Designing sliding spline to minimize torque lock up 6

[ATVextreme]

Developing a CV axle with fixed end joints and a sliding spline in the center. Testing the prototypes found that the cages are breaking in the fixed joints because the splines are not slipping under heavy trottle. Current design is involute splines coated with blue antivibe material (Supplier English version of print says teflon, maybe be nylon). Looking for advice to three questions:
1. First question is would steel to steel splines slip better than steel to coated splines? Assume coating is standard driveshaft blue material. In this application noise and vibe are not the big concerns.
2. Second question would parallel splines reduce the torque lock up significantly versus the current involute spline? Involute splines were origonally chosen to simply manufactoring the axle shafts.
3. Last question is would changing the length of the sliding spline and surface area of tooth to tooth contact change the coeficient of friction?

Thanks for any feedback in advance.

 

[MikeHalloran]

Are you using grease in addition to the coating?
What's the prospect for stronger cages?
Does the coating manufacturer provide PV ratings, so that you might stand a slim chance of actually engineering the connection? In general, more engaged area should help, but numbers would tell you how big a change you need to make.
How do you feel about ball splines?



Mike Halloran
Pembroke Pines, FL, USA

 

[GregLocock]

The wedging action of an involute won't be helping.

My advice would be to develop a spec for breakaway axial thrust vs torque and talk to suppliers.



Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[tbuelna]

Most involute splines are side fit splines, so there is no feature to keep the two halves of the spline aligned axially. Since your center located spline joint has rotational degrees of freedom (ie. the CV joints) at each end of the axle shaft, the spline joint halves can "cock", which causes edge loading and binding in the axial (plunge) direction. A short spline engaged length will make the problem worse.

A remedy to this issue would be to increase the spline engaged length, so that the "cocking" angle is reduced. Or you could switch to a major diameter fit type of spline, which would provide axial alignment.

With regards to static/sliding frictions, surface contact area is normally not a factor.

Hope that helps.
Terry

 

[ATVextreme]

Thank you for the quick responces. Yes there is grease on the splines. There are physical limitations to increasing the size of the cages, but we are considering changing the material from 20CrMnTi to 8620 or 4340. Ball splines were considered initially, but this design was simplier and more cost effective.

The axial alignment may be the issue.

What are thoughts on spline coating versuses steel to steel and is there a standard specification for material and thickness of coating.

Thanks again for the help.

 

[MikeHalloran]

I'm suspicious that pressure developed on the spline flanks from the applied torque may be exceeding the pressure limit of the plastic coating. I'm concerned that it's described as an 'antivibe material', not a 'low friction coating' and not a 'spline coating'. Has the manufacturer actually recommended it for splines, or is this a 'creative' application?

Steel on steel can do very well, provided the grease is actually present and dirt is excluded. Off road, that requires boots, so far not specified.




Mike Halloran
Pembroke Pines, FL, USA

 

[ATVextreme]

Design does have boots as this application is off road and must be water proof. Tier 1 Supplier recommends this coating for friction reduction, but not able to speak to tier 2 supplier applying the coating. Is there a standard call out for thickness or for process of application of teflon on splines. For example .25mm maximun thichness.

 

[CoryPad]

The coated splines will have lower friction. Here is a typical coating for this application:

http://www.wrightcoating.com/services/nylon-coating/drive-line-nvh/default.html

 

[MikeHalloran]

Please forgive me if I'm wrong and/or being excessively rude, but:

It smells like you intend to solve a production problem by blindly changing a purchasing specification, without leaving your desk, based on exactly zero analysis and exactly zero actual data aside from the existence of one or more failures in a part that you may or may not change.

What I mean by 'zero actual data' is that you haven't gone out to the production line with a torque wrench and a force gage and tried to assess, not a friction coefficient directly, but a transfer function between torque and thrust for the splines as used right now.

My smell detector says that if the friction were anywhere near bad enough to be the root cause of the reported failures(s), some binding would have to be noticeable in normal riding as the suspension articulates.

Speaking of that, the suspension probably has rubber bushings instead of, e.g. rod ends, so it's not totally rigid. Which means that the arms may not pivot around the nice centered pivot points in your drawings and CAD models.

It also appears, talking about wobbly suspension bits, that you have not eliminated as a root cause, the possibility that the shafts are just a bit too long, and running out of spline travel is what's killing the joint races.

Have you broken a joint yourself? Go and do so.


Mike Halloran
Pembroke Pines, FL, USA

 

[patprimmer]

While plastics can be slippery, they are also soft and have low melting temperatures compared to steel. They definitely have a maximum PV WELL BELOW steel.

While the surface speed here will be low, the pressure is probably very high.

With a high pressure low speed case, nylon should be better than PTFE (Teflon) as it has less creep and tendency to stiction caused by deformation of the surface by prolonged high load.

Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &

http://eng-tips.com/market.cfm

for site rules

 

[ATVextreme]

Yes, I must say your last reply smells very rude. I am not an ivory tower engineer. The production floor is somewhere in China. The testing was done in house in the USA. The evaluation of the broken parts was by myself and others. There is no question as to the root cause being axial forces acting on the cage under heavy trottle and suspention articulation. We spent much of yesterday in the shop "playing" with different shafts and splines. We hung weights off a bar welded to the shaft to consistently apply torque and measured force to slip. The origonal intent of the post was to ask theoretical questions to better understand best practices from the experts. I believe there is some very good feedback in several of the replies and it is greatly appreciated. I spent the last 10 years in an automotive assembly plant and consider myself very hands on. At the end of the day after all of the data collection what really counts is what specification is documented on the print.

 

[theoldwizard1]

ATV, I'm curious. What did your static testing show ? Were there any original design specs/assumptions ? Do you have any "real world" data showing the actual torque applied to the shaft ? (It could be much higher than expected if the vehicle drive wheel is on rough terrain.)

Those splines need to be cut using tight tolerances and good quality control. The smallest nick/bur could cause them not to slide under high torque. Have you done a close inspection on the spline on the failed parts ?

Better lube is the obvious answer, with the limit being what is practical.

 

[Tmoose]

fairly standard practice nylon coating ca 1979.

FROM SAE "Univ Jt and Driveeshaft design manual" Advances in engineering series #7.

The 1979 nylon may well be The 2010 "blue anti-vibe" material

I'm unclear as to why a CV joint with slippery plunge capabilities is not being used at one end.

 

[ATVextreme]

Thanks again for the added information. As far as loading on the shaft the unit was measured on a dyno to 47hp and 36ftlbs. This does not account for impulse or shock loads. The grease is synthetic Mobile 1. There were no nicks on the splines, but the "blue material" showed a lot of wear after 100 hard miles of testing and was peeling off the steel. The reason the shaft does not use a traditional inner plunging CV is the operating angle is above 30 degrees.

I have been looking for the correct standard to read and reference for the sliding splines sizing and tolerances. Based on feedback in this thread it sounds like a major diameter alignment is better than a side fit alignment. I am not clear which is appropriate the SAE B92.1-1996, ISO 14:1982 , DIN 5463/64, or something else. I plan on purchasing a copy of a standard, but not sure which one. Any feedback to direct me to a standard would be helpful. The current print simply reads 19T, M1.25, 45degrees.

Thank you again for any help.

 

[MikeHalloran]

The 45 degree PA suggests that the spline is rolled, hence doesn't really have a major diameter, just a tip ridge that may be bifurcated.



Mike Halloran
Pembroke Pines, FL, USA

 

[ATVextreme]

Yes current design in protypes is an involute design and is not aligned by the major diameter. The tier one supplier does roll their splines. This shaft is farmed out to a tier 2. This specific spline size was choosen because tooling was available, fit in the physical dimensional requirements, and exceeded strength of the shaft. The female is a 150mm broached sleeve.