I didn't see any search results addressing this specifically so I figured I'd create a thread as a reference for others among us with these questions:
In designing power transmission systems using gears, chains and sprockets, belts and pulleys, and numerous other types of mechanisms, a very important consideration is the center to center distance (here on out referred to as CD) between gears, sprockets, or pulleys to maintain chain/belt tension or a good mesh and therefore maximize efficiency and product life, particularly if looking to eliminate tensioning mechanisms to reduce complexity. I was hoping to get some input from those with lots of experience doing exactly this.
Gears are probably the easiest the deal with. You are given the pitch diameter of the gears you are working and you simply need to make them mesh at that distance. But it's not quite that simple. In FRC experiment, people have found it useful to add a few thousandths to the CD because we don't have a chance to wear our transmissions into a good mesh with the short period of use. Also, what kind of tolerance typically needs to be maintained on this distance?
The next simplest is chain and (timing) belt stages with two pulleys or sprockets of equal size. If the number of teeth is even, it's apparent that each sprocket uses half of its teeth in belt at any given time, and its pretty simple how that can be extended to sprockets with odd numbers of teeth (for large enough sprockets). This basic approach leads to a CD of the belt or chain length minus teeth on one of the sprockets, all divided by two (CD = (length - pulley size)/2 (in teeth) ) appropriately converted via the pitch. This means that the CD should be an even multiple of the pitch. But what tolerance matters in this case? Would experts typically undershoot by that tolerance or some percentage to ensure it is not over tight? Are there other considerations that should be taken into account?
It gets more complicated when the sprockets are no longer the same size. It is no longer trivial to assume how many teeth may be on each sprocket at any time, and also to determine the distances between the points of tangency with the sprockets. Basic trig would make this reducible
if the sprockets have enough teeth that the behavior is not much different from two perfectly round sprockets. But with sprockets with very few teeth, the deviation seems quite significant to me, and the problem is much less tractable. What methods have teams found useful in determining an effective CD to maintain tension in such a situation? Is the variation due to acircular sprockets a big deal?
And then of course there is the situation of a non synchronous belt on perfectly round sprockets, in which case the trig above can be employed, but there remains the question of relevant tolerances. I don't think this type of transmission is used nearly as much in FRC, but is still of some interest.
So I'm just wondering if there is some general wisdom from experience or insight as to how to handle these different situations. What works? What doesn't? What are some common practices? What are some of the known tolerances on tension and the center to center distance?
EDIT:
This thread will probably end up being relevant to the timing belt part of this discussion.
Also, I remembered after posting this that Dr. Joe posted
this once upon a time, which looks like it does take into account the non circular nature of sprockets, though I haven't yet bothered to figure how. Have others found this tool accurate for these smaller sprockets?