Splined Shaft Wear 9

[triumph993]

Folks,

Here's a photo album of the splined shaft from my hydraulic pump. The interior splines of the mating PTO are identically worn.

I'm in the process of identifying the steel, but I believe it to be a high tensile steel (4140 or 4340). My apologies for not being able to narrow this down just yet.

Wondering if these wear patterns tell a story to the metallurgists out there?

https://picasaweb.google.com/dmb2000/PTOPumpSplines?authkey=Gv1sRgCNSloqW9qonCaQ

Thanks,
David

 

[swall]

Based on the red iron oxide debris and the lack of gross deformation in the splines, this looks like a fretting failure. Fretting arises from small magnitude unintended movement between mating surfaces. Corrective action consists of eliminating the movement, providing lubrication and/or making a slight change to the materials.

 

[triumph993]

Thanks swall,

I notice that when looking at a spline tooth in the radial direction, the wear is curved; i.e., there's more wear in the center then the ends - the center is knife edge while the ends still have machine marks on the top edge of the tooth. Is this fretting or load imbalance or something else?

David

 

[Guest102023]

Could be a number of things. You first need to ask how long the spline and its mating part have been in service, as well as other history. Then look at the entire system: alignment, lubrication, etc.

I see blue tint that is suggestive of overheating, but photographing bright metal surfaces is difficult. Brownish-red, which can be well attached to the surface, indicates fretting wear, a.k.a. fretting corrosion.

 

[gearcutter]

While I don't disagree that fretting has been occurring, the amount of spline tooth cross section that has worn away, IMO, cannot be explained by fretting alone.
There is evidence that there are several other factors at play.
Have you fully investigated the possibility of misalignment as there is evidence that there is a non-uniform load distribution along the teeth flanks?




Ron Volmershausen
Brunkerville Engineering
Newcastle Australia

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

 

[mrfailure]

Besides echoing gearcutter's comments, I also wonder if there is a hardness missmatch. You mentioned the PTO had identical wear, but did it really exhibit the loss of material seen on the spline? If not, relative hardness difference could cause this wear.


Aaron Tanzer

http://www.lehightesting.com

 

[triumph993]

Thanks Aaron,

Let me get the PTO cleaned up and I'll post some detailed pics.

David

 

[triumph993]

As promised, here's an album of the wear on the internal splines of the PTO:

https://picasaweb.google.com/dmb2000/ChelseaSplines?authkey=Gv1sRgCNznvcufhd3hJQ

Thanks,
David

 

[mrfailure]

As you said, wear looks to be happening on both sides of the couple.

Based on both albums, this looks to me like adhesive wear where sliding contact is causing material transfer, in this instance on both sides. This mechanism generates friction and heat, accounting for the apparent heat tinting. In the spline pictures, the wear marks suggest relative movement was in the radial (as opposed to axial) direction. Could there be too much slack in the fit-up between the spline and the PTO?

Check out Systematic Analysis of Gear Failures by Lester E. Alban for a more detailed description of gear adhesive wear failures.

Aaron Tanzer

http://www.lehightesting.com

 

[tbuelna]

I would agree with the diagnosis of insufficient lubrication being one likely root cause of failure. The lack of proper lubrication led to fretting.

Fretting failures tend to be self-perpetuating. With splines, there is always a small amount of debris generated due to boundary contact conditions and relative motion. Unless this debris is flushed away from the contact interface it simply acts like an abrasive, causing further surface degradation and friction.

As brimstoner noted, there also appears to be some possible evidence of overheating. The initial fretting damage can lead to increased friction, and if the relative spline motion is significant enough it can overheat the materials, leading to loss of temper/strength.

Hope that helps.
Terry

 

[EdStainless]

The motor wasn't on a VFD was it? I have seen the bad wave form from a VFD accelerate coupling wear like this.
When new is there any rotational play in the coupling?
If they are loose to start with then this happens faster. A very thin coat of a thick synthetic grease will slow this down, but if they don't fit well it will eventually happen.

= = = = = = = = = = = = = = = = = = = =
Plymouth Tube

 

[triumph993]

Thanks everyone for your input.

In real life the metal also has a blue tint, so heat is definitely a factor.

It's not a VFD. The pump hangs off a PTO which is mounted to an Allison transmission with a heavy duty diesel engine as the prime mover.

In a different thread we did an analysis of compressive and shear stresses and found that the spline is undersized for the torque and expected lifetime involved.

My plan is to take the easy route and go with an upgrade from an SAE 'B' to SAE 'BB' shaft (7/8" to 1"). The problem is that the constant mesh PTO is known to have issues with motion in the output shaft. I'm looking into PTO alternatives.

If I had my way I'd go with an SAE 'C' shaft with wet lubrication, but as it stands returning lubrication fluid from an SAE 'B' or 'BB' shaft to the transmission, unfiltered, doesn't seem prudent.

I apologize for the tangent, but thought you might enjoy some background.

Thanks,
David

 

[Carpner]

Hi,

I've been selling Chelsea PTO's for about 20 years now and the shaft wear has been an issue for several years. From what Chelsea engineers are saying, the newer engines have a torsional vibration that causes the wear... I think that is certainly a portion of the problem, another being so much off shore product coming to North America in recent years, the quality of product has decreased...

We recommend pulling the Pump back from the PTO as a yearly maintenance and apply new lubricant. Chelsea Part # 379688 is what the Chelsea people recommend, sorry I don't have technical specifications on that...

I've found the best solution is to go with the DIN5462 PTO output, but depending on your location, the support for that maybe limited. Chelsea output designation "XY" and of course, I'd need to know the pump model to give the part number for it, if even possible... Sadly, Chelsea doesn't offer the SAE C shaft with the B mount on the 277/278 series as of yet...

If you can let me know your application then I can toss some recommendations, not sure if going to an electric over hydraulic shift would help or possible...

I'm not an engineer, just been around and broke enough stuff to know enough to be dangerous... I just found this thread as I was searching for information on a SAE BB shaft twisting and how much torque is required to do that...

 

[gearcutter]

Sounds like the engineers are using a poor excuse to hide the fact that the coupling is severely under designed for the job that it has to do.
I'd like to see the proof that they have to make such claims.

Ron Volmershausen
Brunkerville Engineering
Newcastle Australia

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

 

[gearcutter]

Once off, or rarely applied, applications of grease is the worst thing you could do in these situations.
You'd be better off to run the coupling dry.


Ron Volmershausen
Brunkerville Engineering
Newcastle Australia

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

 

[Carpner]

The PTO is using SAE specifications for the shafts and mount pads, and both Muncie and Chelsea are having this problem and they are the largest manufacturers of PTO in North America. The problem has increased over the last 8 years, in comparison to the 90's and early 2000's... I will see if I can find something from Chelsea...

90% of all truck mounted hydraulics are not specified by engineers, but by parts guys and ordered by owner operators, they look for the cheapest solution... If it was me designing this, the pump would have the SAE BB or DIN shaft...

I'm not a fan of the wet spline, another leak point... The greasable shaft is available but most people use the wrong grease or never grease them anyway...

I will see what I can dig up on the Lubricant supplied with the PTO...

 

[gearcutter]

Have you ever seen the build up of debris that occurs in gaps between the spline teeth and the cavities between the major & minor diameters as a result of grease that has solidified and is trapping the debris?
It can get so bad that the coupling loses the ability to handle any misalignment.
The combination of dried grease and debris can become so hard and tough that the only way it can be removed is with a chisel and hammer.
You might as well assemble the spline with a handful of lapping compound.
If you can’t regularly insert new grease; then adding any grease at all is a recommendation that, IMO and experience, will only cause the joint to fail sooner.
A specification for a once per year application is just plain silly.
It’s a poor attempt at a remedying the problem and is often used as an excuse (lack of lubrication) for covering up the fact that the spline joint is simply under rated.
Once again, show me the proof to the engineer’s claims of torsional vibration being the cause of failure.

We design and build dry spline connections for 600Kw rock crusher shafts that are driven by diesel engines...........now there is a good example of torsional vibration in a system.
The only reason why these splines wear out is as a result of misalignment because of poor fitting practices.


Ron Volmershausen
Brunkerville Engineering
Newcastle Australia

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

 

[Tmoose]

http://www3.dowcorning.com/DataFiles/090007c880002c28.pdf

" bonded coatings containing MoS2, surface pressures up to the yielding point of constructional steel are possible."

Kluber Microlube GL261 survived in our 8000 rpm spindle's spline drive for the life of the spindle, making disassembly possible and even the re-use of the parts sometimes. Previous lubes did not protect against wear so the parts would be found seized solid with fretting debris.
It would not surprise me if Kluber had other products that they might recommend for your application.

BMW motorcycles maintenance includes re"grease"ing the various splines in the driveline every now and then. Some years have plated splines in an effort to fend of corrosion and allow longer re-lube schedules. For a while BMW recommended the Microlube, but a klingy product called Wurth Sig 3000 is earning respect and gaining popularity on some BMW and similarly spliney MotoGuzzi enthusiasts' boards.

 

[EdStainless]

Torsional vibration will destroy splines. That is why I asked about a VFD. The same goes for poorly designed gear boxes (harmonics) and pumps with torsional resonance issues (vane pass feq).
There is no fix for a bad design.

= = = = = = = = = = = = = = = = = = = =
Plymouth Tube

 

[mrfailure]

I earlier noted the failure mechanism appeared to be adhesive wear (and not, let me emphasize, fretting). I wanted to go into a little more detail:

Adhesive wear occurs on sliding surfaces when pressure between the contacting asperities is sufficient to cause local plastic deformation and adhesion. The energy absorbed from the plastic deformation takes the form of heat - hence the blue heat tint observed. The process feeds on itself because the heat softens the surface increasing amount of adhesion and increasing plastic deformation. Eventually the process results in galling, where there is relative movement between the shaft and PTO. The wear marks on both surfaces indicate galling. The wear marks also indicate a radial direction of galling wear contact between the teeth (and not axial).

The mechanism is consistent with a coupling that is inadequately designed to handle the applied load. This root cause may mean a larger shaft/coupling may be required to handle the applied load and prevent initial plastic deformation. The mechanism could also be consistent with a root cause involving inadequate lubrication in the sense that that an appropriate lubricant and maintenance cycle could have the potential to prevent local plastic deformation and frictional heating involved in formation of the adhesive wear mechanism in the first place. Finally, the issue here could be a root cause from the combination of both factors - both inadequate design and improper lubrication.

Aaron Tanzer

http://www.lehightesting.com