"David" wrote in message
...

It seems to me that this explanation, though common, is oversimplified.
What we are talking about is stability. When an aircraft (or a car for
that matter) turns on the ground a sideways force is developed on the
wheels. There is also a force generated by the sideways acceleration at
the cg and, also, in the case of the aircraft, a side aerodynamic force.
If the combined reaction of the first two is behind the cg than the yaw
motion on the ground will be unstable and may or may not be able to be
controlled by the pilot with the rudder.

It is not that the main wheels are in front of the cg so much as that
the side force developed by the tail wheel is very small (even smaller
if it is a castoring tail wheel) and too much of the side ground force
will be developed by the 'main' wheels.

I'm not sure that's the case, David.

The problem with a decelerating taildragger is that the inertial force at
the CG associated with the deceleration occurs behind the retarding force of
the mainwheels on the runway. This is unstable. If the two forces get out
of line, the couple tends to increase. By contrast, having the retarding
force on the main wheels behind the CG is stable.

^
| direction of motion in landing roll.

-------------------------------------------------

Tricycle:

^
|
* CG

+ wheels
|
v

-------------------------------------------------

Conventional (tailwheel):

+ wheels
|
v

^
|
* CG

-------------------------------------------------

Look at the stability to a yaw.

Of course the lateral forces you mention are usually available to control
the situation, which is why the instability isn't manifested as a ground
loop every time. If the aircraft starts to yaw it just gets more draggy,
hence the situation just gets worse.

In principle a tricycle gear aircraft should be susceptible to groundlooping
on take-off, but you have the stabilising advantage of increased drag
hampering acceleration as the aircraft starts to get out of shape.

Julian Scarfe