Simple Behavioural Models: Fluid Coupling
I'm in the early stages of developing a railway-simulation game, and would like to model the physics of the trains considerably more accurately than existing games in this genre. I've already found most of the information I need to do this, but I have some specific questions which are difficult to answer using readily available information. Most of them have at least something to do with hydrodynamics, which is why I'm here.
My first question relates to fluid couplings. I thought I should ask about these first, as they are simpler than torque converters but share the same basic principles. Both of these are heavily used in the railway industry, usually in the driveline of DMUs, but also in some locomotives.
So far I have two sources of information, which are not sufficient to build up a full model to the accuracy level I want.
One is a spec-sheet from Voith Turbo which describes the full-throttle torque-vs-output-speed curve for one of their transmissions. This shows that at relatively high speeds, the coupling between the engine and output shafts is fairly tight, as I would expect. But I also need to know about the behaviour at low speeds, since DMU transmissions from the 1950s use only a fluid coupling, and these appear to allow sufficient slip to allow engine idle with the train stopped. The Voith spec-sheet describes a transmission which switches to a torque converter at low speeds.
My second source of information is the generic formula which describes "the coupling" as related to the fifth power of diameter and either the square or the cube of "the speed". It's not clear whether the speed refers to the slip between the input and output, or the speed of the faster half of the coupling. It's also not clear whether to use the square or the cube formula, or the meaning of the parameter this produces.
What I need to know is the qualitative behaviour of the coupling at various input torques and speeds, and the full range of output speeds. At this point I would accept a reference to a reasonably-priced book that has sufficient detail.
What I don't need to do is perform a full CFD analysis. Such a thing would not run in realtime, which is necessary in the context of a game. :rolleyes:
|All times are GMT -4. The time now is 00:47.|