I've built my wheelbase testing unit with the following possible wheelbases
(starting at 105mm)
which is slightly longer than the wagons I have: 123; 138 and 150mm maximum.
See the holes holding the bearings (grub screws) for adjustment.
I also used John's equation of B=2RS/L with B=wheelbase, R=radius, S=difference between track and wheel width (1.7mm on the wheels), and L which is a separate calculation as a relation between wheel radius and flange depth and mine equated to 16.27.
It gave a wheelbase of 146mm.
BUT from the pics it already seems not feasible.
Nevertheless I will run all the wheelbases and report back as soon as the postman arrives.
There is a system used by some modellers of vintage passenger stock that have 3 rigid axles (1 near each end and 1 'amidships') that I have seen used successfully.
Each axle/axle box unit looks rigid but is actually pivoted at mid axle to allow some swivel around that central point. These axle 'units' (2 axle boxes + 1 axle) are then laterally connected together by a light longitudinal spring wire under the floor that maintains some light control over the axle units but allows sufficient pivot to negotiate curves. I've not used this system myself. I don't know how they manage the solebar suspension springs, brake shoes & rodding etc. Should also work on 2 axle vehicles ??
Coupler swing - If the head of the coupler is outside the outer rail then you are very likely to end up with derailment problems due to the amount of sideways loading transferred back to the chassis from the coupler.
distance of axle from end of vehicle - the closer you can get your axle to the end of the vehicle then the more you can reduce the amount of coupler swing - check out the difference at each end in the photos you have posted.
The further you axles are out toward the end of the wagon, then the you need a bigger clearance for the wagon on the inside of the curve. You can quickly test this by fastening a pencil to the middle of the wagon on the inside of the curve, and move the wheels to a different position and watch the path traced by the pencil along the curve. (Remember to move the pencil so that it is central to the changed wheelbase for most obvious "worst case"results)
Salada has mentioned the linkages between swivelling axle boxes. Look for "Cleminson" suspension and "radial truck" for more information.
Also consider using check rails that bite on the back of the inside flanges. This should help to avoid flange-climbing on the outer rail. For a quick and dirty test, strip a length of rail, bend to follow the inner rail and butt the foot of this rail up against the moulded rail fasteners - supuerglue this check rail in position. It won't be accurately gauged but it should be "close-enough". In that the flange should pass through in the gap OK and the back of the wheel should rub against the inner edge of the check rail enough to prevent the leading flange on the outer rail from climbing.
Just some other stuff to play with while you are in "experiment mode". It's an interesting test rig and I'll be honest on that I haven't seen done before.
Unanderra in oz
Well done John. "Cleminson" is the name of the method but I'd forgotten that.
Yup, 'working' check rails DO work. My fledgling colliery model uses them on the bits built so far; also 'working' Xing guard rails.
Increasing the weight of wagons of doesn't prevent flange climbing, neither in 'O' Gauge (tried it) nor 20" gauge cable hauled mine incline - see photo below:
Mine trams always try to climb this outer switch rail, not helped by the cable roller position. I loaded the little swivelling/end tipper seen in the distance to about 12 cwt but it made no difference. Looks like breakfast was burning !.
Photo by Salada.
Last edited on Tue Sep 19th, 2017 11:49 pm by Salada