Spline fretting wear pattern 3

[driller1]

I just joined after reading a recent thread - Splined Shaft Wear (thread 330-318697). I'll apologize in advance as I"ll likely miss nuances of posting etiquette.

I have been dealing with a significant wear problem on the splines driving some hydraulic pumps on a large piece of equipment. We've figured out a lot of the causes - mostly related to the equipment manufacturer not following the specific recommendations the pump manufacturers make for the mating female splined shaft in the pump drive. The wear pattern on the splines is sort of a compound curve rather than straight like you might expect when two straight splines are rubbing. I think it is probably related to misalignment, but haven't been able to get a good explanation of the mechanism (movements?)that would create the pattern.

Background: The biggest problem was that in a new model of pump drive for the machine's hydraulic pumps, the equipment manufacturer decided a 35 Rc 4340 shaft was good enough. The pumps (Sauer-Danfoss, Eaton, Parker) all have SAE D 1.75" or SAE C 1.25" splined input shafts with hardness near 60 Rc & all of these pump makers call out for a hardness in the range of Rc 55-62 for the mating female shaft in the pump drive.

This was compounded by 1. A lack of lubrication other than the initial assembly with a high temperature anti-seize. 2. A shorter length of spline engagement less than recommended and 3. A concentricity specification that allowed a runout of .010 TRIR between OD diameter of hollow shaft where it is supported in taper roller bearings and the female spline pitch diameter on the inside of the shaft. The prior model of their pump drive had a carburized 8620 shaft matching the hardness of the pumps' input shaft & the other factors didn't impact nearly as much- average life > 10,000 hours vs <3,000 hours now.

Here is a link to a few pictures showing the wear pattern.

https://www.dropbox.com/sh/hfsspv2vqod2f8b/-HBaWFOE6J

I'm looking forward to the input & hopefully will be able to provide some to others from time to time.

 

[Metalhead97]

driller,
I'm assuming I'm looking at the male-spline, which is directly attached to the pump. Am I correct? What's confusing is that you said the pump driver shaft (the female end) was significantly softer. How much wear are you seeing on that side of the couple? The harder material should wear less. Slight misalignments are self-correcting through the break-in wear period if the assembly is designed correctly. However, what I'm seeing goes well beyond that.

If the information you provided is correct I suspect that the softer material (Rc 4340) was worn pretty good. In addition, I'm assuming that the alignments are set by a fixture (which allows some slop), and that hasn't changed. I would bet the other compounding problem is that you’re getting a bunch of Rc 4340 debris, and this is creating a 3-body wear situation. Wear debris is oxidized, cold-worked material, which means that the debris will act like an abrasive to the harder steel surface. Technically, with this much wear you are far from a fretting type of wear situation as fretting wear requires that surfaces be in constant contact with small relative movements. However, in this case you probably are in some sort of reciprocating wear situation due to the extra sliding as the shaft starts and stops. I hope this helps.

MH

http://www.linkedin.com/pub/luke-autry/1b/510/566

 

[driller1]

Metalhead97,
Thanks for the reponse. It is the male splined shaft going into the pump. It is a Parker P2 axial piston pump plugged into a pump drive which does NOT allow much slop, the pilot of the pump is a tight fit in the matching pilot diameter of the pump drive and it is held solidly against the drive so there is no axial movement. But, if there is misalignment-say like a shaft that where the female splines are not concentric with the OD of the shaft which is where it rides on bearings- then it is very hard to correct the misalignment.

Also. you are correct on the softer material wearing more quickly. It may start as fretting, but quickly goes to a nasty mix of oxidized fretting debris mixed with metal splinters.

The machine's application does create a lot of changing torque forces on the splines, but that wouldn't seem likely to create the wavy pattern of wear. The literature suggests that the alignment needs to be nearly perfect in this tyoe of rigid mounting,so our next step is to get some very accurate measurements next time we take the machine down.

 

[mfgenggear]

driller

see this thread

http://www.eng-tips.com/viewthread.cfm?qid=318416

mfgenggear

 

[driller1]

mfgenggear,

Thanks, it was that thread that lead me to join & post. Our company has a quite a few PTO driven pumps in similar application to his, so we've been down that road. The solutions were similar to what was posted, although the posts weren't as clear or definitive as they might be.

If the engine is idling, there often isn't enough torque to keep the splines loaded & you can actually hear the clatter from 20 feet away. We copied Southern California Edison's solution since they have a huge fleet & installed a high idle solenoid that automatically kicks engine up to about 900 rpm as soon as PTO is engaged which solves most of the problem.

Lubrication is also a big deal, if they can't do an oil bath, then most of the PTO manufacturers recommend periodically pulling pump back out of splines (you usually don't have to remove the hydraulic hoses) & reapplying the special high temp moly-disulfide type grease.

The wear pattern on those splines was pretty straight, which is what I'd expect to see, and makes the wave pattern of wear on our splines a mystery.

Thanks again for responding.

 

[mfgenggear]

Driller

do research on dry film lube, IE coating the splines
a tighter fit (class 4 ) splines would also help. less backlash
standard splines are class 5
a major dia fit spline may also help, but cost would raise significantly
it should be good for this type of environment such as this.

all I can do is advise since I am a mfg of gear & splines, not a designer.

Mfgenggear

 

[gearcutter]

driller1,

I have seen and reverse engineered many splined couplings in this type of application. For the majority of the time; the couplings are simply under-rated for the load that they are expected to transmit. I feel that if a splined coupling, in an industrial setting, is designed correctly and sees no axial sliding during its life, then it should never need lubrication.

The link which Mfgenggear posted shows how to conduct a basic design validation of a splined coupling.
I've never thought it wise for an engineer to solely rely on data from a product catalogue without first conducting a validation of what the data claims. I would suggest that this is the first place you start.

Here is another link where similar issues are discussed. In one of my posts you'll find a possible explanation as to the uneven contact seen in your images -

http://www.eng-tips.com/viewthread.cfm?qid=318697

Ron Volmershausen
Brunkerville Engineering
Newcastle Australia

http://www.aussieweb.com.au/email.aspx?id=1194181

 

[mfgenggear]

Gearcutter

That is a great recommendation.
I can see how an engineer can get caught up with the mfg data.

about the lubrication I been doing research on dry film lubrication.
It would seem to me that in this application it would help prevent wear or fretting, and
for a dirty environment it would not attract foreign debris.

Mfgenggear

 

[Metalhead97]

driller,

One of your biggest problems is that your female spline shaft has a 35 RC hardness. That shaft will wear quite a bit and create a lot of abrasive debris during the process. As soon as the female spline shaft starts wearing significantly you move away from your design conditions. Going over what you said in your OP the female spline shaft hardness was the most a significant change it seems.

MH

http://www.linkedin.com/pub/luke-autry/1b/510/566

 

[driller1]

Gearcutter

I went back and re-read the other thread you recommended. (

http://www.eng-tips.com/viewthread.cfm?qid=318697)

"Angular misalignment generates a small oscillatory movement, between mating spline teeth, which traces the path of a figure of 8."

I can see how that would happen. Would a concentric misalignment also create a similar wear pattern? I have a harder time visualizing that.

The hydraulic pump manufacturers have warnings like this one from Sauer Danfoss: "Alignment between the mating spline’s pitch diameters is another critical feature in determining the operating life of a splined drive connection. Plug-in, or rigid spline drive installations can impose severe radial loads on the shafts. The radial load is a function of the transmitted torque and shaft eccentricity." Sauer Danfoss' spec for concentricity of the two shafts diameters is .04 and we have recently learned that .010 is the tolerance specified for the concentricity between the female splines and the outside diameter on the female shaft wheres the bearing that support it in the pump drive. Having the female splines have a concentricity specification that is 250% of the pump manufacturer's specification doesn't sound like a good idea.

I will get a copy from the library of Alban's "Systematic Analysis of Gear Failures" that mrfailure recommended, are you aware of anything else published that might have detailed information on wear patterns?

Thanks to all for your posts.

 

[gearcutter]

Concentric misalignment will generate radial contact lines.

When using the major diameter to check spline runout; first ensure that there is no runout of the major diameter relative to the pitch diameter of the spline. A poor manufacturing procedure can result in there being runout between these two diameters.

When you are showing dimensions please specify metric or imperial as I don't know what you mean by 0.010 & 0.04.

Albans's book is OK but it's very basic and falls short of helping explain a fair few of the gear & spline failures that I've seen.

Below are several images that will help you to better understand what's going on. The images relate to angular misalignment gear couplings but the principles are the same.

TOOTH SWEEP OF GEAR COUPLING

LOADING CYCLE OF MISALIGNED COUPLING

Ron Volmershausen
Brunkerville Engineering
Newcastle Australia

http://www.aussieweb.com.au/email.aspx?id=1194181

 

[gearcutter]

The images in my previous post are from a paper 'The Evolution of Gear Spindles' by Jon R. Mancuso & Joe Zilberman - Kop-Flex, Emerson Power Transmission Corp.

Below is an excellent table showing the contact pressure distribution of partially misaligned spline teeth.
Notice how much the pressure increases with only 0.2 degrees of angular misalignment.
The image is from a paper 'An analysis of misaligned spline couplings' by S Medina & A V Olver - Tribology Section, Department of Mechanical Engineering, Imperial College of Science, Technology and Medicine, London, UK

Ron Volmershausen
Brunkerville Engineering
Newcastle Australia

http://www.aussieweb.com.au/email.aspx?id=1194181

 

[mfgenggear]

Gearcutter

Very Very good information.

Mfgenggear

 

[mfgenggear]

apparently having misaligned gears or splines is very bad.
and is the cause of many failures.

lack of lubrication is also very bad & causes premature failure.

Mfgenggear

 

[Metalhead97]

Driller,

You said yourself that the concentricity of the female spline with the OD of the shaft it rides on is the type of misalignment that's a concern for you. Does the female spline shaft appear to be wearing evenly?

MH

http://www.linkedin.com/pub/luke-autry/1b/510/566

 

[Tmoose]

Each element of a typical 2 element gear coupling sith spacer is only subjected to angual misalignment. I think most manufacturers forbid installations that subject a single element gear coupling to parallel misalignment

 

[mfgenggear]

Tmoose

that's the issue in an nutshell, & yet it rears it's ugly head again & again.

If it's not permissible yet it seems some mfg's are ignoring or not really making sure
the splines are concentric to each other. It's like the lessons learned are ignored.
& short cuts, or lack up knowledge (naive) is prevalent.
& why "as Gearcutter points out" are these spline so under sized.

There has been statements that a spline is nothing more than a key way, I have heard this statement at various times
through out my career, only to now realize that spline are more complicated. and can not be
so lightly considered.

It is a privilege to receive the knowledge , & experience from all of you.

Mfgenggear