For tight radius curves, it is better to use set track than flexitrack, as the setrack by design allows for a little bit of gauge widening on the curves and ensures that the radius of curvature is constant and good rail joints without gaps or sudden transitions in radii.
I suspect that the derailment problem may have something to do with the gauge narrowing that arises when you bend flexi-track through too tight a radius. Also with flexitrack you have to very careful to ensure that you do not introduce any slight kinks in the curve. Derailments occur when one of the wheel flanges finds a place where it can climb out from between the rails. This could be anywhere that there is a rail joint or a sudden change of rail height.
Perhaps this needs a bit more explanation. Consider a set of wheels standing on a piece of straight track. The flanges are not in tight contact with the inside edge of the rail and as they move they stay parallel to the inside rail edge. The coning on the flat part of the wheel tread (present on good wheel sets) ensures that gravity keeps the wheel in the centre of the track and it should not hunt from side to side. In theory there should be no derailments at all.
Now consider a pair of wheels travelling on a curved piece of track. There is a lateral force seeking to push the wheel set towards the outside of the curve. At reasonable speeds this is balanced to a greater or lesser extent by the gravitational effects of the tread coning, and a state of equilibrium exists. If there are no sudden increases in the radius of curvature, the flanges are kept away from the edge of the rail. With tighter curves and at greater speeds, the flange leading edge is forced against the edge of the rail more often and hunts back and forth..
Whenever the leading edge of the flange is forced against the inside edge of the rail and there is a rail joint present with enough roughness or gap to give the flange edge a purchase, the wheel may climb out of the four foot and you get a derailment. I mentioned that the wheels may hunt from side to side when travelling too fast. This explains why the derailment does not always occur in the same place. The laws of chance indicate that the right conditions for a derailment will only occur from time to time, all we can do is ensure that these can occur as infrequently as possible, by laying smooth long radius curves, with transition and ensuring that the back to back of the wheelsets, the squareness of our chassis, the freedom of movement of bogie pivots and so on are all at the optimum.
Using set track helps with sharper radii as the gauge widening helps to keep the flange leading edge off the edge of the rail.
Child Psychology: A child will soon lose interest if the running is unreliable. Yes, but they will also lose interest if everything runs smoothly all the time. They will want to run the locos flat out or connect all the stock up to the locos in a double header. Often the intention may be to create a derailment and get adults attention.
I hope this helps.
Colombo
I suspect that the derailment problem may have something to do with the gauge narrowing that arises when you bend flexi-track through too tight a radius. Also with flexitrack you have to very careful to ensure that you do not introduce any slight kinks in the curve. Derailments occur when one of the wheel flanges finds a place where it can climb out from between the rails. This could be anywhere that there is a rail joint or a sudden change of rail height.
Perhaps this needs a bit more explanation. Consider a set of wheels standing on a piece of straight track. The flanges are not in tight contact with the inside edge of the rail and as they move they stay parallel to the inside rail edge. The coning on the flat part of the wheel tread (present on good wheel sets) ensures that gravity keeps the wheel in the centre of the track and it should not hunt from side to side. In theory there should be no derailments at all.
Now consider a pair of wheels travelling on a curved piece of track. There is a lateral force seeking to push the wheel set towards the outside of the curve. At reasonable speeds this is balanced to a greater or lesser extent by the gravitational effects of the tread coning, and a state of equilibrium exists. If there are no sudden increases in the radius of curvature, the flanges are kept away from the edge of the rail. With tighter curves and at greater speeds, the flange leading edge is forced against the edge of the rail more often and hunts back and forth..
Whenever the leading edge of the flange is forced against the inside edge of the rail and there is a rail joint present with enough roughness or gap to give the flange edge a purchase, the wheel may climb out of the four foot and you get a derailment. I mentioned that the wheels may hunt from side to side when travelling too fast. This explains why the derailment does not always occur in the same place. The laws of chance indicate that the right conditions for a derailment will only occur from time to time, all we can do is ensure that these can occur as infrequently as possible, by laying smooth long radius curves, with transition and ensuring that the back to back of the wheelsets, the squareness of our chassis, the freedom of movement of bogie pivots and so on are all at the optimum.
Using set track helps with sharper radii as the gauge widening helps to keep the flange leading edge off the edge of the rail.
Child Psychology: A child will soon lose interest if the running is unreliable. Yes, but they will also lose interest if everything runs smoothly all the time. They will want to run the locos flat out or connect all the stock up to the locos in a double header. Often the intention may be to create a derailment and get adults attention.
I hope this helps.
Colombo
