Will L wrote:
I don’t really believe anybody really thinks that current springing methods are in any way a representation of the real thing, and anybody who has tried has quickly found out that the physics of mass related things just doesn’t scale.
I thought the way the original mention was written, it ambiguously implied that model springing followed prototype practice - which was therefore "better" than flexi-chas if only for that reason.
What I don’t understand is why your so set against using primary springs? The reason why vehicle get sprung is that it decouples the mass of the body from shocks transmitted when the wheels hit any irregularities.
That certainly sounds like obvious "common sense", but for simple linear springs, this is where the devil is in the details:
(1) In order to decouple the body, the wheel spring has to compress almost all of the height of the "bump" in order to not raise the body height. I.e. the force to compress the spring past its static equilibrium level has to be VERY much less that the force needed to lift (or at least begin to lift) the body. To me this implies a body mass/weight that is relatively heavy compared to the "rate" of the spring holding the wheel down, or a very "soft" spring. I would suggest that to work well, the body weight would ideally be an order of magnitude greater than the spring rate.
Of course, if the springs are relatively hard, then you tend to have track holding that is not much different from a rigid chassis. Ditto if the vehicle is light weight, no matter how soft the springs are.
But:
(2) To hold up the body in the first place, and allow wheel movement, the spring used should be in static equilibrium around about its 50% compression point. Which means you have to know or measure the body weight per wheel to get the spring strength in the ball park. However if the spring rate is "soft" as above, then a small error in the spring rate or body weight will have the spring equilibrium position sitting too close to, or even at the Up/Down "stops", and no effective springing at all.
This of course is the opposite of the requirement to decouple the body from track irregularities.
(3) From 1 and 2 above, you end up with having to compromise between track holding and "smoothing" of bumps, which means you are not going to get anything like the best of either. Plus, there are the issues of knowing the weight, adjusting the spring rate(s) (per wheel individually) to achieve the balance of holding/smoothing performance (per wheel) and making sure the entire chassis is stable. It would seem there is little possibility in the way of a one size fits all, or even an easy, simple to understand and install/adjust solution
The prototype gets round most of those issues by using significantly non-linear springs, such as leaf springs. These can act with soft spring rates in the mid range of their operating position, but exhibit much higher spring rates if the movement goes much beyond the mid range.
Precision non-linear springs are not exactly easy to obtain or make in small scales, but my suspicion is that CSB's are somewhat non-linear at larger deflections and that is why they are so much easier to set up and be stable on P4 models than simple linear springs. That however doesn't necessarily imply that either the springing or the track holding is optimal. Unless there is a body of analysis that has been done to actually predict both the equilibrium loads and dynamic spring rates of all the various wheels in a chassis. A practical test of the springing effectiveness of a CSB vehicle would be to see if pressing down with a light finger pressure , or some temporary extra weight (coins) , actually easily lowers the body slightly. If it doesn’t, then it probably is working in a similar way as the 70’s “Alan Gibson” half springing method.
(4) Springing of course doesn’t actually decouple at static, or relatively slow, speeds. It’s mostly a speed proportional effect. The faster the vehicle moves, the greater the delay of the springs trying to raise the body, with the greater likelihood that the wheel direction reverses before the body inertia is overcome. The more common, shorter P4 layout runs typically operate on the slow side of prototype speeds. And especially so when shunting. So then it’s not necessarily a major benefit or visually apparently different.
So overall, it’s not that I am arbitrarily set against primary springing. I just add up the pros and cons above, and see only disadvantages compared to the use of equalization, with fewer components and no need to match/adjust spring rates to suit the particular weight/speed/size/complexity of the vehicle chassis. In addition, using tilting beams that incorporate the wheel bearings, obviates the need, cost and installation accuracy of working vertically sliding hornblocks.
Will L wrote: Your favourite 4 wheel bogies needs this less than anything else as the pivot between bogie and vehicle body goes some way to providing this insulation, and, if you must, Mr Bedford does a bogie vehicle secondary springing unit which will complete the job. But on classic 4 wheel wagon, equalisation/compensation is just not so effective. I can accept that the improvement in running available from a sprung suspension can be quite subtle especially on well laid track, so, in the end, the question becomes how easy is it to achieve?
The real reason why sprung suspension has now become the method of choice for many who are happy to do more than just substitute wheels is that it actually represents a better modelling solutions. Certainly no harder to do than classic compensation, avoiding that difficult gap between axle box, spring and sole bar that comes up with a rocking W iron, and leaves the underside clear for the wagon detailing enthusiast to add all those brake gear straps, hangers and assorted gubbins. With the best will in the world, two rocking W irons plus and a couple of cross beams doesn’t sound as if it represents any form of modelling simplification
That's the part I'm looking into. Thinking with my inherently lazy attitude and my low volume manufacturer experience hats on, I'm trying for the fewest possible, all finished parts, no soldering, no measuring or uncertain positioning, self alignment, and easy assembly in a few minutes.
The ball's in my court to find out if that's an economic possibility. If it is, you might get a lot more satisfied newbie S4 Society members.
Andy
