Tweet
Hi, all. New to the forum.
I've got a 2010 Kymco Yager GT 200i scooter that I'll be experimenting with. It's a fuel-injected, liquid cooled 174.5 cc engine, a Honda GY6 derivative. The engine's been immaculately maintained, with the fuel mileage and all maintenance done to it tracked from Day 1.
I've ordered the programmer for the ECU, so I can change timing and lambda, etc. It'll be here around the 15th of April.
I'm thinking of going with the Brisk Racing BR14ZC spark plugs, unless a more knowledgeable forum member has a better suggestion. Right now I'm running an NGK Iridium IX DPR7EIX-9.
What I want to do is build an exhaust heat recovery system... it'll dump exhaust heat into the coolant via an air-to-air cross-flow heat exchanger. The air will be pulled through the heat exchanger into the intake of the engine.
In this way, I get quicker warmup on a cold start (by dumping exhaust heat into the coolant), a warm air intake that's regulated to a specific temperature (there'll be a bimetallic-actuated throttle body type valve that opens to allow cool outside air into the intake if the air's too hot), will allow me to preheat the gasoline to the same temperature as the coolant for better dispersion, and will allow me to preheat water.
I'm having built a microcontroller electric pump setup with two small mag-drive water pumps to replace the built-in coolant pump. It'll monitor coolant, head and exhaust temperatures and put both pumps and both radiator fans to 100% if any overheat is detected. With this, I can vary coolant temperature in 1 F increments and monitor temperatures as I lean the lambda.
I'll be buying an OEM throttle body and modifying it to hold a water injector, in addition to the fuel injector. After the coolant pump circuitry build is complete, I'll be having built a microcontroller that takes the PWM pulse width for the fuel injector from the ECU, and allows it to be varied to power the water injector, so I can vary the proportion of water injected as a proportion of the fuel injected. In this way, I don't have to determine the proper amount of water to inject at any given engine speed or load... it'll be proportional to the fuel being injected. The microcontroller will also have the ability to modify the PWM pulse width going to the fuel injector, so I can dynamically lean the fuel injected without messing with the ECU fuel map.
One thing I'm not sure about is how to pressurize the water for injection... because it's on a small scooter, electricity generation capability is limited, so a pump is probably out of the question. I'll be running a 30 watt LED H4 headlight in place of the 55/60 watt H4 bulb to save 30 watts. 25 watts of that will be consumed by the water pump setup (with both pumps running at full tilt... they're very small pumps, only 15 LPM each, but each still pumps more than the built-in pump. The pump microcontroller takes something like less than a quarter watt). So after all is said and done, I've only got about 5 extra watts to spare over stock. The voltage controller is a ground-shunt type, so I have to be very careful in balancing the power consumption, in order to prevent burning it out by shunting too much current to ground by consuming too little on the bike itself.
So, getting creative, I decided I'm going to try to capture exhaust pulses right off the exhaust manifold as it exits the engine proper, put them through a check valve to capture the pressure without it bleeding back, and pipe that into the water tank. Thus, no power required... but I'm not sure how much pressure will be created or if it will be enough for water injection.
Inside the customized throttle body, I'll mount a piezo that'll be powered by a simple variable-frequency oscillator, to force the fuel and water molecules to remain as small as possible prior to intake, rather than condense.
After all of that is done, I'll have the plasma spark circuitry built. I'm hoping that I can lean the lambda enough to get 150 MPG. I'm getting about 70 MPG now.
So... getting to the questions that required that long explanation...
1) How much power does the plasma spark require?
2) Can I upgrade my existing coil to something that gives a hotter spark, as I think I'd fry the poor little sucker that's in there if I try to experiment with it. Or will a coil that draws more power burn out my ECU? It appears that the coil is powered directly from the ECU. I've got an electrical diagram of the bike for anyone who's interested.
3) For the piezo in the throttle body to prevent fuel and water condensation, is there any research as to what frequency would be best? What power output?
4) Are two piezos required, each operating at a different frequency, in order to keep both fuel and water in a fine mist?
5) When water is injected and burned as fuel, does that vary the oxygen content of the exhaust stream, and if so, in what way? I'll need to know this prior to having the fuel and water injector PWM pulse width modulator built.
6) Is there anything I'm missing, or am I taking a wrong path on any of what I've planned?
Aaron, look for my order of your "Ignition Secrets" book soon.
Thanks, guys.
I've got a 2010 Kymco Yager GT 200i scooter that I'll be experimenting with. It's a fuel-injected, liquid cooled 174.5 cc engine, a Honda GY6 derivative. The engine's been immaculately maintained, with the fuel mileage and all maintenance done to it tracked from Day 1.
I've ordered the programmer for the ECU, so I can change timing and lambda, etc. It'll be here around the 15th of April.
I'm thinking of going with the Brisk Racing BR14ZC spark plugs, unless a more knowledgeable forum member has a better suggestion. Right now I'm running an NGK Iridium IX DPR7EIX-9.
What I want to do is build an exhaust heat recovery system... it'll dump exhaust heat into the coolant via an air-to-air cross-flow heat exchanger. The air will be pulled through the heat exchanger into the intake of the engine.
In this way, I get quicker warmup on a cold start (by dumping exhaust heat into the coolant), a warm air intake that's regulated to a specific temperature (there'll be a bimetallic-actuated throttle body type valve that opens to allow cool outside air into the intake if the air's too hot), will allow me to preheat the gasoline to the same temperature as the coolant for better dispersion, and will allow me to preheat water.
I'm having built a microcontroller electric pump setup with two small mag-drive water pumps to replace the built-in coolant pump. It'll monitor coolant, head and exhaust temperatures and put both pumps and both radiator fans to 100% if any overheat is detected. With this, I can vary coolant temperature in 1 F increments and monitor temperatures as I lean the lambda.
I'll be buying an OEM throttle body and modifying it to hold a water injector, in addition to the fuel injector. After the coolant pump circuitry build is complete, I'll be having built a microcontroller that takes the PWM pulse width for the fuel injector from the ECU, and allows it to be varied to power the water injector, so I can vary the proportion of water injected as a proportion of the fuel injected. In this way, I don't have to determine the proper amount of water to inject at any given engine speed or load... it'll be proportional to the fuel being injected. The microcontroller will also have the ability to modify the PWM pulse width going to the fuel injector, so I can dynamically lean the fuel injected without messing with the ECU fuel map.
One thing I'm not sure about is how to pressurize the water for injection... because it's on a small scooter, electricity generation capability is limited, so a pump is probably out of the question. I'll be running a 30 watt LED H4 headlight in place of the 55/60 watt H4 bulb to save 30 watts. 25 watts of that will be consumed by the water pump setup (with both pumps running at full tilt... they're very small pumps, only 15 LPM each, but each still pumps more than the built-in pump. The pump microcontroller takes something like less than a quarter watt). So after all is said and done, I've only got about 5 extra watts to spare over stock. The voltage controller is a ground-shunt type, so I have to be very careful in balancing the power consumption, in order to prevent burning it out by shunting too much current to ground by consuming too little on the bike itself.
So, getting creative, I decided I'm going to try to capture exhaust pulses right off the exhaust manifold as it exits the engine proper, put them through a check valve to capture the pressure without it bleeding back, and pipe that into the water tank. Thus, no power required... but I'm not sure how much pressure will be created or if it will be enough for water injection.
Inside the customized throttle body, I'll mount a piezo that'll be powered by a simple variable-frequency oscillator, to force the fuel and water molecules to remain as small as possible prior to intake, rather than condense.
After all of that is done, I'll have the plasma spark circuitry built. I'm hoping that I can lean the lambda enough to get 150 MPG. I'm getting about 70 MPG now.
So... getting to the questions that required that long explanation...
1) How much power does the plasma spark require?
2) Can I upgrade my existing coil to something that gives a hotter spark, as I think I'd fry the poor little sucker that's in there if I try to experiment with it. Or will a coil that draws more power burn out my ECU? It appears that the coil is powered directly from the ECU. I've got an electrical diagram of the bike for anyone who's interested.
3) For the piezo in the throttle body to prevent fuel and water condensation, is there any research as to what frequency would be best? What power output?
4) Are two piezos required, each operating at a different frequency, in order to keep both fuel and water in a fine mist?
5) When water is injected and burned as fuel, does that vary the oxygen content of the exhaust stream, and if so, in what way? I'll need to know this prior to having the fuel and water injector PWM pulse width modulator built.
6) Is there anything I'm missing, or am I taking a wrong path on any of what I've planned?
Aaron, look for my order of your "Ignition Secrets" book soon.
Thanks, guys.