Roller shaft failed 6

[sokenyou]

In a paper converting mill, one of our draw roller shaft failed at the weakest point (near the bearing and chamfer with the radius of 0.5). Rachet marks (but shiny) indicates that fatigue cracks were initated at a few locations along the groove. The eccentric pattern tells me that the load on the shaft was imbalanced BUT it's only pressing on tissue papers, and to our knowledge, there was no excessive loads. Two shafts failed within a month... it's really frustrating.

 

[EngineerErrant]

Do you have a picture for us?

"Engineers like to solve problems. If there are no problems handily available, they will create their own problems." -Scott Adams

 

[sokenyou]

Thanks Scott, yes I have picture for you to look at. In fact, I was told that it was fretting corrosion that caused it but we are iffy about this because it's not subject to high loads with other parts and it was in the mill for two weeks.... again, I still haven't figured out what to do next to resolve this issue.

 

[asimpson]

Roller could be resonating at its natural frequency. Was much vibration noticed before breakdown? Have conditions changed like speed or loading. Higher loading may even dampen out vibration.

 

[sokenyou]

Thanks asimpson....speed was up 30% but not loading. Just a note too although speed was up 30%, it's still way below the designed speed, so it's just mind boggling and I don't see where the fretting corrosion came from? Between the shaft and the bushing????
Thanks.

 

[JohnRBaker]

Just remember, one of the criteria for the 'design speed' may have been to find a point at which there were no harmonic issues. However, reducing the speed may actually have dropped it to where a harmonic could have become a problem. If there's a potential for a natural harmonic frequency situation, lowering the speed is not always the right answer.

John R. Baker, P.E.
Product 'Evangelist'
Product Engineering Software
Siemens PLM Software Inc.
Industry Sector
Cypress, CA
Siemens PLM:
UG/NX Museum:

To an Engineer, the glass is twice as big as it needs to be.

 

[tbuelna]

sokenyou,

It was a bit difficult to see details of the fracture zone due to the picture's low resolution. After looking at the picture, can you confirm a couple things for me?

First, it looks like the bearing used is a ball bearing unit with a self-aligning flange mount and a set screw on the inner race. Is that correct?

Second, what was the fit between the shaft and bearing bore? If there was clearance and relative movement, another possible cause of the fracture (besides contact fretting) may have been shaft surface gouging damage produced by the set screw tip.

Regards,
Terry

 

[sokenyou]

Hi Terry,

Yes, the bearing used is a ball bearing unit with a self-aligning flange mount and a set screw on the inner race. The fit between the bearing zone and the shaft is a couple of thou according to what the Mechanic was saying.

However, I don't understand why this (clearance and relative movement) is an issue, it's been like this for a good number of years according to what I heard (I just started this job a month ago). Well, we have seen signs of the shaft surface gouging damage produced by the set screw tip though...

I am puzzled about this issue as well...not understanding why it wasn't an issue before and now it's giving us problems.

Thanks everyone if you have an idea to determining the root cause of this.

Thanks.



Regards,
Kenneth

 

[redpicker]

One of the things you will find out is that all problems are "new".

"They've been doing it like this for years and never had a problem"

"We've just started seeing this; it's never happened before"

For nearly all of these types of problems, if you dig deep enough, you will find that either
1) They haven't always been doing it like this. They've either used a higher (or lower) rated motor/shaft/some other component or have increased the loading/production/or some other usage variable.
OR
2) They've always had problems with it.
Usually, both.

The reality is that nearly all of the problems are "old". They rarely ever get solved, someone just puts a band-aid on it and they go on. Sometimes the band-aid will help, sometimes not, but since the broken component has been replaced with a new one, it works for a while. People forget/get moved to another job/some new guy comes along.

Easy problems get fixed and since they no longer exist, nobody remembers them.
Tough ones get recycled.

rp

 

[JohnRBaker]

Even if the clearance was only a "couple a thou" it could still result in some degree of eccentricity which would exacerbate the potential for harmonics. Anything 'high-speed' should really use press-fit bearings.

John R. Baker, P.E.
Product 'Evangelist'
Product Engineering Software
Siemens PLM Software Inc.
Industry Sector
Cypress, CA
Siemens PLM:
UG/NX Museum:

To an Engineer, the glass is twice as big as it needs to be.

 

[tbuelna]

sokenyou,

Clearance between the shaft and bearing bore is a big issue, since it's this clearance that sets up the conditions for fretting to occur. Rather than installing the shaft into the bearing bore with clearance and relying on a set screw to hold things secure, I would suggest using a bearing that has a locking tapered sleeve on the inner race. This would eliminate the possibility of fretting on the shaft surface. These bearing are offered by SKF (and others) in a flanged self-aligning pillow block design like the one you are currently using.

As for this situation never occurring in the past, try to be optimistic about it. You had a great run of good luck!

Hope that helps.
Terry

 

[mfgenggear]

Sokenyou

I wanted to add to the very good advice giving all above.
To my experience a bearing that spins in or outside of it's mounted surface,
is a very bad to occur. Like instant failure.
I like both recommendations above, but feel a pressed bearing(with Loctite) or thermal assembled, is the the better method.
This is just my opinion & experience.

Mfgenggear

 

[Tmoose]

Hi Kenneth,

Can you provide pictures of OD the hunk of broken shaft that was in the bearing (the bearing "seat"), and the OD of the broken stub on the roll, and the bearing ID?

In the picture it looks to me like the shaft extends out a ways, for a coupling or gear or sheave or something.
Do you have details of that area too, like sheave width and number of belts, belt tensioning method, type of cou0pling, etc.

Dan T

 

[sokenyou]

Hello Dan T,

Please see attached - dimensions of the shaft and the roller.
Thanks.

Regards,
Kenneth

 

[sokenyou]

Hello All,

We found some cracks on the shaft very close to the bearing sleeve. We are now leaning towards fretting corrosion, BUT here is the thing, we are pointing our fingers at everything that could possibly leading to microscopic motions of tightly fitting parts, however, why didn't it happen for the last so many years before we turn up the speed of the mill and this shaft started failing?

The thing is, the design for the machine is to run it at 2 times of what we are running at right now!!!

It's just mind boggling....


Thanks.

Regards,
Kenneth

 

[Compositepro]

http://www.mt-online.com/july2012/failure-analysis-of-machine-shafts

 

[Tmoose]

Hi Sokenyou,

I,m still interested in pictures of the damaged shaft and bearing race.

If the bearing ID is a slip fit on the shaft (diametral clearance greater than 0.0002 inch) and the bearing uses any commercial combination of setscrews, eccentric collars, or even slit race extension and clamping collars then radial forces from a belt drive or even just roll weight will be enough to create micromotions that will first slowly, then more quickly eat the shaft, setscrew and bearing race.
The greater the clearance, the more rapidly wear will progress.
The greater the radial load the more rapidly wear will progress. Excessive belt tension can be a killer two ways - hastening the micromotion, and greater bending stress on the shaft.

regards,

Dan T

 

[Occupant]

I tend to agree that this is probably a resonance problem, but want to point out that "j6" is not a slip fit. It falls under the category of interference fits and is a shaft fit recommended for normal to heavy loads.

 

[Walterke]

If they are running at (exactly) half the design speed, wouldn't the resonance created there also be there at the normal speed? (Been a while since I did any vibration work but I seem to remember something like this)

@Sokenyou:
Look up old orders on that machine and try to find out how frequently the shaft broke before you arrived there. If it indeed wasn't a problem before, find out what they changed and, if possible, change it back.

NX 7.5.5.4 with Teamcenter 8 on win7 64
Intel Xeon @3.2GHz
8GB RAM
Nvidia Quadro 2000

 

[sokenyou]

The shaft broke within two weeks last time and the shortest time it lived there was 5 days. The one we have in now has been there for 10 days and we are praying hard that it will run for a long time.

Is there a possibility that the root cause of the problem is heat treatment of the steel or the heat treatment wasn't done properly?

Also, what are the consequences if I make the shaft with a more ductile material??? what would be the trade-off? I am not solving the problem but if the shaft distorts then I know something is wrong and can plan ahead to change it, to buy some time.

Please advise if you can.
Thank you