Sunday, May 27, 2012

Two New Steam Engine Designs

You may know I am fascinated with all manner of steam machines, from pop-pop boats to locomotives and rockets.  I've had a number of inventions in this area, but today I'm going to write about two designs for a boilerless steam engine.  

This all began when, in a moment of weakness, I came into possession of a seventeen-foot dory powered by a steam engine.  The boat needed work and I also wanted to convert the boiler from kerosene to gas in order to make it a little less messy and a bit easier to fire up.  As I began to rework the firebox and fix other things around the boat, it began to bother me having a box full of high pressure steam in the boat with me.  This became a meditation and led to my first design for a boilerless steam engine.

Design #1: The Boilerless Steam Engine
One area of technology that was missing in the golden age of steam was fast-acting solenoid valves and electronic controls that are used to provide electronic fuel injection in modern gasoline and diesel engines.  The idea is simple:  a fuel pump delivers fuel to the injector valve under pressure, and a computer commands the solenoid inside the injector to open a valve and spritz a certain amount of fuel into the airstream entering the engine.  With today's technology the amount of fuel delivered can be carefully controlled even at speeds to match the an engine at 6000 rpm (that's 10 milliseconds pre revolution).

So my idea was is:  instead of heating up a whole lot of water in a boiler and storing lots of dangerous steam, heat up a block of metal in the cylinder head, and then use a fuel injector to spritz water into that hot metal on the way to the cylinder.  Only two controls are needed:  a coarse one to keep the cylinder head at a heat of 300-500 degrees  C (by throttling a gas burner), and a fine one to fire the injector at the right time and for the right duration.

[schematic image of a boiler and steam engine yet to be supplied]

[schematic image of Boilerless Steam Engine yet to be supplied]

While I've never actually built one of these engines, there can be no doubt that it would work.  However, I think I've convinced myself that it wouldn't work very well.  The problem is being able to transfer enough heat to turn a pea-sized drop of water completely to steam in, say, 25 milliseconds (corresponding to 1/4 revolution of an engine at 600 rpm)  I don't know that this won't work, but I've seen hot water dance on a griddle, and I think that sort of surface flash effect would keep the water droplets from making good contact unless you forced them through very small holes, and then that might well dissipate much of the energy that was supposed to do work on the piston.  I'm sure it could work, but I just don't think it would work well.  It would be fairly easy to test the idea by simply experimenting with injecting water drops into a channel in a ver hot piece of metal.  But now I have an even better idea...

A little applied physics
We pause here for a lesson on superheated water and the critical point.  You know that if you put a pot of water on the stove and heat it up, it starts to boil when it gets to 100 degrees C.  The more heat you supply, the more water boils, but it never gets hotter than 100 degrees.  In a pressure cooker, the container holds a pressure at 15psi above atmospheric pressure, so the water gets up to 120 degrees or so before it boils, and a pressure relief valve prevents the pressure from getting any higher for safety reasons.

But what if you have a really strong tank, and *don't* release the pressure, but just keep heating the water up?  Well it just gets hotter and hotter and the pressure goes up and up.  If you let some of this water out, you will find that it boils when it is released, because it already has enough extra heat content to boil it [this is what happens when you take the radiator cap off an overheated automobile engine].  Time out for two useful facts:

    It takes 1 calorie of heat to raise 1cc of water by one degree C
  It takes 540 calories (!) to turn 1cc of 100 degree water into steam
  [bonus fact] that same 1 cc of water becomes 1675 cc(!) of steam

Now if you keep heating the water up to 374 degrees C (it will be at a pressure of 3200 psi) you get to what is called the critical point.  At this temperature if you release that water into the air it will spontaneously turn completely into steam.  Recalling that this involves an expansion by a factor of over 1600, you can see that this will be an explosive change.

Steam rockets (off topic, but inspriational...)
If you made it this far, you will probably enjoy seeing what some people have done by just putting a nozzle on a tank of superheated water to build a rocket...

Here are some whacky back yard experiments:
and here is some serious rocketry:

While it may seem like an uncontrollable thing, the steam rocket is actually remarkable for its controllability.  It is as simple as a solid fuel rocket to start, and yet it can be throttled down or stopped at any moment, unlike solid fuels.  This is why Evel Knevel used it for his Snake River crossing attempt (it failed due to premature parachute deployment).

Design #2:  The Superhot Water Engine
If you think like me, you are probably anticipating the second design.  Here, instead of injecting cold water into a very hot cylinder head, hoping that it will flash to steam, we inject super heated water into a normal cylinder where it will flash spontaneously from its internal heat content.  It may sound scary to be dealing with water that is hotter than molten lead at a pressure of 3000 psi but, there will be no large reservoir of this water.  Here is the the setup:

The pressure is generated by a water pump (no tank involved), the heating takes place in a relatively small pipe (again, no large accumulation of hot water), the pipe goes straight to an injector valve into the engine's cylinder.

[schematic image of Superhot Water Engine yet to be supplied]

Baby steps
In order to conduct a few experiments, we need a pump, a heater, and an injector.  We can do the first experiment just shooting steam out of the injector into the air.  Then as a next step, we can shoot it into the spark plug hole of a leaf blower motor (I happen to have several leaf blowers left from an earlier experiment ;-). 

Parts list
Pump:  AR383 pressure washer - 1900 psi at up to 1.5gpm
    available for $165 from Wallmart
Pipe: 1/4-inch copper tubing - small diameter gives both high rating
    and good thermal transfer, and minimizes the amount of stored fluid.
Thermometer: Pyle PIRT30 high temperature IR thermometer
    50 tp +550 degrees C, available for $70 from Sears
Heater:  Bernzomatic TS4000T trigger-start torch
    available for $35 from Walmart
Injector valve:  ?

The injector valve is the one missing ingredient, and affordable pumps fall short of the desired 3200 psi.  We should be able to observe plenty of the spontaneous injection effect at 2000 psi, though, with the only downside being that we will produce somewhat wet steam and somewhat less power from the engine.

Collaboration?
I'm eager to start experiments and build one of these engines.  However because of a heavy work schedule and family obligations, I won't get much done in the next few months.  If you would like to work with me, or even try some of this on your own, please get in touch with me and share the fun!

11 comments:

  1. Perhaps you already know, but the surface flash effect on the griddle you mentioned is called the "Leidenfrost effect".

    ReplyDelete
  2. No I didn't know that and thanks. I've had a lot of fun observing it.

    BTW: Several people have suggested diesel engine injectors, and I plan to try them, but I can find no "physicist's guide" to injectors - i.e. info on current, response time, operating pressure etc, instead of just what engine they work with. Any suggestions out there?

    ReplyDelete
  3. Apparently my Design #1 is one in a series of reinventions of the "hothead" concept first proposed by R Seaton Taylor in 1911. It has been made to work, although never very effectively, I think for the reasons that I mention.

    Still searching for any record of a superhot water engine.

    ReplyDelete
  4. Replies
    1. This link is a commercial steam engine called the cyclone. Its similar to your idea. Any progress?

      Delete
  5. I am doing Steam Rockets
    San Jose
    Can we discuss on e-mail ?
    rocketmanlewis@yahoo.com

    ReplyDelete
  6. This comment has been removed by a blog administrator.

    ReplyDelete
  7. you make a reference to a "One area of technology that was missing in the golden age of steam was fast-acting solenoid valves and electronic controls that are used to provide electronic fuel injection in modern gasoline and diesel engines" are you aware Flashsteam.com has made such a valve. The valve is also variable lift to provide efficient throttling. The valve can be fitted to any engine and installs in a spark plug like port.

    ReplyDelete
  8. "Design #2: The Superhot Water Engine

    If you think like me, you are probably anticipating the second design. Here, instead of injecting cold water into a very hot cylinder head, hoping that it will flash to steam, we inject super heated water into a normal cylinder where it will flash spontaneously from its internal heat content. It may sound scary to be dealing with water that is hotter than molten lead at a pressure of 3000 psi but, there will be no large reservoir of this water. Here is the the setup:" check these two links

    http://flashsteam.com/forums

    http://flashsteam.com/steam-engine-project-part-3

    ReplyDelete
  9. About 9 years ago I had an idea to use a diesel engine and intake hot air then injecting water instead of diesel fuel thus flashing the injected water into steam. I obtained a small, approx. 10hp rated Kabota 2 cyl from a truck refrigeration unit that I tried it on. I was unable to get the intake air hot enough to make it run but I can't think of any reason it wouldn't work if engine and intake air were sufficiently heated. My goal was to create a small prime mover that could be used to turn smaller sources of industrial waste heat to shaft power as there aren't any cost effective means now that I'm aware of. I used a small amount of water soluble machine tool coolant-lubricant in the feed water to avoid damaging the fuel pump. No modifications were done to the engine. I ended up selling the engine to a guy who was going to make it into a stationary power unit for an Amish farmer.

    -Randal Recker

    ReplyDelete
  10. Biggest issue with this right now is one of thermodynamics. You can heat, for example, the engine head and inject water onto it, but most of the water will bounce off and go into the cylinder. What little water flashes into steam is going to condense on the cylinder walls and the top of the piston. To really make this idea work, you'd have to heat the entire engine. No real efficiency advantage there.

    How 'bout a flash steam boiler instead? You could use relatively small amounts of water and simply run the Rankine cycle engine as usual. As it is a boiler, you do not have to use distilled water (you would with a heated engine) and injecting steam oil is not an issue.

    I'm thinking that electronically controlled valves/solenoids that inject steam instead of fuel could well boost efficiency, although compounding probably works as well.

    ReplyDelete