I am new to posting on ETCG, and I tried to search for answers to my specific problem, but I can’t seem to find what I am looking for in the search function.
The code 43 and high long and short term fuel trims have originated from a swap of a 95 obd1 GS-R motor in a 91′ obd0 DA Integra.
I used a speed factory obd0 to obd1 jumper harness and spliced in everything accordingly; Also, I am able to see a live reading of the short and long term fuel trims via the Hondash bluetooth scanner which has confirmed the very rich mixture that I can smell at idle. The short term fuel trim goes as high as 46.9% and the long term fuel trim is at 23% right now.
A very interesting occurrence happens when I drive the car at about 65 mph and over 3k rpm; after a couple minutes of those driving conditions, the short term fuel trim will go to a constant 1.6% and read from the O2 sensor voltage (closed loop), until the engine is turned off. But after turning the engine back on immediately, the computer for some reason goes back into open loop and richens the short term fuel trim again, its like the computer completely forgets the calibration that it did to get the short term fuel trim to go to a constant 1.6%. I have double checked that there is constant power going to the D1 pin for the P72 ECU for the “backup” power, which I believe saves the long term fuel trim and what the computer has learned after operation? At first I thought that the computer was not getting power or that there was something wrong with the computers ability to retain learned values, like the long term fuel trim, to reference every time that the engine was started; however, I sent the P72 ECU off to HA Motorsports and they confirmed that the ECU is functioning property and that it retains all of the learned values like it should.
I believe that the short term fuel trim should constantly fluctuate within a +-5% at all times when the computer switches to closed loop, but maybe that’s different for obd1? And the long term fuel trim should be as close to 0% as possible?
The obvious suspects like the O2 sensor, leaky injectors, and inappropriately spliced wires have already been addressed (using a Denso O2 sensor, replaced injectors, and double checked the wiring and pinouts to make sure they correspond correctly.)
What I have noticed now is that when I drive the car in the above mentioned conditions, the short term fuel trim will “switch” to a constant 1.6% with a check engine light (code 43)
To provide some more supplemental info about possible causes here are some things that I have check and ruled out so far: Map, coolant temp, all sensors in distributor, and vacuum in the intake manifold have all been tested according to the Haynes manual and have tested good. I have checked for vacuum leaks multiple ways (propane, carb spray, and just replacing most of the vacuum lines), but I don’t believe there could be a vacuum leak because of the fuel trims. When at idle the short term fuel trim is lower, and then when I accelerate the short term fuel trim increased rapidly. If there was a vacuum leak I believe that the short term fuel trim would decrease when accelerating because any minor vacuum leak would be unsubstantial.
Anyway, I had the O2 sensor relocated into the y-pipe to try to eliminate the possibility of an exhaust leak, but that has not changed anything. Does this sound like an exhaust leak? I torqued the header with a Honda gasket in a crisscross pattern to 23 lb ft and the Y-Pipe with Honda gaskets to 40 lb ft (according to the Haynes manual), and those are the only two possible exhaust leak origins (except if the header or Y-pipe themselves have a crack, but I can’t find any sign of that).
Fuel pressure is @43-44 PSI at idle, @46-47 part throttle, and @50-51 at full throttle. Those pressures should check out according to the Haynes manual, and I am running stock 91′ DA Integra fuel pump, would that cause any issues with fuel delivery? I am asking because the code 43 from the P72 ECU says “Fuel supply”. Also the code 43 is the only CEL that I have gotten from this swap that has not been solved.
Please let me know if you think you have a suspicion of what this issue may be pointing to, I would greatly appreciate it.
You must be logged in to reply to this topic.