Debris from carburized gear 3

[UCengno1]

A product we produce in specific configurations utilizes a splash lube planetary gearbox for speed reduction. The sun (nitralloy) and fixed ring gear (8620) are nitrided and the planets (9310) are carburized. In order to facilitate identification of lube refill we have recently added sightglasses which has led us to another discovery, significant dark colored debris in the gearbox. The debris is so dark and pervasive that it completely eliminated the usefulness of the sightglass.

We do not have wear related issues but this discoloration is very disconcerting to the customer and we have a competitive product that does not have this issue.

An analysis is in process but I am assuming the debris is worn carburize. I am also assuming that this has been an issue long before addition of the sightglass so I need to be careful to avoid solving one problem while creating another.

Am I correct in the assessment of the debris as a result of worn carburized layer?
Is there another wear issue that could generate the dark debris?
Would a completely nitrided set eliminate dimish the dark debris?

Trying to start down the right road from the begininning of an investigation. Thanks.

BCK

 

[metengr]

BCK;
Why not have the debris analyzed by a lab that specializes in wear metal/lub oil analysis to aid in determining the origin?

 

[mrfailure]

I hate jumping to a root cause and corrective action without knowing the facts. Before assessing the cause of the debris, you need to confirm what the debris is itself (EDS analysis would be my first evaluation technique). I also find it very helpful to look at such debris in a stereo binocular microscope to characterize debris morphology.

Aaron Tanzer

http://www.lehightesting.com

 

[gearcutter]

Are you suggesting that carbon from the carburized layer is some how being released into the oil causing the oil to change colour?

Ron Volmershausen
Brunkerville Engineering
Newcastle Australia

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

 

[UCengno1]

Ron,

That is what I am considering. Do that make sense?

Regarding analysis.
1) Particulate analysis pick up an even mix of non-magnetic and magnetic media.
2) Magnetic media was very fine matter. Partial consituent was 9310 but a large amount was unidentified. SEM in process to define the unknown.
3) Non-magnetic media was shiny matter that turned ou to be primarly silver (Ag) flakes. We are waiting for final confirmation of this from SEM but assume this is from a degrading bearing.

Thanks.

BCK

 

[mrfailure]

The 9310 obviously is consistent with the planetary gears. Silver intrigues me because I have not experienced it as a constituent in bearings. What technique did you use to identify the Ag? Is it possible that Ag was a misidentification from whatever analytical technique you used? If Ag is confirmed, potential source(s) in your system for it will go a long way in identifying root cause.



Aaron Tanzer

http://www.lehightesting.com

 

[UCengno1]

I get my SEM results back on Monday, 10/3. Will post those results when avaiable. That will shed some light on the situation.

I was trying to be pre-emptive on whether anyone had experienced this type of black staining in the lube due to carburize.

Thx. BCK

 

[EdStainless]

any finely divided material in the oil will look black.

I suggest that you start grabbing oil samples from old units also and have them analyzed. Like you said you don't want to create new problems.

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

 

[hydtools]

Consider also any products used during gearbox parts cleaning and assembly.

Ted

 

[Guest102023]

Don't just analyze the oil, analyze it at regular intervals. There ate many such labs that will do that and alert you when undesirable trends develop.

 

[Guest102023]

There are ...

 

[gearcutter]

I doubt very much that somehow carbon is being released from the carburized layer in enough quantities to cause the oil to change colour.

Are you using an EP oil?
Over what period of time is the oil changing colour?
Do all the machined surfaces within the casing still have the original as machined finish?

It sounds to me like you either have an adverse chemical reaction going on or the oil is being exposed to high temperatures.
It is possible for oil to be exposed to excess high temperatures within a gearbox without changing the bulk temperature.

Ron Volmershausen
Brunkerville Engineering
Newcastle Australia

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

 

[UCengno1]

Ron,

Gearbox currently utilizes a synthetic lube per Mil-L-7808. The time to color change is disturbingly short, only 3-5 minutes of operation. Good question on the platings as all of the major components are unplated but I need to verify the small stuff; i.e. spacers, lock-rings, etc... Will check and reply.

High temperature is a possiblity as the input speed of the unit is reasonably high, 30k rpm on a 13 mm gear. Localized heating may occur. We are setting up a test to confirm time to steady state lube temp so more data will be available soon. However that is the bulk temperature.

Thx.

BCK

 

[gearcutter]

Based on the short time to discolouration and speed of the mesh; I would suggest that you have a problem with the design of the gear geometry and lube you've chosen.
Any elevated levels of carbon, or soot, found from the oil analysis is most likely as a result of the oil being exposed to very high temperatures at the point of meshing. This will not necessarily increase the bulk temperature.
Have you been able to closely inspect the gear teeth?
Is there any evidence of scuffing/scoring?


Ron Volmershausen
Brunkerville Engineering
Newcastle Australia

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

 

[tbuelna]

UCengo1,

Since you are using MIL-L-7808 turbine engine oil, I'm assuming this is some sort of aircraft gearbox, possibly an EMAD. As such, there are many possible sources of the silver particles in your oil. The most likely source is from bearing cages. Aircraft gearbox rolling element bearings commonly use alloy steel cages with silver plating. Another source may be from metal locknuts used inside the housing. Silver plating is commonly used on metal locknuts as an anti-seize, and some of it gets scuffed off of the threads of new locknuts when they are installed. Splines are also sometimes silver plated to inhibit fretting, so that may be another potential source.

As for the black residue collecting on your oil level sight glass, it's not likely ferrous debris from gears or bearings. Ferrous debris is very dense and would quickly settle out of the oil. Thus it would not collect on the oil reservoir sight glass, which is presumably at the top level of the oil in the tank.

If the debris is dark and of low density, I might speculate that it is possibly from a seal. Does your gearbox have any carbon face seals or lip seals?

Finally, as gearcutter suggests, if there is excessive heating of the oil, it will quickly be evident in gear tooth scoring. A gear that is subject to scoring will fail very quickly, and such a failure will be evident in scuffing of the gear tooth flanks.

Good luck.
Terry

 

[gearcutter]

Terry,

Some very good points you've made.
Star for you.

Ron Volmershausen
Brunkerville Engineering
Newcastle Australia

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

 

[UCengno1]

Terry,

You make a great point about the media "suspended" in the lube being non-metallic. Kind of an "ah ha!" moment. We have a standard Chicago Rawhide oil lip seal on the forward end, viton with the garter spring for contraction on shaft. We have observed grooving in the shaft that could certainly contribute to the debris. We also have a carbon face seal on the high seal on the high speed input. Wouldn't you expect potential for more debris out of the viton seal? I was considering a tempeoray replacement of the viton seal to see if the desbris is reduced/eliminate. I would not expect the same out of a PTFE element seal would I?

I think you are spot on with the silver as the bearings have some significant run time on them and could be showing wear. These are ABEC 7 bearing made for high speed operation so the use of silver plating would have high potential.

BCK

 

[tbuelna]

UCengno1,

Based on your last post, I'd take a close look at the carbon face seal. They work well when installed and operated correctly. But they are very sensitive to preload, misalignments, insufficient cooling, and excessive roughness on the mating faces.

A few things you might want to check with your carbon face seal:

1) The finish of the carbon ring and gland's mating sealing surfaces. These surfaces must be extremely flat and smooth. Your seal vendor can give you spec's for these, but because they are so smooth you may need some specialized inspection equipment.

2) The axial preload of your installed seal. Most carbon face seals use something like a wave spring to provide preload, although there are some that use magnets. Do a dimensional check of your test article parts to establish if you have the proper axial installed length. Check the shaft end play. And perform a tolerance stack-up analysis based on your drawing dimensions, both at RT and OT.

3) Check to make sure your seal has adequate cooling. If the seal sliding speeds are high, or if the seal has excessive preload, it can easily overheat. This results in coking of the lube oil (ie. dark residue) around the inside surfaces of the seal. It is common to provide a small impinging oil jet to cool carbon face seals.

Good luck.
Terry

 

[gearcutter]

UCengno1,

Any chance of letting us all know how the investigation went and what your conclusions were?

Ron Volmershausen
Brunkerville Engineering
Newcastle Australia

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

 

[tbuelna]

UCengno1,

I agree with gearcutter. Don't leave us hanging. If you don't tell us what happened it just takes all the fun out of armchair engineering.

Terry