Hi All
I was hoping that I could get some guidance on a project that I have.
I'm working on a engine controller that is a piggyback to a standard OEM ecu.
Not a ignition/fuel controller by the sense of the term, but in the end I hope to reduce gas mileage by means of cylinder deactivation. This is what I have planned.
Physical side
1) 1995 Nissan Maxima, 3.0L V6, OBD-I
2) Replaced the lower manifold with California specification emissions. This manifold has a set of butterflies (swirl valves) that reduce the size of the intake tract to each cylinder. I have modified this with custom CNC full-faced butterflies to replace the OEM butterflies. This will now seal the cylinders off from incoming air. The butterflies are activated by a solenoid.
3) Since this is a V6, the exhaust system is segregated by cylinder banks. A exhaust valve will be placed between the manifold and "y-pipe" on the deactivated side to seal off this end.
From what I have read, OEMs seal off the cylinders through the valve system. These systems usually consist of hydraulic lifters or shafts that disengage the valves to prevent them from opening. Of course, I'm not able to manufacture anything like this with out excessive amounts of money. Sealing the intake valves is particularly necessary as it seems most mileage increases are due to decreased pumping loses. Most OEMs seem to claim that by also closing the exhaust the remaining gases in the cylinder will become a spring and negate the power loss by an "empty" powerstroke.
Electronically, in theory, I think this needs to be done.
1) Activate the butterfly solenoid (12V)
2) Activate the electronic exhaust valve, also 12V but I think it needs a bit of amps. I do not have the specs yet.
3) Shut down the fuel injectors in every other cylinder in the firing order (my case is 2,4,6)
3) Avoid Check Engine Lights
4) Excessive fuel adjustments by the car's ECU
5) Make sure the ECU stays in closed loop operation
***By disabling 2,4,6 a power event on this engine will be every 120 degrees of crankshaft rotation. I decided not to disable 1,3,5 as the timing sensors on the engine use cylinder 1 as a reference. To be safe.
Having that said, I think I need access to the following engine sensors.
1) LH, RH O2 sensors (RH cylinder bank will be disabled) 0-1V (linear)
2) Throttle position sensor 0-5V (linear)
3) 3 fuel injectors 0-12V (oscillating?)
4) Mass Air Flow (?) 0-5V (linear)
With cylinders 2,4,6 deactivated (RH bank), the RH O2 should theoretically go immediately lean. Which the ECU would respond by adding more fuel. This would over fuel cylinders 1,3,5. So I think that I would need to intercept the RH O2 sensor when the system is activated and replace the signal with the LH O2 signal. Hopefully this will keep the ECU from adding too much fuel. In addition since the remaining bank is running and should be receiving more air, it may run lean. In which case I am counting on the ECU to richen the mixture.
The RH O2 indicating lean is something I observed by people that tried to just disconnect fuel injectors to receive fuel savings. All indicated worse gas mileage.
I would use the throttle position sensor to disengage the system if the throttle was wide open (WOT). So when the TPS is reading 5V to have the system disengage and resume normal engine operation until reactivated.
I am debating whether or not that I need the Mass Air Flow sensor's involvement. If the OEM ECU adjusts the mixture automatically for the increased air in the active cylinders I would not need this. I would have to experiment first without it to find out.
There we go....what i have thought of so far.
What I need help with is actually building a electronic controller to operate the system. I have very basic electronics knowledge, at least nothing this complex. My idea of how to start is the following.
1) Hard switch for main control
2) Remote Soft switch for activation/deactivation
3) Switching between O2 signals for the RH bank.
4) Switching the fuel injectors off
5) deactivate the system based on 5V input from the TPS
My thought was to use PNP transistors but I have no experience with them. I did not know what I could to activate them (keep them saturated) next I didnt know if the current was low enough (however I figured there is no relays in the ECU itself so it should be okay). Also how to use a soft switch...
More than anything I dont know where to get started. I'm sure Digikey has what I need, but I dont know the best parts either (automotive grade?). Any recommendations?
If anyone can give me some tips or point me in the right direction, that would be great.
I was hoping that I could get some guidance on a project that I have.
I'm working on a engine controller that is a piggyback to a standard OEM ecu.
Not a ignition/fuel controller by the sense of the term, but in the end I hope to reduce gas mileage by means of cylinder deactivation. This is what I have planned.
Physical side
1) 1995 Nissan Maxima, 3.0L V6, OBD-I
2) Replaced the lower manifold with California specification emissions. This manifold has a set of butterflies (swirl valves) that reduce the size of the intake tract to each cylinder. I have modified this with custom CNC full-faced butterflies to replace the OEM butterflies. This will now seal the cylinders off from incoming air. The butterflies are activated by a solenoid.
3) Since this is a V6, the exhaust system is segregated by cylinder banks. A exhaust valve will be placed between the manifold and "y-pipe" on the deactivated side to seal off this end.
From what I have read, OEMs seal off the cylinders through the valve system. These systems usually consist of hydraulic lifters or shafts that disengage the valves to prevent them from opening. Of course, I'm not able to manufacture anything like this with out excessive amounts of money. Sealing the intake valves is particularly necessary as it seems most mileage increases are due to decreased pumping loses. Most OEMs seem to claim that by also closing the exhaust the remaining gases in the cylinder will become a spring and negate the power loss by an "empty" powerstroke.
Electronically, in theory, I think this needs to be done.
1) Activate the butterfly solenoid (12V)
2) Activate the electronic exhaust valve, also 12V but I think it needs a bit of amps. I do not have the specs yet.
3) Shut down the fuel injectors in every other cylinder in the firing order (my case is 2,4,6)
3) Avoid Check Engine Lights
4) Excessive fuel adjustments by the car's ECU
5) Make sure the ECU stays in closed loop operation
***By disabling 2,4,6 a power event on this engine will be every 120 degrees of crankshaft rotation. I decided not to disable 1,3,5 as the timing sensors on the engine use cylinder 1 as a reference. To be safe.
Having that said, I think I need access to the following engine sensors.
1) LH, RH O2 sensors (RH cylinder bank will be disabled) 0-1V (linear)
2) Throttle position sensor 0-5V (linear)
3) 3 fuel injectors 0-12V (oscillating?)
4) Mass Air Flow (?) 0-5V (linear)
With cylinders 2,4,6 deactivated (RH bank), the RH O2 should theoretically go immediately lean. Which the ECU would respond by adding more fuel. This would over fuel cylinders 1,3,5. So I think that I would need to intercept the RH O2 sensor when the system is activated and replace the signal with the LH O2 signal. Hopefully this will keep the ECU from adding too much fuel. In addition since the remaining bank is running and should be receiving more air, it may run lean. In which case I am counting on the ECU to richen the mixture.
The RH O2 indicating lean is something I observed by people that tried to just disconnect fuel injectors to receive fuel savings. All indicated worse gas mileage.
I would use the throttle position sensor to disengage the system if the throttle was wide open (WOT). So when the TPS is reading 5V to have the system disengage and resume normal engine operation until reactivated.
I am debating whether or not that I need the Mass Air Flow sensor's involvement. If the OEM ECU adjusts the mixture automatically for the increased air in the active cylinders I would not need this. I would have to experiment first without it to find out.
There we go....what i have thought of so far.
What I need help with is actually building a electronic controller to operate the system. I have very basic electronics knowledge, at least nothing this complex. My idea of how to start is the following.
1) Hard switch for main control
2) Remote Soft switch for activation/deactivation
3) Switching between O2 signals for the RH bank.
4) Switching the fuel injectors off
5) deactivate the system based on 5V input from the TPS
My thought was to use PNP transistors but I have no experience with them. I did not know what I could to activate them (keep them saturated) next I didnt know if the current was low enough (however I figured there is no relays in the ECU itself so it should be okay). Also how to use a soft switch...
More than anything I dont know where to get started. I'm sure Digikey has what I need, but I dont know the best parts either (automotive grade?). Any recommendations?
If anyone can give me some tips or point me in the right direction, that would be great.