Galling/Seizure of Ductile Iron Shaft

[MTUMetEng99]

I am investigating a rash of failures that involves a rotating, ductile iron shaft seizing in an aluminum housing.
The belt-driven shaft rotates on several cast aluminum journals. Seizure is occurring at the journal closest to the drive belt.
Examination shows a galling pattern ~3 or 4 mm wide and located 1 or 2 mm away from the edge of the contact surface that extends all around the cast iron journal. The mating cast aluminum journal contains a corresponding pattern with a build-up of ferrous debris, presumably from the iron shaft.
The shaft is composed of pearlitic ductile iron (~90 HRB); the aluminum is an aluminum-silicon alloy similar to A319. The system is lubricated with motor oil.
There is no indication that any contaminants entered through the oil feed hole of the affected journal.
Any suggestions for what to look at next?
Thanks,
BergiusMaximus

 

[syapnh]

According to your description the load is too big and generates a too high temperature in the zone where the galling occurs. It is exactly in this zone that pressures are the highest. For a given rpm value and a diameter the product pressure x sliding speed is proportional to the energy lost in friction which has to be evacuated if the lubricant is expected not to loose its portance.
The excentric load on the bearing generates a non uniform pressure distribution with a peak on the side where the load acts. If the shaft diameter is bigger there are several positive aspects:
- shaft is more rigid so that the pressure peak can decrease
- the contact area is bigger and the pressure decreases but
- the sliding speed encreases in the same ration
- a bigger area with the environment will reduce the thermal resistance and allows a lower temperature
It could be possible to reduce the effects by use of a higher vicosity lubricant or with high pressure additives.

 

[electricpete]

Out of curiosity, is the damage on the stationary aluminum shell focused around the load zone?

Another curisotiy, in this case you believe the shaft is softer material than the shell?

For sleeve bearings, there are conditions necessary to establish minimum film thickness. Stribeck's number S = mu * N / P needs to be kept above a minimum value , where mu is viscosity at operating etemperature, N = speed, P = load.

There are plenty of design charts available that address these factors. Maybe you can find some on google. If you post the shaft diameter, oil viscosity grade, estimated operating temperautre, speed in RPM, bearing length, and load we can get a rough idea fromt those charts.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[electricpete]

Also, is the temperature of the bearing shell monitored? If yes, how far away from the working surface and the load zone? Any vibration history?

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[electricpete]

A description of the belts would provide a clue to estimate the load.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[JTreglio]

A suggestion: sputter deposit a solid lubricant such as moly disulfide or tungsten disulfide on the shaft. These materials love high loads, especially tungsten disulfide, and will eliminate the galling as well as give you lower friction levels overall. There are companies that offer the coating as part of their coating service. Be careful to get the sputter-deposited material. Not sure the spray-on type can take the load.

Jim Treglio

 

[syapnh]

The matching of the 2 materials is not the best.
I do not know how much will cost the sputtering of WC but the best solution is to use the right combination of materials and give the right harness differences between the 2 surfaces.
The iron used is soft and the Al2O3 layer is hard so combined the higher temterature due to the too high load, the reduction in film portance due to the lower viscosity and the presence of hard particles lead to the galling.
As usual a wrong start leeds to expensive repairs.
The normal relationship is : hard shaft and soft journal ib this case it is inverted.

 

[electricpete]

The normal relationship is : hard shaft and soft journal ib this case it is inverted.

A clarification of terminology, the journal is the portion of the shaft that sits in the bearing.

But I agree that usually we see a hard shaft and a soft bearing shell. As far as I know, I think the main reason is that it is usually much easier to replace or machine a removable bearing shell than replace or machine or buildup the shaft.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[syapnh]

Sorry for the misuse. The reason the journal has to be hard and the bearing soft can be explaines with a similitude with turning on a lathe:
to work the cutting edge has to be harder than the piece which turns.
It is exactly what happens in the analysed case.
In an assembly shaft + bearing it is expected to have the contrary: i.e. the not rotating piece should wear first.

 

[electricpete]

You may be right, but I have a question: The soft piece will wear, whether the soft piece is the rotating piece or the stationary piece, right? Is there some reason galling is more likely when the soft piece is rotating vs stationary?

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[syapnh]

The conditions are not exactly the same. If the soft part turns the surface defects generated by the stationary part will cover all the circumference. The lubrication will be destroyed all over and the risk for galling due to too high friction is higher. If the hard piece turns the wear is local, only on the spot where the contact between shaft and support occurs. If temperature stays between limits and if the bearing material has not the tendency to weld to the shaft then destruction is limited since the shaft surface stays "clean". If particles are generated the softer bearing will deform under local pressure and the pressure peaks between particles and rotating surface are limited. If the rotating surface is hard (that means resistent to those pressure peaks due to a yield limit high enough) it will not be affected. The worst case is a pair with a soft bearing but not too soft and a soft shaft: particles generated by wear will penetrate the bearing surface and exerce on the shaft pressure peaks which -since the shaft is not hard enough- will deteriorate the surface and destroy the lubrication film portance, it follows a rise in temperature a reduction of the viscosity and the destruction loop goes on.

 

[unclesyd]

Have you looked at surface hardening of your Ductile Iron Shaft in the bearing area?

 

[rlh42]

Localised relief of the bearing to allow a better load distribution whilst also creating an entrainment area for the lubricant may help.

 

[coatingengineer]

Hi.

In determining which surfaces to coat, consider the following: For a 'hard-soft' bearing system, coating the softer of the two surfaces will extend the life of the system. However, the harder a substrate material, the more effectively the lubrication will perform.

Generally speaking, the coefficient of friction will be slightly higher if both mated surfaces are coated than if only one surface is coated. However, wear life will increase if both surfaces are coated.

So, you will have to determine which is more important: longer wear or greater efficacy in lubrication.

Let me know if you need help with coatings. Because, that's what I do.

William
IndustrialCoatingsWorld

 

[coatingengineer]

Hi.

I'd forgotten to include this link:
<a href="

http://www.industrialcoatingsworld.com/Low-Friction-Coatings/MoS2-Low-Friction-Coatings.html">MoS2

Low Friction Coatings</a>

Good luck!

William
IndustrialCoatingsWorld

Hi.

In determining which surfaces to coat, consider the following: For a 'hard-soft' bearing system, coating the softer of the two surfaces will extend the life of the system. However, the harder a substrate material, the more effectively the lubrication will perform.

Generally speaking, the coefficient of friction will be slightly higher if both mated surfaces are coated than if only one surface is coated. However, wear life will increase if both surfaces are coated.

So, you will have to determine which is more important: longer wear or greater efficacy in lubrication.

William
IndustrialCoatingsWorld