing hope.
Going to try putting in my old spark plugs / coils today (have not been able to bum a ride to the parts store) to see if it shows any improvements, the plugs worked fine when I replaced them.
After this, sadly going to have to put it out to pasture for the time being so I can work on the 4runner, I am out of ideas for the Ford outside of the ECU itself.
Well, just swapped to the old coils and plugs, no change. I would of expected some kind of change if they were the issue.
The really strange part is that once again, it runs perfectly for the first few seconds after a cold start when still in open loop. Then it goes into closed loop and the issues start.
Pretty much rules out any mechanical, fuel, clogged exhaust or timing issue as those would effect open loop as well.
Just replaced the MAF. I could not find my vacuum gauge but the readings seem on point for the MAP sensor or at least close enough to not cause this big of an issue.
Anyone else have any bright ideas?
Only thing I can figure at this point would be the ECU itself but that is a lot to spend on a long shot. I have replaced the ECU in a few cars in the past with strange issues and it never worked before.
Hard to say exactly as I have not timed it, ~15 seconds or so I would guess today from a cold start. When it is already hot it is shorter.
I was watching the OBD at one point when it left open loop and the instant it did is when the issues started.
I have noticed this since I got it, just not as pronounced. When I would first start it up in the morning it would be really peppy and drive great until I was at the end of the block, by the time I took off from there it would be back to it’s old down on power ways.
Okay. Well let’s try and force it to stay in open loop for an experiment. Try disconnecting coolant temp sensor and see if that keeps it OL. If not, then disconnecting the upstream O2 sensor should. These will obviously cause a CEL but this can really point us in the right direction
EDIT: if it only took about 15 seconds from cold, then it’s likely the heated o2 sensor determining the switchover point. Coolant temp may not help
Interesting idea, I was wondering how I would go about forcing it into open loop but I worked with standalones or tuneable ECU’s for so long I never really learned what would cause a stock ECU to do that.
I will try the temp sensor first, assuming I can get at it. I remember disconnecting a plug next to the tstat when doing the intake manifold and not sure I can get to that plug with the manifold in place.
It was hard to get exact before/after numbers but it seemed to read a bit lower on average. It would still go to 100% with some throttle but seemed like it took a little more then before.
Hmm. So a minute of googling tells me there may be a fuse for the o2 heater circuit. That was on the v6, but the i4 may have that as well. That can be another way to disable/delay closed loop.
Nope, no codes which I found strange as well, although maybe I didn’t leave it running long enough, I can check for pending codes next time I am out there.
Gonna let it cool off and try disconnecting the O2 sensor tomorrow most likely.
Oh that’s interesting… both getting no code and the idea about neutral/park.
As for the O2, I think I see the connector on the firewall behind cylinder 4 in an image search. Should be another easy one to test. If it goes into closed loop without an AFR sensor, the ecm actually becomes an suspect in my book. As of now, I’m looking elsewhere.
Open loop does 2 things: generally runs rich to warm up things in a hurry from a cold start, and (duh) uses open-loop preprogrammed values for everything. Once you go closed-loop, it leans out and uses soft trim-values instead.
So if open-loop is good, more fuel means better acceleration, etc., and those values are programmed at the factory, normal expected values but not trimmed for the specific engine.
Remember the old (rather stoopit) trick of putting a wedge under the accelerator for “better fuel economy”, simply to keep people from stomping hard on the accelerator pedal? Sounds almost like that’s what’s happening closed-loop, ie, something’s tricking the ECM to limit fuel-flow somehow? (Eg, load sensor, etc.?)
I forgot if your engine runs rough or does anything untoward, other than just having a lack of power.
I heard of cases where the VSS or cam/crank sensor, etc., “jumped a cog” and threw off valve timing. Ie, the sensor might read a good timing value, but it’s just off enough to let the engine run, but be horribly weak.
Case in point, miscalibrated temp sensor on Andouille-based lights. My new SC31pro was reading and blinking out 44C-45C, but it was really 23C (then slowly sneaking up to 25C from just sitting in my hand when I was playing with it). So it was thinking the light was already running pretty warm even though it was sitting in rather cool air undisturbed. But!, it was doing exactly what its sw was telling it to do.