grovenor-2685 wrote:As well as being out of place in a discussion on vehicle suspension the discussion above does raise some issues with the standards that perhaps need a bit more exploration. i will come back to it after a bit of preparation.
i wrote this after reading a post from Tony Wilkins that I thought was on here, however, I find it is actually on the P4 v S4 topic.
The relevant sentence was this:
I feel I should point out that there is actually an error with the P4 standards which has been there from day one. In the published standards the quoted figure for the Check Gauge (BB plus EF) is 18.15mm Since the min BB is 17.67mm and the EF is 0.38mm this adds up to 18.05mm not 18.15mm. This means that for a wheelset with 17.67mm BB at a check rail the flange of the opposite wheel wiil not be in contact with the running rail but 0.1mm away from it. This may not seem important, but if the wheelset is running through a crossing with the flange in contact with the stock rail, the back of the opposite wheel flange will strike the knuckle of the crossing as it overlaps it by that 0.1mm difference assuming that the track gauge is the nominal value.
This caused me to look at the standards as I had thought that the set of equations used to validate the figures was complete. Examination revealed that the condition given by Tony does NOT in fact get checked by any of the equations and an additional equation is needed. This would be:
(BBmin + EFmin) - (TGmin - CF min) >/= 0. This is effectively the converse to equation 2.
Putting the figures to these shows that it is required to eliminate the wide tolerances that the P4 standards give to BB, EF and CF to allow both this new equation and equation 2 to be satisfied.
See attached calculations.
The reason this does not give significant trouble in practice is that the problem will only occur when the wheelset is running up against the stockrail, ie on the larger radius of a curved turnout where the crossing is on the inside rail of the curve. Also if using CF gauges correctly CFmin should never be encountered.