Page 1 of 2

Closed Loop begins too early

Posted: Sun Apr 29, 2007 9:08 pm
by Synoptic
If the car is hot and closed loop on. turn the car off, let the Stock O2 cool down for 1 min. Then start the car, the rom will go into closed loop since ECT is in the closed loop range, but since the O2 sensor is still cold and not yet heated, it will read rich and the ECU will start removing fuel, leading to a stall. Until the O2 is hot again, the engine will want to die at idle. I have a couple of customer having the same problem. Calvin, I think it has something to do with the code you added for closed loop Wideband... Maybe we should have a Voltage range for Closed loop operation, and not simply a target. That way, when the Stock O2 is cold an dreading rich (3.8v), do not go into closed loop, until Voltage is in the 0 - 1v range.

Posted: Mon Apr 30, 2007 1:02 am
by xenocron
Are you saying that a stock O2 sensor is sending more than 1 volt to the ECU?

Posted: Mon Apr 30, 2007 6:45 am
by hackish
Normally there should be a minimum run-time before enabling closed loop. This is how it is most often implemented. Maybe we can only enable closed loop after the engine has run for 60 seconds. -Michael

Posted: Mon Apr 30, 2007 8:03 am
by Synoptic
The dataloging view shows > 1v when when Key On Engine Off while O2 Sensor is getting colder, until it reaches 3.83. I know Calvin had to remove a lot of 1.5v checks in the cose to enable closed loop wideband. Maybe he needs to put some of those checks back in the code when having stock O2 installed and closed loop enabled.

Posted: Mon Apr 30, 2007 12:50 pm
by calvin
all are put back if \"use closeloop wideband\" is unchecked

Posted: Mon Apr 30, 2007 2:59 pm
by Synoptic
calvin wrote:all are put back if "use closeloop wideband" is unchecked
Then, what cause the problem ?

Posted: Mon Apr 30, 2007 4:26 pm
by Gaskleppie
Got the same.... min ECT is not enough. I think a timer would be best. Some WB will react different than others. some will give 3.5volt, some give 0volt at the heating period. A timer wil always work.

Posted: Mon Apr 30, 2007 5:02 pm
by nate
Sounds like we need something similar to the O2 FB Time Vs Temp table in AEM EMS?

Posted: Thu May 03, 2007 2:29 pm
by calvin
i think i found the bug. I want a tester that can test narrowband closeloop.

Posted: Thu May 03, 2007 3:32 pm
by Gaskleppie
kom maar op!

Posted: Thu May 03, 2007 7:03 pm
by calvin
sending it over

Posted: Mon May 07, 2007 7:56 am
by calvin
I hope this is fixed... I need testers

Posted: Mon May 07, 2007 8:24 am
by Gaskleppie
Tested! Seems ok now! I started my cold engine. First 3.8volts, When the sensor was warm it went to 1volt, still no reaction on the o2 trims. Waited till engine was hot, stil no o2 trim. Touched the throttle to drive away, then it went in closed loop. With warm engine it was the same! No closed loop until throttle movement. I think this is the same as normal. Last month I watched a guy doing a dutch smog test. On al stock Honda cars he had to touch the throttle before the smog readings were ok, otherwise o2 levels were way to high (open loop). Now I know why he had to do that! ;) Its in the stock code! This was a NORROW BAND test! So now we can work on step 2, the wideband closed loop test.

Posted: Thu May 10, 2007 8:09 pm
by calvin
fixed hopefully:) nick test

Posted: Mon May 14, 2007 10:03 am
by Synoptic
Saturday I tuned a car using eCtune r5 and the included 0.40 base bin in the wizzard. When hot, the car had no problem. If I turned the car off for like 2 - 3 mins, the AF would go lean, even if closed loop was disabled. Kinda weird. the leaning happens AFTER post fuel, and before some other point. I dont know why the car would lean-out itself like that. I'll do more testing this week as I have a full schedule of tuning.