416 on 304ss fretting 2

[geesamand]

I have an application in which a 416 ss sleeve is attached to a 304 ss shaft with close clearance fit and 316ss cup point set screws. (3" shaft with .003" diametral clearance). A radial bearing is attached to the sleeve.

We occasionally see the joint wear/yield at the tips of the set screws and wear between the sleeve and shaft. The shaft wear surface leaves no obvious debris and has a grey, wavy, velvety texture. Because the 416 sleeve is harder than the 304 shaft, the shaft loses most of the material. We haven't yet been able to catch whether the set screw points or the sleeve clearance is the primary and which is the secondary wear mode.

Any suggestions on what to look into? We seemed to have no incidents of this loosening back when the materials were nickel plated 1018 shaft, nickel plated 4140 sleeve, and 316 ss set screws. We changed to stainless for the corrosion resistance and to get away from nickel plating.

 

[EdStainless]

The 316 set screw into a 304 shaft does not sound like a formula for success to me.
I also would worry that now your shaft has a much higher thermal expansion than the sleeve, where before they were matched.
I believe that the set screw is working loose (or the shaft is yielding under it).
I would also believe that the wear is self limiting, the 316 will work harden a lot, and the wear will slow way down.
Is the 416 working? The sulfides in it give it very poor fatigue properties.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[geesamand]

Thank you for the answer.

The 416 is "working" relative to the other parts.

We recently started to switch from a 316 set screw to a knurled cup point alloy steel set screw. There's a chance it could corrode more but we're willing to take that chance. Time will tell if it holds up better.

Any suggestion how to pursue the question of 304 shaft with 316 set screws? Does it make you less comfortable than 316 set screws in a nickel-plated 1018 steel shaft, for example? My google search efforts turn up nothing so far other than set screws for sale.

 

[EdStainless]

We used to key sleeves in place, they were loose, but restricted from turning or moving longitudinally.
If the set screw is in the 416, then as things get warm the screw and shaft expand a lot more than the sleeve. After a few cycles of this the set screw is no longer tight.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[tbuelna]

Why don't you thermal fit the sleeve onto the shaft with some interference? Using a clearance fit and set screws to mount a sleeve to a shaft is the worst approach I can think of. Plus I would imagine your bearing requires a slight interference fit between the sleeve OD and the inner race ID to perform properly.

 

[geesamand]

Shrink fits and bonding are the obvious choice for a rotating connection, however in this machine it needs to rapidly detach from the shaft using hand tools. This sleeve and set screw arrangement had very good real-world success, and since changing to 416 on 304 we believe there is a trend in this fretting / loosening. That suggests the mechanical design is not the problem so much as the materials or another factor.

We are investigating other shaft attachment methods (tapers, locking rings, compression collars, etc) but they each have their own drawbacks.

 

[geesamand]

Also, I described my wear between the 304 and 416 as fretting. Perhaps it could be galling, or mechanical wear due to grit/abrasive dust.

I have attached a photo of the 304 stainless shaft. The right hand section is where the wear occurred and the band of clean material in the middle is a portion of that same shaft journal absent of wear.

I prefer options that keep the 304 shaft material in place.

 

[EdStainless]

That look about right. I have seen brass sleeves cause the same wear on high strength shafts. It is the abrasives and the slight rocking or scuffing motion that causes it, and it isn't always the softer material that wears.
I would look at using a key to prevent rotation, you may still need a set screw to hold axial position (though there are other ways).

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[geesamand]

There is a key present - this wear involved at least rocking motion in the advanced stages. The key is on the opposite side and there was little/no wear on the shaft journal around the key. The key transmits tiny torque (bearing rotating friction) but significant wear was present on the driving sides of the key. The heaviest wear was on the side opposite the key, which suggests the rocking shaft pivoted on the key and caused sliding wear on the side opposite the key.

David

 

[EdStainless]

It sounds like these sleeves are looser than they should be.
Maybe double keys would help also. Since torque is very low they can be very shallow.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[geesamand]

We've taken the sleeve clearances down as far as practical without grinding or honing them to fit. They aren't loose unless the sleeve fasteners come loose, and being mechanical in nature we it's reasonable to say this will happen once in a while no matter how well the fasteners are constructed.

I've studied the literature and our damage and I don't see galling at all. This is not surprising since 416 and 304 have very good galling resistance. Unfortunately the soft 304 will degrade in dry sliding wear/fretting.

I'd appreciate any suggestions for an affordable hard finish process I can add to the 304 shaft journals. It needs to minimize the wear of the shaft journal in the event the 416 s/s sleeve gets loose. The 416s/s sleeves are no harder than 250 HBn. I'm not looking to "prevent" shaft wear per se, as much as get the wear to occur in the sleeve, and hopefully allow the user to recognize the condition before the shaft is trashed. I will consider change the sleeve to something softer and still machinable like 303 just to help push the balance further.

(I've been doing research on hard finish methods and it seems everyone who does coatings is a specialist in one-two coating methods. I'm hoping for a broader-scope suggestion so I don't have to get lost in numerous quotes from all kinds of coatings just to find the lower-cost technologies)

Thanks to all who have contributed.

 

[EdStainless]

How hard are the fretted areas of the shaft?
There are a few ways to get the shafts harder, and preserve the corrosion resistance.
The first would be to use cold drawn bar for shafting. You can buy Nitronic 50 shafting readily since it is used as boat prop shafting. This will be more expensive, but it should machine better than annealed bar.

Another would be to 'overlay' (HVOF) the desired areas with a hard facing (ceramic or carbide). These would need to be prepared by machining out some metal, and then finish ground after coating. Lots of process step involved, and the final grinding isn't cheap or easy.

Or you could hard finish the entire shaft with on of the low temperature carburization processes. These are different from convention treatments in that no carbides form in the structure so corrosion resistance is not harmed. They are a thin but hard surface zone of the metal it self. These shafts will be difficult to machine after treating so all work will need to be done before, and then re- straightening can be difficult.

There are down sides to each of these approaches, so you will need to weigh the benefits.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[geesamand]

Thanks for the ideas.

The current 304 shaft is annealed condition so approx. 120-130 HBn.

I have some experience with the powder met coating process and I think that's too expensive.

This shaft is 304 for the corrosion resistance but at the sleeves, any level of corrosion resistance is probably enough. Low temp carburizing is interesting and I will look into it.

I hadn't considered BSQ Nitronic 50 for the shaft but I will also look into it.

Should I consider Colmonoy or Stellite type of overlays?

 

[EdStainless]

We were doing this on cold drawn 625 shafts, and we we found that any coating helped, and none was that much better than another. We tested metal bonded carbides as well as ceramics. They were all HVOF or similar (no simple flame spray). They all failed in heavy abrasive conditions, but they all helped.

The Nit 50 (Aquamet 22) will cost more, but maybe you can find cold finished 304 (201 might even be better) to use.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[geesamand]

Hi Ed,

I appreciate your help with this subject in the past. We're still puzzled by what's going on and would like to find an expert opinion on our issue. Do you how I might locate a consultant(s) to help us?

Thanks, David

 

[Tmoose]

What is the nature of the radial loading applied to the shaft and bearing?

http://www.nskamericas.com/cps/rde/...ng/CNSK_Pocket Fits06-Fits and Clearances.pdf

What is the orientation of the set screws? two at 90 degrees, three at 120 degrees, etc. All on one end of the sleeve, or some on each side of the bearing?

If the bearing outer race stays stationary and the shaft rotates and there is a significant one direction radial load ( like a belt drive or gravity ) then clearance between the sleeve and shaft, or sleeve and bearing, and set screw retention is just looking for trouble.

 

[geesamand]

A bit of an update:
We switched to an alloy steel set screw and problems aren't gone yet. I missed something very important in Ed's first post: thermal expansion. Running some numbers suggests this really is a risk. Given that the set screw points yield the shaft material during torqueing, it's reasonable to believe it will yield more when temperatures cause the shaft to grow relative to the sleeve, and that it will be loose from that point on when temps swing back. We will test this to confirm it's a factor.

We performed tests for galling wear (we even ran the machine without set screws at all) and could not generate any in a reasonable time. However, this did not include the real-world effects of dust and dirt that get blown around and into the joint. Therefore the visible wear could be strictly abrasive.

Assuming for a moment this is the case, it seems we could go to 17-4PH for the shaft and 416SS for the sleeve. Those CTEs match well and the physical properties are very good.
For machines where we need to keep the 300 series shaft, I need to find a suitable sleeve material. It needs only nominal corrosion resistance (416 was amply good). Based on CTE compatibility I see only 300 series, 201, and Nitronics. I did price out Nitronic once and found it impractical. Out of the 300 series, I'd start with 303 for its machinability but the dimensional stability is unknown and galling resistance vs. 304 is poor. Maybe we can nitride it to resist galling. 201 is interesting but I have no experience with it and it's not well documented. Thoughts on a sleeve for a 304L shaft?

Tmoose,
"If the bearing outer race stays stationary and the shaft rotates and there is a significant one direction radial load ( like a belt drive or gravity ) then clearance between the sleeve and shaft, or sleeve and bearing, and set screw retention is just looking for trouble."
I know, if I was involved in the early design phase I'd have quashed it on principle. But it works well in this system. We've investigated many other shaft attachment methods and haven't settled on another that meets all of the basic requirements.

 

[EdStainless]

I can't believe that cold drawn Nitronic 40 bar is more expensive than 17-4PH that has been aged.
The Nitronic alloys are 200 sseries.
You are right that 17-4 will have slightly lower CTE, but at higher strengths I am not sure that it is as significant.
Stay away from 303, its corrosion resistance is nearly as bad as 416.


= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[geesamand]

Thanks. I will look for Nitronic 40 grade.

I have no concern about the corrosion resistance of the sleeve - it's not exposed to the corrosives. We just want to avoid rust that slows down the mechanics who have to service it.

Other than corrosion resistance, what concerns would you have for 303? Does the machinability help with dimensional stability?

 

[EdStainless]

If you want free machining look at Outokumpu's Prodec alloys. They make versions of 304 and 316 that machine well but don't rely on massive S inclusions.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube