Re: [CAD_CAM_EDM_DRO] Re: Belt tensioners on toothed belts

A belt flexing up and down, depending on it's length, would indicate a belt that was under-tensioned. SD-PI/Sterling has a PDF that shows the tension recommendations for different belts and the specs are quite high. Although, it's a trade-off in application. A belt run more tightly is more precise, but the resistance in transmission of torque is higher. Also the higher side loads shorten bearing life. So, one needs to find a happy medium between, tight enough to be accurate and loose enough, not to rob power.

Introducing an idler into the transmission surely increases frictional losses. Pre-loading this idler with any sort of spring is unneccesary. A bi-directional belt relies on perfect timing in both rotations. In one direction under load where an spring loaded idler is present, the idler/spring will relax under load, allowing mis-timing between driver and driven. Simply tightening an adjustable device into a held position should be ample. Wether it be an idler or simply the motor itself. The latter generally being the most efficient apart from direct coupling.

As a toolmaker, I learned early on and was told by my predecessors.."Avoid springs" they should always be a last resort and not a first option. It's amazing what can be accomplished with a simple counterweight and ever present gravity.

Ron



caudlet <thom@...> wrote:
--- In CAD_CAM_EDM_DRO@yahoogroups.com, "Ken Campbell" <deltainc@...>
wrote:

>
>
> ----- Original Message -----
> From: "caudlet" <thom@...>
> >
>
> ----- Original Message -----
> From: "Chris Horne" <chris@...>
> Subject: [CAD_CAM_EDM_DRO] Re: Belt tensioners on toothed belts
>
>
> > The error introduced is directly related to the slackness of the belt,
> > the diameter of the pullies and the screw pitch.> Chris
> *******
> Hi ... we are in the r/c car business , many use toothed belts (

usually HTD

> type.. strong for their weight ) .. anyway, tensioners are common, and a
> couple of flanged ball bearings also common and do the job fine ; For
> these bigger apps, consider a spring loaded arm, and stop by your local
> hobby shop and check out the small shock absorbers ( coil overs

available )

> used on dune buggies ... that would stop harmonics from building up,
> besides being " real cool" to show your buddies.
>
> (g) ken campbell, deltawerkes.
>

The difference here is that:

1. Postional integerty is important. It's probably not a problem on
an RC car if the position of the drive wheel is off by .005 in rotation.

2. The CNC usage is under conditions where the direction of drive
changes hundreds of times in a minute.

I was dismayed watching the belt bow up and down under the direction
changes by as much as 1/2". That change in drive length has to go
somewhere. I guess that it might tend to cancel out over a lot of
direction changes but could be cumulative over a long run if not
exactly equal. I would think the repeatability of the system would
suffer if belt flexing is allowed to happen.

Since this is the first time I have seen this design in a CNC machine
(spring loaded idler arm) I didn't know if the designer had discovered
a new approch or just wanted to avoid having to design a movable motor
mount. I have seen rigid rollers on belt drives to force a fuller
wrap around the drive pinion but the tensioning mechanism is at the
ends of the belt and can be tightened enough to prevent any flex.






[Non-text portions of this message have been removed]