[TheForge] Re: Flatbelt excitement

[xlch58 at swbell.net]

Mike Spencer wrote:

>Bob> If I remember right the idler should be on the powered side of 
>Bob> the belt rather than the slack side.
>
>xlch58> Just the opposite....
>
>That's right.  Tensioner keeps the slack side from flapping.  In easy
>cases  it's just "stick it there".  But I might have to think about or
>experiment with spring vs. gravity, spring tension and mass of the
>idler.  And maybe it's not the way to go anyhow.
>
>All the previous flat belt stuff I've done has worked first try, plus
>or minus a bit of tuning up and none of it ever needed a tension idler.
>
>
>- Mike
>
>  
>
My guess is that the problem is not over running, but is probably a 
resonance issue.   Given that the natural frequency of the belt system 
is probably already pretty low, I expect that you need to find a way to 
raise the natural frequency of the system.   Shorten the belt which will 
raise the spring constant,  or lighten the  flywheels, thus lowering 
mass moment of  inertia.  Or you could raise the rpm-- it would flap a 
bit as you came through the resonant rpm, but would settle down if you 
got enough above it.   You might still need an idler if you can't 
establish enough distance between the resonant rpm and the running 
rpm.   There are a lot of materials out there on torsional vibration.

Charles