Polygon Vs Involute spline question
Polygon Vs Involute spline question
(OP)
We are looking at an OEM part, for a customer, which uses a polygon spline between shaft & hub. The original configuration used an involute spline but this was superseded.
The hub, having a relatively thin wall section, is showing signs of cracking between the polygon profile and the outside diameter.
This problem has only begun to appear since the change-over of spline profiles.
We have been asked why this might be happening and I suspect it may be due to the polygon profile creating greater radial & tangential forces than what an involute profile might.
We have software which calculates the separation forces of mating involute profiles but nothing for a polygon profile. The calculations are based on the pressure angle. My assumptions about the polygon profile are based on looking at the profile as being a large pitch & large pressure angle, 3 teeth spline. We have been able to model something close to a polygon using the gear/spline software we have but unfortunately the software can only cope with a certain amount of profile correction before it effectively crashes.
Would I be correct in assuming that a polygon profile does create greater separation forces and how could this be proven?
Any help would be greatly appreciated.
The hub, having a relatively thin wall section, is showing signs of cracking between the polygon profile and the outside diameter.
This problem has only begun to appear since the change-over of spline profiles.
We have been asked why this might be happening and I suspect it may be due to the polygon profile creating greater radial & tangential forces than what an involute profile might.
We have software which calculates the separation forces of mating involute profiles but nothing for a polygon profile. The calculations are based on the pressure angle. My assumptions about the polygon profile are based on looking at the profile as being a large pitch & large pressure angle, 3 teeth spline. We have been able to model something close to a polygon using the gear/spline software we have but unfortunately the software can only cope with a certain amount of profile correction before it effectively crashes.
Would I be correct in assuming that a polygon profile does create greater separation forces and how could this be proven?
Any help would be greatly appreciated.
Ron Volmershausen
Brunkerville Engineering
Newcastle Australia
http://www.aussieweb.com.au/email.aspx?id=1194181





RE: Polygon Vs Involute spline question
Can you show an example of a polygon spline?
Is it simply 1/2 of a hexagon giving you a
30 degree pressure angle? Does it have any
significant radii in the corners?
RE: Polygon Vs Involute spline question
http://www.generalpolygon.com/profiles.htm
These are usually used with thick-walled hubs and the stresses generated are mainly contact (compression) type. If you have a thin-walled hub, then sure, there could be high radial and tangential stresses.
There have been journal articles that cover this area:
J Mat Pro Tec 109 (2001) 30-37
J Mech Des 122 (2000) 130-135
The second one in particular is a direct comparison between polygon and involute spline.
Regards,
Cory
Please see FAQ731-376: Eng-Tips.com Forum Policies for tips on how to make the best use of Eng-Tips Fora.
RE: Polygon Vs Involute spline question
http://ww
Regards,
Cory
Please see FAQ731-376: Eng-Tips.com Forum Policies for tips on how to make the best use of Eng-Tips Fora.
RE: Polygon Vs Involute spline question
The diference between a spline conection and an internal gear one is that the presure angle of the spline does not induce a radial force on the hub as all power is transmited only as if a keyway was used (in this case several around the shaft) in an internal gear(hub) external gear (shaft) conection the presure angle of the gear teeth will push the hub radially outwards, in the polygon shafts what you are looking at ar 3 or 4 teeth gears with a very high presure angle, and we know the higher the pressure angle the greater radial force is produced on the surfaces involved, what in this case will transform in a failure of the hub because of this new force (the splices caused practically none)
Regards
SACEM1
RE: Polygon Vs Involute spline question
http
http://www
RE: Polygon Vs Involute spline question
You mention you are experiencing cracking between the polygon female socket and the shaft OD it is located within. A couple of questions:
- Where do the cracks originate from? If the polygon coupling has a loose fit, the cracks may be propagating from fretting/pitting at the coupling contact surfaces, and not from high hoop stress concentrations at the thin wall sections around the coupling lobes.
- What is the function of the OD surface around the coupling socket? If it is something like a bearing shaft, that must maintain an accurate shape/profile when loaded, then you are better off with the involute spline. The involute spline, with a low effective pressure angle, will not cause high hoop stress and distortion of the OD profile. The polygon coupling (as SACEM1 noted) has a high effective pressure pressure angle, thus the high induced hoop stress.
-What is the material and heat treat of the coupling mating parts? If your budget allows, I would recommend case hardened and ground contact surfaces for the coupling.
- Is there a requirement that the coupling halves be free to slide axially? If not, then you might want to produce the polygon coupling with a tapered fit, and draw the two halves together tight with a nut or bolt. That will help to eliminate fretting issues. As a minimum, I would also recommend a thin layer (.0002-.0004 inch) of soft plating, such as silver or tin, on the coupling contact surfaces.
- What provisions are there for lubrication? If the spline can be oil wetted, then you should dam both ends of the coupling and run the coupling interface submerged in oil. The constant flow of oil through the coupling interface will flush out any metallic micro-debris that is inevitably generated by any sliding fit spline or coupling. If the coupling is grease lubed, make sure the coupling interface is tightly sealed and packed full with an EP grease at assembly.
- What is the L/D ratio of the coupling? If the coupling Length/Diameter ratio is more than about 1.5, it is likely not uniformly loaded along its length, due to torsional wind up. Keep the L/D ratio below about 1.5 if you can. Also make sure both ends of the coupling have a similar radial stiffness in their cross section if possible.
Good luck.
Terry
RE: Polygon Vs Involute spline question