Pathfinder R50 vg33e after transmission swap

[onurgocmen]

Hello all,

3 years ago I bought a pathfinder r50 3.3 converted from automatic to manual. I've been working on it for a long time. It has the old auto engine ecu and wiring harness. The automatic transmission computer(tcm) was removed/disappeared during the swap. In the wiring harness, only the positive and negative poles of the park neutral switch are connected, the vehicle can start in any gear. (For my problem, which I will talk about later, I tried to disconnect these poles after the vehicle started, but it did not solve the problem.)

I have no problems with my car other than minor hesitations that started later. I used it for a long time without these hesitations, my car was good. I'm investigating to resolve the hesitation issue.

I connected 2 separate vehicles to the computer as in the photo. 1 of them is my vehicle that works with a manual transmission automatic engine control unit and automatic ecu cable set. The other is a vehicle with an automatic transmission and an automatic brain, just as it left the factory. In my converted car, the iacv valve stays open 20% when idling. But as soon as I touch the gas pedal, it opens 100% regardless of the rpm. In the stock vehicle, on the other hand, it is 20% open at idle, but it opens gradually at 30-50-80-100% according to the throttle response. For my car this creates a serious vacuum leak. And since it is 100% open during the ride, it warms the computer and the valve. I've looked through all the factory service books over the years. I changed the tps sensor, I made the tps adjustments, I changed the iacv valve, I installed my friend's auto ecu, but this rpm jump did not fix it. It is very difficult to find manual ecu and harness in my area. Any ideas how I can run IACV by tps position? I'm hoping it can be done with a few jumps or resistors on the harness but frankly I don't have a deep knowledge of electronics. I think this situation occurs because tcm is disabled. I looked at the tcm and ecu pinouts in the service book, but I couldn't find any explanation such as tcm controls the iacv.

[mdmellott]

So ... you have a '98 Pathfinder converted from having an automatic transmission to having a manual transmission without all the right parts for a seamless integration and you have this Pathfinder able run on LPG or gasoline with separate fuel injectors for each fuel type and you installed an EGR delete plate. The multitude of possible complexities and issues is mindboggling and I applaud you undertaking such a daunting project. The difference between Frankenstein's monster and Pinocchio is that Dr. Frankenstein was a mad scientist that create an incomplete thing that was eventually destroyed but Mr. Geppetto carved a marionette out of enchanted wood that became the misguided, mischievous Pinocchio until he learned to do right and became a real boy. I was going to make a point here but I lost myself.

[macgiver]

Very interesting & astute theory of TPMS controlling IACV .
Now many TPS have a switch for "fully closed " , then resistor POT giving 1-100% opening (we'll forgo here) , then you have a switch @ WOT - these 'switches' may be NO , NC , OR combo NC + NO .
You can completely separate / segregate the TPMS circuitry , probably by small ctrl relays that have NC/NO combo poles AND run the IACV OFF these relays with a resistor divider for Idle setting ( volt parameter) , and another resistor divider for off-idle running .
Say you keep the 20% for idle but through experimentation you have a ' working ' IACV voltage to sustain your off-idle driving .
I dunno ?.........Capeeche ?

p.s. If PWM (pulse width modulated ckt) on the IACV - all bets off , I think 1998 your well before that era ??
I dont even know IACV ever went PWM , but just sayi'n ...

[macgiver]

Could also be a "threshhold-trigger " ckt working directly off TPS resistor-divider , to achieve idle vs off-idle voltage parameters . And it shall have high enough impedance looking towards the TPS circutry so as not to overload nor disturb TPS signal to ECM .
There are " basic canned schematics " on the web , but you gotta have sufficient knowlegde ( like a VStar here) to implement properly . I just try to give a good working theory to look into .

[onurgocmen]

Dear Macgiver, thank you very much for your valuable suggestions. When I examine the TPS circuit the top socket actually seems to do this task. I tried unplugging it completely. But the ecu could still see the open/closed position from wherever it was getting data, so I plugged it back in. There is no data on my computer about wot. The service book says wot is only used for A/T control. Judging by the diagram, the system enters wot as soon as it somehow comes out of the off position. I can cut a wire instead of removing the whole socket And I'll consider your other suggestions, thanks again.

And dear mdmellot, when I bought it I did not know about this conversion because the assemblies were done very professionally. Unfortunately, people in my region have to make these transformations because they can't find parts,dont doit for pleasure. I know dozens of similarly converted vehicles around me. All of them have this 100% opening issue and they will burn the ecu eventually. But in their swapped vehicles has no hesitation.Just like in my car before.. Although I cannot fix my hesitation problem, I am trying to prevent future ecu problems of all my friends. And I love my vehicle. Thank you for your information and all your opinions.You're right about the mess, but that's the car and conditions in front of me.

[macgiver]

Right , I was going to add the end result oughtta have the ECM to IACV disconnected , AND maybe not left "open " , but put a suitable load resistor slightly higher value maybe than the resistance of the IACV - just to satisfy the transistors load circuitry , ya know . Right , many an ECM x-sistor outputting to some IACV 's were blown under certain conditions , I dont pretend to know those details tho .

[VStar650CL]

The TPS sensor isn't the problem, translating it into an IACV position is. The IACV-AAC on your ride is a PWM-able solenoid controlled by a signal from the ECM, so to convert it to a TPS input you'll need a device to convert voltage from the TPS into PWM for the IACV. Then you'll also need a resistor with the same DCR as the IACV coil, to fool the ECM into believing the solenoid is still there and keep it from throwing a P0505. That part is straightforward, but the only device I can think of that will convert voltage into PWM for you is one of these:

https://www.ebay.com/itm/143561612623

It's meant for controlling brushless fans, but you can "teach" it the voltage values from a TPS just like teaching it the ramp of a temperature sensor. Since it doesn't have any meaningful current drive, you'll also need a small Solid State Relay (SSR) to actually drive the solenoid coil. I just looked at the "minimum on" signal output for the controller and it's 15% at a 100 Hz PWM rate, 90% at maximum. So it will probably do what you want, although it may need a bunch of experimenting and fiddling to get it exactly where you want.

[onurgocmen]

Thank you all. I'm trying to understand why the ECM only applies the idle/wot mode for the IACV since only the tcm is disabled. Because it doesn't seem to have any connection with wot. Even if TPS is turned on at 5-10% on my odb device, ECM immediately runs iacv to wot. But when I examine the fuel trims, it seems to take the tps clearance into account. Because I do not encounter any extreme rich-poor situation. Frankly, I was thinking that I could do it with a few cable jumps or by completing a missing grounding, but it won't.

I understand the logic of the device you are talking about very well, thank you. This setup may be beyond me for now, but it's a good start. If I can even command 20%-50%-90% with a very stupid setup, I'm sure it will be much better than its current situation.

[onurgocmen]

what i understood from the last drawing is that the 'wot' circuit is no longer grounded by the tcm so it may be sending too much voltage to the ecm via the 'other' path? but in this case, when I removed the upper socket, it should have been fixed.

[VStar650CL]

Only the "closed" position on that switch would affect anything. The WOT section only closes at WOT and the signal only goes to the TCM, so its ground is irrelevant. Most TCM switch signals are set up opposite from other controllers. Generally switch-detection circuits are pulled up by the device and forced to ground by the switch, but TCM's are usually pulled down by the TCM and forced to 12V by the switch. That's the case with both of those switches, so your problem may be from lack of a pull-down bias on the OR/W wire. Without a pull-down source, the ECM may see the switch as closed even when it's in the "other" position. Try putting a 10K 1/4W resistor to ground from OR/W and see if it helps.