Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Cylinder blowdown pressure?

  1. Aug 13, 2014 #1
    Hi, all.

    I'll be experimenting with water mist injection (amongst other things) on a small single cylinder engine. Because the electrical generation capability of the engine's alternator is limited, I have to find some other way of pressurizing the water for injection (not enough juice for an electric pump)... my idea is to capture the high pressure pulse just as the exhaust valve opens (cylinder blowdown) by tapping into the exhaust pipe just outside the engine proper.

    I'd trap the blowdown pressure with a check valve, route it to a pressurized water tank, and use that pressure to inject the water via a modified fuel injector in the intake manifold, controlled via a reprogrammed MicroSquirt that controls both the fuel injector and water injector.

    It is my hope that by leaning the fuel:air ratio then injecting just enough water to keep combustion temperatures below where NOx is created, the expanding steam will add to cylinder pressure, helping to offset power loss due to burning lean, while at the same time making for very clean emissions (lean burn has low HC and CO2, but high NOx due to burning hot... cool that down and you get low HC, CO2 and NOx). A side benefit is the water acts as an octane booster to prevent lean burn knock.

    The intake manifold will have a piezo that breaks the water and fuel into a fine mist, for more complete combustion and to force the water to flash to steam quicker. The injected fuel will be heated to 250 F, the intake air will be heated to approximately 150 F, and the injected water will be preheated to just below boiling point using exhaust heat, so we get the most expansion in-cylinder with as little heat input as possible (ie: it uses more water to cool combustion temperatures, giving as big a boost to cylinder pressure as possible).

    The engine is 174.5 cc.
    Compression ratio is 11.2:1.
    Compression pressure is 227.2 psi.
    Exhaust valve opens at 28 degrees BBDC.
    Exhaust valve closes at 8 degrees ATDC.
    Maximum RPM is 9200 (rev limiter)
    Redline RPM is 8000.
    The cylinder sits at approximately a 45 degree angle from vertical.
    Exhaust pipe is 1.2" in diameter.
    There is a 90 degree downward bend in the exhaust pipe about 2 inches after where it attaches to the engine, then a 90 degree bend to the right about 4 inches after that, then a third 90 degree bend at the right side of the bike before it goes to the muffler.

    Not sure if any of the above info helps.

    Is there any way to calculate what peak blowdown pressure would be present in the exhaust pipe just outside the engine proper?

    What shape and orientation of pipe (that will be grafted to the exhaust pipe) would be most beneficial (prior to the check valve) for capturing the pressure pulse? Directly in line with the exhaust port, at an angle? Expanding bell shape, straight pipe, reducing bell shape?

    Would spacing of the check valve from the exhaust port be a consideration (ie: is closer better or farther away)?

  2. jcsd
  3. Aug 17, 2014 #2
    The closer to the valve the better I suppose, and I don't think shapes will matter much, but I don't think there will be enough energy there unless the exhaust system is heavily restricted(more so than would be reasonable to do).

    I think you could scavenge more pressure/energy from the crankcase when the piston comes down, effectively compressing the gasses under the piston. In a single cylinder, this is especially profound, as there are no opposing pistons going up at the same time. A two stroke uses this energy to help force the charge into the cylinder.

    I still don't know about using a fuel injector though. You need a lot of pressure to be able to spray through an injector. You might need to figure out another way to regulate the water volume. If you are utilizing the crankcase pressure, it would vary with respect to rpm and load, so you should be able to get pretty accurate volume per cycle just by pushing the water directly into the intake tract. The vacuum will be pulling on it anyways, so you will be restricting it heavily at low throttle positions if injecting after the throttle. I would try and inject it right near the throttle, using a port that is only exposed to vacuum as it opens. You would be making a water carburetor of sorts. You wouldn't even need the pressure from the crankcase or exhaust for this.

    How is the piezo going to be used? Maybe it could be manipulated so as to also help regulate the volume?
    Last edited: Aug 17, 2014
  4. Aug 19, 2014 #3
    The piezo will sit in the intake manifold just downstream of the fuel and water injectors. Its job is to turn the fuel and water into microscopic droplets (a "fog") and keep them in suspension in the air. Smaller fuel droplets means faster and more thorough combustion. Smaller water droplets means a faster flashing to steam. Think along the lines of those cool misters you see for humidification, except the piezo is designed to keep running without melting down when it's hot and dry.
  5. Aug 19, 2014 #4


    User Avatar
    Homework Helper

    but then you'd lose the energy from the compressed air when the piston goes back up. Air is resonably elastic during compression and expansion, with most of the losses in the crankcase due to the path of the air flow and momentum changes.

    For drag racing vehicles, using an engine driven vacuum pump to reduce the amount of air in the crankcase increases the overall power output of an engine (note this is a full power, high rpm situation).

    As mentioned, trying to get pressure from the exhaust will restrict the exhaust flow. If that pressure was being used to drive a turbo charger, then overall power will increase. I'm not sure that the pressure from a turbo charger would be enough to pressurize the water enough for injectors, so doing this may require an engine driven pump (should be more efficient than a larger altenator driving an electrical pump).
  6. Aug 20, 2014 #5
    Yeah, I guess if it is a sealed crankcase, the pressure when the piston is down will give a slight boost pushing it back up, but I haven't really seen any sealed crankcases. Most engines, even small single cylinder ones have some kind of crankcase ventilation. Sealed crankcases tend to blow out seals and vent to wherever, defeating the idea of a pressurized case.

    Anyways, I don't think the lack of pressure from a sealed crankcase will hurt overall engine efficiency much, even if it was running that way now. The main idea was just to find a way to force some water in.

    I agree that if a fuel injector is used to regulate the water, a high pressure pump will be needed(at least 10psi, better with 30+). that's why I say ditch the injector and just spray it through a restriction directly into the manifold or upstream of it.
    You will be losing some energy in the pump, as well as to fire the injector.
  7. Aug 21, 2014 #6
    There are several objectives to precisely metering the amount of water being injected... preventing too high a cylinder pressure, preventing too much combustion heat from being absorbed (then thrown out the exhaust), keeping flame front speed high, and preventing NOx.

    The reason I wanted to use an injector to inject the water is that I can measure relative humidity of the incoming air versus the intake air temperature, and have the ECU adjust the amount injected to that precise amount needed.

    I was hoping that the high pressure blowdown pulse that occurs just as the exhaust valve opens would provide enough pressure (there wouldn't be an awful lot of flow diverted from the exhaust system... we're not injecting a lot of water) to keep the water tank pressurized enough that a pump wouldn't be necessary.

    I guess I'll just have to empirically obtain the data to see if it'll work.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook