What's up with the "gear bearing" idea? 1

[Windward]

Why are there no large gearboxes using the gear bearing concept? Say in the wind turbine industry? I can't find any.

Early reports did not indicate that there should be no upper limit on the rating of gearboxes of this kind. In 2002:

"Planetary transmissions could be simpler, cheaper, and more rigid. Gear bearings are conceptual mechanical components so named because they function as gears and as roller and/or thrust bearings. Gear bearings will be essential components of the next generation of compact, large-mechanical- advantage gear drives."

http://www.techbriefs.com/component/content/article/ntb/tech-briefs/mechanics-and-machinery/5848

Note that it says large ratio, not large rating. That still seems to be the case. In 2013:

"For medium-range applications, such as in power tools, cast gear bearings could reduce cost and size. For low-tolerance applications, such as toys, injection molded plastic gear bearings could substantially reduce cost. Beyond reducing parts, the high-load capacity of gear bearings can further reduce cost by enabling the use of less expensive, lower strength materials."

http://www.gearsolutions.com/article/detail/6321/the-future-of-gear-bearing-technology

There are later patents for this device, but here is the 2003 US patent number for the basic arrangement of a gear bearing gearbox:

6,626,792

If the gear bearing idea is practical for small gearboxes, why isn't it practical for large gearboxes?

 

[MikeHalloran]

It appears that it requires the gears to run with zero lash, which in practical terms, makes them impossible to manufacture in quantity.

My friend Wayne spent most of WW2 rebuilding planetary supercharger drives by selective fitting of planet gears. The gearboxes still fragged themselves with regularity; I don't think that's a reflection on Wayne's skill, just on what was then possible.





Mike Halloran
Pembroke Pines, FL, USA

 

[Windward]

That makes sense to me, Mike. The gear bearing gearbox has no means to compensate for unequal load sharing among the planets. This inequality would be inevitable because of manufacturing tolerances and other reasons. For example, to reduce planet gear load imbalance, Timken has the Flex Pin for adjusting the position of the planet gears during operation. [It appears that Timken did not invent the Flex Pin, but maybe they have improved it.]

Without any means for adjusting the gear bearings to reduce load imbalance, a gearbox of this kind would require perfect parts and perfect assembly. That may not necessarily mean zero backlash in all of the meshes, but it would mean the same backlash everywhere, taking into account the different ratios in the stages, so I believe your observation is correct.

Then what happens after the gearbox is in service, when the meshes will be affected by the load, wear, and differential expansion from changes in temperature? Probably the situation would get worse.

These problems would not be as serious in smaller gearboxes, where the benefits of the gear bearing might overcome the disadvantages. But the disadvantages would be too great in a large gearbox.

 

[tbuelna]

NASA studied and tested the Self-Aligning Bearingless Planetary (SABP) concept back in the 1980's. You can read more about it here.

 

[Windward]

Thanks tbuelna. This is a high-quality study, but I have only skimmed it. It mentions a flexible output ring gear and straddle-mounted planets to improve load sharing.

Since John Vranish patented his gear bearing in 2002, his invention must be somewhat different from the gear bearing of this 1988 study. I cannot determine from the study exactly what their gear bearing looks like, but again, I have not read it in detail. They did not conduct full load testing, only spin testing to measure noise and some other effects.

I don't know and can only wonder whether any gearboxes of this kind were ever used in helicopters or elsewhere. If not, it would seem to indicate that a gear bearing gearbox is not suitable for the power levels and speeds in this study, about 500 hp, input speed 35,350 rpm and output speed 348. The gear bearing section, last in the train, produced less than 1/5 of the overall reduction ratio.

In other words, in the thirty years since this study, apparently no one has shown that the gear bearing gearbox is useful for anything other than low power, low speed, low precision, low cost applications. Not that there is anything wrong with that.