I note (from the patent) that Cornish runs a 500 cc
  motorcycle engine on his hydrogen.
  Normal consumption at full trottle would be 250 cc per
  revolution.  Actual consumption is 1 liter per minute.
  Now if the revolutions are 1000 RPM this is 1cc per rev.
  That is 1/250th of normal full trottle Consumption or
  1/25th of 1/10th throttle normal consumption.
  Lets see how this compares to Peter Lowrie's consumption.
  Peter is producing About 5 liters per minute.
  His Honda engine is 1600 cc.
  This is 800 cc per rev.
  Consumption at 1000 rpm should be 800,000 cc or 800 liters per
  minute.
  At 1/10 throttle this would be 80 liters per minute
  He is producing about 1/16 of this.
  The difference between the two is a ratio of 1.5625
  Cornish being about 2/3 of what Peter Lawrie is operating on.
  Allowing for the fact that Cornish is producing only
  hydrogen and using the oxygen in the air, the two are almost
  identical.

  How can they both be getting this efficiency?
  The e-gas fuel when burned is like a two edged sward. Both
  sides can be used. If the collaps to water after the shock
  explosion is also used, the power out would be approximately
  doubble.

  Consider this as to how it is possibly done:

  The cams can be changed to doubble lobe to make the
  engine two cycle. The timing can be set at a time "T" befor
  bottom dead center. The exhaust valve would be locked closed.
  Engines with waist spark are now desirable because the other
  spark will be used in the 2 cycle mode.
  The time "T" is the time after initial explosion that the
  implosion occures. This can be measured with a high speed
  camera on a baloon that has had a very small amount of gas
  locked in. Two wires in a  cork in the neck of the baloon
  can be fired with high voltage, with  the camera running.

  If time "T" is mechanically locked in on the rotation of
  the shaft the motor will have to be run at a constant
  speed. Very good for AC power generators as you require a
  fixed speed. Electronic timing would alow the speed to be
  varied. A small resivoire must be provided to catch the
  water.  This can be drained in betwen runs.

  The valves may have to be modified to ball valves or the
  spring at least be made stiff enough that they would not
  open under a complete vacuum,
  With this theory little or no water vapor is wanted in the fuel
  mixture.  No air is wanted either.

                            (C) 2006
The above is copywrited by Willard J. Elliott and may not be reproduced in
any form without his express written permision. March 6th 2006