Hello,
I'm a physics student at Michigan State and I started an applied physics club. One of our first major projects is converting a 4-stroke combustion engine into a pneumatic engine to drive a motorcycle. We completed the prototype using a 4.5 hp tecumseh 139cc motor, a 5lb CO2 tank, some regulators, and a custom one-way valve opened by the piston at its peak. We're now working on a 1984 Yamaha Virago 1000 cc motor (2 cylinder, 8.3:1 compression ratio) and need some guidance!
Basic Plan:
We plan to inject pressurized CO2 twice per cycle, once at a low pressure (15-100 psi) during the intake phase and again at a high pressure (~800 psi) in place of the ignition phase. This way we don't have to modify the cam shaft to convert to a 2-stroke motor; the motor will require very little modification. We estimate the maximum chamber pressure of the combustion motor during ignition to be around 1000 psi so we'll try to emulate that with CO2 and expect to sit around a maximum of 800 psi when the piston is at the peak just before the engine would usually ignite the fuel. We will put together an optical tachometer and observe the flywheel to time the injections. We hope to regulate pressure and/or flow in response to the throttle being turned.
Problems:
1) Flow or pressure regulation: We expect the 1000 cc motor at 2000 RPMs to require a very large flow rate. The up side is we won't require regulators with especially quick response times or high precision. We really just need a regulator that has a range between 0-900 psi that can be controlled electronically.
[ul]
[li]From the flow estimates we can decide on which type of regulator would make the most sense in our case[/li]
[li]We need to source the regulator once we decide on the requirements[/li]
[/ul]
2) Gas phase transitions: This is mostly just something to consider but as it turns out that at around room temperature gaseous CO2 has a phase transition to liquid around 880 psi. While we won't be operating at 880 psi we expect the cooling of the gas as it comes out of the tank (stored as liquid) to lower the transition temperature and result in a gas/liquid mixture in our lines. The higher density may help reduce the required diameter of our fuel lines given a required mass flow rate but this is hard to model. I'm open to suggestions on how to think about this and what problems/benefits we might expect. My intuition tells me that if we inject liquid CO2 it will expand into a gas as the piston lowers causing rapid cooling; is this a problem?
3) High frequency on/off valve: If we operate at 2000 RPM we need to inject high-pressure gas 33 times/second which may be challenging. Are there high frequency, electronically controlled on/off valves that allow flows listed below?
Calculations:
We're mostly concerned with filling the volume of the chamber when the piston is at its peak with 800 psi CO2 at around 2000 rpm. We expect that we'll have to settle for lower performance but that is the ideal case.
Using bernoulli's equation I can estimate the flow velocity given a pressure differential of 785 psi and plugging in different densities. This calculator tells me the density of CO2 considering the pressure and temperature so I plug in different temperatures to get different velocities. Once I have the velocity I use this calculator to estimate the required pipe diameter. At 2000 RPM we need to fill 1/8.3 liters 33 times per second or a flow rate of ~4 L/s.
At 70 F -> 264 m/s -> 0.17" diameter (mostly gas)
At 60 F -> 115 m/s -> 0.26" diameter (mostly liquid)
Lower temperatures don't change much given that the density won't change much (same diameter for 40 F, for example)
We're open to any suggestions or questions you have. Thanks for reading!
I'm a physics student at Michigan State and I started an applied physics club. One of our first major projects is converting a 4-stroke combustion engine into a pneumatic engine to drive a motorcycle. We completed the prototype using a 4.5 hp tecumseh 139cc motor, a 5lb CO2 tank, some regulators, and a custom one-way valve opened by the piston at its peak. We're now working on a 1984 Yamaha Virago 1000 cc motor (2 cylinder, 8.3:1 compression ratio) and need some guidance!
Basic Plan:
We plan to inject pressurized CO2 twice per cycle, once at a low pressure (15-100 psi) during the intake phase and again at a high pressure (~800 psi) in place of the ignition phase. This way we don't have to modify the cam shaft to convert to a 2-stroke motor; the motor will require very little modification. We estimate the maximum chamber pressure of the combustion motor during ignition to be around 1000 psi so we'll try to emulate that with CO2 and expect to sit around a maximum of 800 psi when the piston is at the peak just before the engine would usually ignite the fuel. We will put together an optical tachometer and observe the flywheel to time the injections. We hope to regulate pressure and/or flow in response to the throttle being turned.
Problems:
1) Flow or pressure regulation: We expect the 1000 cc motor at 2000 RPMs to require a very large flow rate. The up side is we won't require regulators with especially quick response times or high precision. We really just need a regulator that has a range between 0-900 psi that can be controlled electronically.
[ul]
[li]From the flow estimates we can decide on which type of regulator would make the most sense in our case[/li]
[li]We need to source the regulator once we decide on the requirements[/li]
[/ul]
2) Gas phase transitions: This is mostly just something to consider but as it turns out that at around room temperature gaseous CO2 has a phase transition to liquid around 880 psi. While we won't be operating at 880 psi we expect the cooling of the gas as it comes out of the tank (stored as liquid) to lower the transition temperature and result in a gas/liquid mixture in our lines. The higher density may help reduce the required diameter of our fuel lines given a required mass flow rate but this is hard to model. I'm open to suggestions on how to think about this and what problems/benefits we might expect. My intuition tells me that if we inject liquid CO2 it will expand into a gas as the piston lowers causing rapid cooling; is this a problem?
3) High frequency on/off valve: If we operate at 2000 RPM we need to inject high-pressure gas 33 times/second which may be challenging. Are there high frequency, electronically controlled on/off valves that allow flows listed below?
Calculations:
We're mostly concerned with filling the volume of the chamber when the piston is at its peak with 800 psi CO2 at around 2000 rpm. We expect that we'll have to settle for lower performance but that is the ideal case.
Using bernoulli's equation I can estimate the flow velocity given a pressure differential of 785 psi and plugging in different densities. This calculator tells me the density of CO2 considering the pressure and temperature so I plug in different temperatures to get different velocities. Once I have the velocity I use this calculator to estimate the required pipe diameter. At 2000 RPM we need to fill 1/8.3 liters 33 times per second or a flow rate of ~4 L/s.
At 70 F -> 264 m/s -> 0.17" diameter (mostly gas)
At 60 F -> 115 m/s -> 0.26" diameter (mostly liquid)
Lower temperatures don't change much given that the density won't change much (same diameter for 40 F, for example)
We're open to any suggestions or questions you have. Thanks for reading!
